Surgical Management of Aortic Pathology: Current Fundamentals for the Clinical Management of Aortic Disease 978-3709148723; 3709148723

Citation preview

Surgical Management of Aortic Pathology Current Fundamentals for the Clinical Management of Aortic Disease Olaf H. Stanger John R. Pepper Lars G. Svensson Editors

123

Surgical Management of Aortic Pathology

Olaf H. Stanger  ·  John R. Pepper Lars G. Svensson Editors

Surgical Management of Aortic Pathology Current Fundamentals for the Clinical Management of Aortic Disease

Editors Olaf H. Stanger Fmr Royal Brompton Hospital London UK

John R. Pepper Royal Brompton Hospital London UK

Lars G. Svensson Heart and Vascular Institute Cleveland Clinic Cleveland Ohio USA

ISBN 978-3-7091-4872-3    ISBN 978-3-7091-4874-7 (eBook) https://doi.org/10.1007/978-3-7091-4874-7 Library of Congress Control Number: 2019932806 © Springer-Verlag GmbH Austria, part of Springer Nature 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-Verlag GmbH, AT part of Springer Nature. The registered company address is: Prinz-Eugen-Str. 8-10, 1040 Wien, Austria

Foreword

Treating aortic disease has always been an important but often complex area of cardiovascular surgery. This is not only because of the aorta’s role as the body’s largest and most vital artery but also because aortic disease can take a variety of forms with a broad range of consequences, from slowly progressive, asymptomatic disease to sudden death by rupture. The surgical treatment of aortic disease, particularly aortic aneurysm, is one of my long-time interests. In 1949, during my residency at Johns Hopkins, I was assisting surgeon Grant Ward with an emergency procedure and ended up performing one of the first excisions of an aortic aneurysm. Later, during my early years at Baylor College of Medicine in the 1950s, my friend and mentor Michael E. DeBakey and I, along with some of our colleagues, developed techniques for replacing aneurysmal aortic segments, first with homografts and later with synthetic grafts. In addition to treating aortic aneurysm, we used these techniques in the first successful repair of a case of chronic aortic dissection in 1954. Also of great interest to us were methods of preventing ischemic injury, particularly to the brain and spinal cord, during these procedures. It pleases me to see how much progress has been made in aortic surgical techniques, technology, and protective adjuncts since those days. In this book, Surgical Management of Aortic Pathology, Olaf H. Stanger and my long-time colleagues John R. Pepper and Lars G. Svensson have assembled the knowledge of many great minds (and hands) in the field of aortic surgery and related disciplines. This volume contains valuable information on a wide range of topics, including the biological underpinnings of aortic disease; modern diagnostic methods; open surgical, endovascular, and hybrid techniques of repairing the various aortic segments and the aortic valve; and methods of protecting vital organs against ischemia during these often complex procedures. I congratulate Drs. Stanger, Pepper, and Svensson, as well as the many contributors, for creating a comprehensive work on a vital subject. Denton A. Cooley Houston, TX, USA

v

Preface

Current knowledge suggests that the aorta, long seen from a purely mechanistic view as a more or less rigid tube, is in fact a highly functional, metabolically active, hemodynamically responsive, and adaptive structure with laminar flow. The editors have sought to review and understand the aorta as an organ per se. Neither disease nor treatment can be understood in isolation because any change caused by pathology or intervention inevitably affects upstream and downstream segments. Accordingly, this comprehensive work not only includes fundamentals of anatomy and development, but is further focused on advances in imaging, genetics, physiology, molecular biochemistry, and all current treatment options and strategies. Over time numerous concepts have emerged to explain certain aortic pathologies, but most conditions are much more complex than previously thought, with interactions and potential overlap of mechanisms adding to the complexity. Perhaps the most important evolution has been brought about by advanced imaging tools visualizing flows, forces, lesions, and changes with previously unthinkable precision. Each patient represents a highly individual case with multiple conditions that influence function, morphology, adaptions, interaction, progression, and risk. Genetic understanding has also grown rapidly and patients, particularly those with hereditary diseases, are increasingly well informed. In an era of rapid genetic analysis, they seek advice on how to prevent aortic rupture, dissection, and death. This presents new challenges as it is difficult to counsel individuals who are seen at such an early stage of their disease that means of accurately predicting their specific outcomes have yet to be developed. Given the great diversity of disease and treatment concepts, the challenge of individual decision-making calls for highly specialized interdisciplinary management involving cardiac and vascular surgeons, interventional cardiologists and radiologists, imaging experts, geneticists, and others. Ideally, each patient will have an individually tailored treatment concept. Aside from acute aortic events, most aortic diseases are indolent and often discovered by chance. At the same time, as more and more patients are diagnosed and undergo treatment, the group of “survivors” grows constantly, with the ever-present risk of pathology progression and exposure to future complications. Since patients must be followed longer, with management of disease

vii

Preface

viii

progression and associated complications, specialist care of patients with aortopathy is a life-long commitment. With expanding insight into the mechanisms that underlie aortic diseases, myths and paradigms are challenged and questioned, reviewed, and adapted. Clearly, not one technique or standard protocol will fit all patients’ conditions. Conventional concepts, i.e., excision and local replacement of an aortic dissection entry tear, are now considered insufficient to cause false lumen collapse and subsequent remodeling with a near normal prognosis in most, if not all, cases. New techniques, particularly interventional and hybrid techniques, are developing rapidly and taking their places in the toolbox. But do they translate into better overall outcome? “The more we know the less we understand,” but progress is clearly being made in parallel with new controversies to resolve. We are honored and thankful that these many experts on aortic medicine and surgery readily agreed to contribute to this comprehensive volume, sharing their knowledge, experience, and modern outlooks. We hope that this book will provide the reader with stimulating and valuable new insights into the ongoing quest for optimal patient care. We wish to acknowledge the valuable support of Genie Lamont (Graz, Austria), and Wilma McHugh (Springer; Heidelberg, Germany) in the preparation of this book. And foremost we are very grateful to the editorial skill and persistence of Sara Baumberger (Project Coordinator; Berne, Switzerland) which proved indispensable in the compilation of this extensive work. London, UK; Munich, Germany London, UK Cleveland, OH, USA 

Olaf H. Stanger John R. Pepper Lars G. Svensson

Contents

1 A Brief History of Aortic Pathology and Surgery����������������������    1 Olaf H. Stanger Part I Anatomy 2 The Surgical Anatomy of the Aortic Root������������������������������������   49 Robert H. Anderson, Diane E. Spicer, and Shumpei Mori 3 Morphology and Surgical Anatomy of the Aorta and Pulmonary Trunk ������������������������������������������������������������������   63 Siew Yen Ho 4 The Normal Aorta and Changes with Age����������������������������������   77 Mary N. Sheppard Part II Mechanisms 5 Molecular Mechanisms of Aortic Valve Pathology ��������������������   87 Ghada Mkannez, Deborah Argaud, Marie-­Chloé Boulanger, and Patrick Mathieu 6 Functional and Morphological Interplay of the Aortic Valve, the Aortic Root, and the Left Ventricle����������������������������������������   99 Marie-Annick Clavel and Philippe Pibarot 7 Interaction Between the Haemodynamics of Coronary Flow and Aortic Valve Pathologies��������������������������������������������������������  115 Christopher J. Broyd and Justin E. R. Davies 8 Left Ventricular Fibrosis in Patients with Aortic Stenosis ��������  127 Vassilis S. Vassiliou, Calvin W. L. Chin, Tamir Malley, David E. Newby, Marc R. Dweck, and Sanjay K. Prasad 9 Novel Understanding on Thoracic Aortic Diseases from Bioengineering Concepts ��������������������������������������������������������������  141 T. M. J. van Bakel, F. J. H. Nauta, Michele Conti, Rodrigo Romarowski, Simone Morganti, J. A. van Herwaarden, C. Alberto Figueroa, F. Auricchio, and Santi Trimarchi

ix

x

10 Mechanics of the Thoracic Aortic Wall����������������������������������������  149 Bulat A. Ziganshin and John A. Elefteriades 11 Role of Biomechanical Stress in the Pathology of the Aorta ������������������������������������������������������������������������������������  163 Giuseppina Caligiuri, Bernard P. Levy, Antonino Nicoletti, and Jean-Baptiste Michel 12 Insights into the Pathogenic Mechanisms of Acute Dissection ������������������������������������������������������������������������  181 Daigo Sawaki and Toru Suzuki 13 Triggers of Aortic Dissection��������������������������������������������������������  191 Bulat A. Ziganshin and John A. Elefteriades Part III  Principles 14 Medical Treatment of Thoracic Aortic Pathologies��������������������  207 Alan C. Braverman 15 Frailty in Aortic Surgery ��������������������������������������������������������������  225 Michael Mack 16 Biomarkers of Acute Aortic Syndrome����������������������������������������  233 Toru Suzuki and Daigo Sawaki 17 Post-operative Management of Aortic Valve Disease������������������  243 Susanna Price 18 What the Surgeon Can Expect from the Pathologist and Vice Versa��������������������������������������������������������������������������������  251 Yara Banz and Vera Genitsch 19 Indications for Aortic Surgery and Use of Guidelines����������������  259 Joon Bum Kim and Thoralf M. Sundt III 20 Lessons Learnt from the International Registry of Acute Aortic Dissection (IRAD)������������������������������������������������  277 Xun Yuan and Christoph A. Nienaber Part IV Imaging 21 Trans(o)esophageal Echocardiography (TOE/TEE) in the Diagnosis of Aortic Pathologies������������������������������������������  295 Raimund Erbel and Sophiko Erbel-Khurtsidze 22 Left Ventricular Function and Aortic Valve Replacement������������������������������������������������������  313 Xu Yu Jin, Jiang Ting Hu, and John R. Pepper 23 Imaging of the Thoracic Aorta������������������������������������������������������  333 Helen Dormand and Raad H. Mohiaddin

Contents

Contents

xi

24 Cardiovascular Imaging in Aortic Diseases: Multimodality Approach in Clinical Practice ����������������������������  371 Arturo Evangelista, Laura Galian, Gisela Teixidó, and José Rodríguez-Palomares 25 Multidetector Computed Tomography of the Aorta������������������  385 Alistair C. Lindsay, Arjun Nair, and Michael B. Rubens 26 Aortic Root Assessment with Computed Tomography in the Context of TAVR������������������������������������������������������������������  409 Paul Schoenhagen, Lei Zhao, and Xiaohai Ma 27 Three-Dimensional Rotational Angiography������������������������������  427 Konstantin von Aspern and Lukas Lehmkuhl 28 4D Flow MR: Insights into Aortic Blood Flow Characteristics ����������������������������������������������  435 Florian von Knobelsdorff-Brenkenhoff and Alex J. Barker 29 Impact of Aortopathy and Aortic Valve Disease on 3D Blood Flow and Wall Shear Stress in the Thoracic Aorta: As Assessed by 4D Flow MRI ��������������������������  447 Michael Markl, Paul W. M. Fedak, and Alex Barker 30 Emerging Tools to Assess the Risk of Rupture in AAA: Wall Stress and FDG PET������������������������������������������������������������  465 Alain Nchimi, Thomas Van Haver, Christian T. Gasser, and Natzi Sakalihasan 31 Exploring the Thoracic Aorta with Advanced Magnetic Resonance Imaging Beyond Routine Diameter Measurements ����������������������������������  487 Alban Redheuil Part V Congenital and Connective Tissue Aortopathy, Bicuspid Aortic Valve 32 Congenital Aortopathy������������������������������������������������������������������  503 Matina Prapa and M. A. Gatzoulis 33 Aortic Connective Tissue Histopathology������������������������������������  513 Mary N. Sheppard 34 Clinical Aspects of Heritable Connective Tissue Disorders ����������������������������������������������������������������������������  523 Aline Verstraeten and Bart Loeys 35 Bicuspid Aortic Valve: Timing of Surgery ����������������������������������  531 Elizabeth H. Stephens and Michael A. Borger 36 Biscuspid Aortic Valve Repair������������������������������������������������������  541 Nhue Do, Joel Price, and Lars G. Svensson

xii

37 Morphology of the Ascending Aorta in Bicuspid Aortic Valve Disease��������������������������������������������������  553 Matina Prapa and M. A. Gatzoulis 38 Genetics of Marfan Syndrome and Loeys-Dietz Syndrome������������������������������������������������������������������  561 Aline Verstraeten and Bart Loeys 39 The Exostent Concept for the Marfan Syndrome����������������������  567 John R. Pepper, Conal Austin, Mario Petrou, Filip Riga, Ulrich Rosendhal, Jan Pirkf, and Tom Treasure 40 Aortic Dissection in Patients with Disorders of Connective Tissue����������������������������������������������������������������������  575 Florian S. Schoenhoff and Thierry P. Carrel Part VI Aortic Valve and Root 41 Diagnostic and Therapeutic Targets for Aortic Valve and Ascending Aorta Pathologies: Challenges and Opportunities��������������������������������������������������������������������������  591 Giovanni Ferrari and Juan B. Grau 42 Concepts of Aortic Valve Repair��������������������������������������������������  609 Stefano Mastrobuoni, Laurent de Kerchove, Munir Boodhwani, Emiliano Navarra, and Gebrine El Khoury 43 Surgical Repair of the Bicuspid Aortic Valve: Predictors of Failure����������������������������������������������������������������������  631 Maude Pagé and Jean-Louis Vanoverschelde 44 Native and Prosthetic Valve Endocarditis������������������������������������  643 Amina Khalil and Jonathan Anderson 45 Reconstruction of the Left Ventricular Outflow Tract for Aortic Root Abscess����������������������������������������  663 Shinichi Fukuhara and Michael A. Borger 46 Stented Bioprosthetic Valves ��������������������������������������������������������  677 John R. Pepper 47 The Stentless Valve Concept����������������������������������������������������������  685 Alberto Repossini 48 Challenges and Lessons from Preoccupation with the Stentless Aortic Valve Prosthesis Concept��������������������  699 Olaf H. Stanger 49 Management of Aortic Prosthetic Leaks��������������������������������������  719 Alberto Pozzoli, Maurizio Taramasso, Michel Zuber, Shingo Kuwata, André Plass, Marco Russo, Fabian Nietlispach, and Francesco Maisano

Contents

Contents

xiii

50 Aortic Root Enlargement��������������������������������������������������������������  731 A. Parsee and J. Price 51 Current Role and Techniques of the David Operation ��������������  741 Tirone E. David 52 The Yacoub Operation������������������������������������������������������������������  749 Magdi Yacoub, Heba Aguib, and Ahmed Afifi 53 Effect of Aortic Root Repair on Aortic Valve Function��������������  757 Michael Scharfschwerdt 54 The Ross Operation ����������������������������������������������������������������������  765 John R. Pepper Part VII TAVI and New Technology 55 TAVI: A European Perspective ����������������������������������������������������  773 Leanne Harling and Andrew Chukwuemeka 56 Transcatheter Aortic Valve Replacement in the USA����������������  783 Jose F. Condado, Hanna Jensen, and Vinod H. Thourani 57 Transcatheter Aortic Valve Replacement: Management of High-Risk Patients and Complex Procedures ������������������������  795 Brandon M. Jones, Lars G. Svensson, and Samir R. Kapadia 58 New Technology: The Sutureless Valve Prostheses ��������������������  807 Paolo Berretta and Marco Di Eusanio Part VIII Ascending Aorta 59 Recommendations on Thresholds for Elective Surgery for Thoracic Ascending Aortic Aneurysms����������������������������������  821 Loren F. Hiratzka 60 Impact of Ascending Aorta Replacement on Ventricle Load and “Windkessel” Function��������������������������������  831 Michael Scharfschwerdt 61 Surgical Strategies in Acute Type A Aortic Dissection ��������������  837 Mark Field, Deborah Harrington, Omar Nawaytou, and Manoj Kuduvalli 62 Acute Type A Dissection Repair ��������������������������������������������������  857 Syed T. Hussain and Lars G. Svensson 63 Management of Aortic Malperfusion Syndromes in Aortic Dissection������������������������������������������������������������������������  875 Hector W. L. de Beaufort, Arnoud V. Kamman, and Santi Trimarchi

xiv

64 Cerebral Protection During Surgery of Acute Type A Aortic Dissection������������������������������������������������  885 Jean Bachet 65 Type-A Aortic Dissection After Previous Cardiac Surgery: Features and Optimal Management����������������������������  913 Olaf H. Stanger 66 The Elephant Trunk Concept in Type A Aortic Dissection����������������������������������������������������������������������������  921 Roberto Di Bartolomeo, Paolo Berretta, and Marco Di Eusanio Part IX Aortic Arch 67 Aortic Arch Surgery Under Warm Conditions (Moderate to Mild Hypothermia)������������������������������������������������  935 Ali El-Sayed Ahmad, Razan Salem, and Andreas Zierer 68 Infection of Ascending Aortic and Aortic Arch Prostheses ����������������������������������������������������������  943 Maximilian Luehr and Maximilian A. Pichlmaier 69 Elephant Trunk Technique(s)��������������������������������������������������������  965 Christian Hagl and Sven Peterss 70 Arch Surgery and Elephant Trunk Technique and Alternatives for Second Stage������������������������������������������������  987 Michael Z. Tong and Lars G. Svensson 71 Management of the Aortic Arch in the Modern Era������������������  997 Joshua B. Goldberg, Steven Lansman, and David Spielvogel 72 Debranching Concepts and Techniques for Arch Surgery���������������������������������������������������������������������������� 1009 Ehsan Benrashid, Nicholas D. Andersen, Richard L. McCann, and G. Chad Hughes 73 Total Endovascular Aortic Arch Replacement: Branched Arch Endografts ���������������������������������������������������������� 1027 Christine Herman and Cherrie Z. Abraham Part X Descending Aorta 74 Blunt Thoracic Aortic Trauma����������������������������������������������������� 1039 Mina L. Boutrous, Rana O. Afifi, Ali Azizzadeh, and Anthony L. Estrera

Contents

Contents

xv

75 Surgical Anatomy of the Blood Supply to the Spinal Cord�������������������������������������������������������������������������� 1049 Germano Melissano, Luca Bertoglio, Enrico Rinaldi, and Roberto Chiesa 76 Surgical Strategies in Acute and Chronic Type B Aortic Dissection �������������������������������������������������������������� 1061 Mark Field, Debbie Harrington, Omar Nawaytou, and Manoj Kuduvalli 77 Current Open Treatment of Thoracoabdominal Aortic Aneurysms�������������������������������������������������������������������������� 1075 Maral Ouzounian, Scott A. LeMaire, Scott A. Weldon, and Joseph S. Coselli 78 Spinal Cord Protection in Thoracic Aortic Surgery ������������������ 1091 Jean Bachet 79 Monitoring Spinal Cord Function in Open and Endovascular Treatment of Thoracoabdominal Aortic Pathologies������������������������������������ 1127 Barend Mees, Geert Willem Schurink, Noud Peppelenbosch, Werner Mess, and Michael Jacobs 80 Infected Aortic Grafts in the Descending Thoracic Aorta���������� 1143 Rana O. Afifi, Kristofer M. Charlton-Ouw, Hazim J. Safi, and Anthony L. Estrera 81 Hybrid and Redo Strategies for Descending and Thoracoabdominal Aorta������������������������������������������������������ 1157 Rana O. Afifi, Hazim J. Safi, and Anthony L. Estrera Part XI TEVAR 82 Endovascular Treatment of Aortic Diseases�������������������������������� 1171 Andreas Mitsis and Christoph A. Nienaber 83 Secondary Procedures After Primary Thoracic Endovascular Aortic Repair (TEVAR): Pathologies and Management ������������������������������������������������������ 1187 Michal Nozdrzykowski, Friedrich-Wilhelm Mohr, and Jens Garbade 84 Stenting of the Descending Aorta ������������������������������������������������ 1201 A. D. Godfrey, N. J. Cheshire, and C. D. Bicknell 85 Complications of TEVAR�������������������������������������������������������������� 1211 Rana O. Afifi, Ali Azizzadeh, and Anthony L. Estrera

1

A Brief History of Aortic Pathology and Surgery Olaf  H. Stanger

1.1

 he Earliest Descriptions T and Concepts

“What’s past is prologue” (William Shakespeare, “The Tempest,” Act 1, Scene 1). Several early authors are credited with the first mention of the aorta and its abnormal variations, but at times when even the function of vessels was obscure, it is difficult to say what role and relevance were attributed to their observations. The oldest written narrative of aortic pathologies is found in the Ebers Papyrus, named after the German Egyptologist Georg Ebers (1837– 1898), who purchased the 2 meter scroll with 108 columns of text in 1872  in Thebes. He brought the papyrus to Leipzig where it is still kept in the university library. It is among the oldest and most important medical papyri from ancient Egypt, written in hieratic script and dating to 1550 b.c., but believed to have been copied from earlier texts. The papyrus contains a basic description of the human heart with “vessels attached for every member of the body” [1]. Furthermore, the text identified peripheral and abdominal aneurysms, i.e., “when his abdomen palpitates, it is caused by a swelling therein” [2]. The term “aorta” seems to have first been applied by Aristotle (384–322 b.c.). In his works O. H. Stanger (*) Fmr Department of Cardiac Surgery, Royal Brompton Hospital, Imperial College London, London, UK

Historia animalium, De somno, and particularly De partibus animalium, Aristotle refers to the heart as the center and origin of blood connected with two great vessels (later identified as veins and aorta) and defined the aorta as primary (arterial) outflow of the left ventricle. He then describes all the other vessels as branches of the great vessel and the aorta, and he assumed a two-­ way blood transfer (blood produced from nutrition travels toward the heart only to be collected and distributed into the vessels) [3]. Whereas he envisioned the heart as the container and origin of blood, he stated that all blood leaves the heart but none returns, and so was unaware of circulation. Although he was not fully clear about the function of the mentioned vessels, the existence of a “great artery” and the “aorta” is clear. At that time veins and arteries were first distinguished, and the term “aneurysma” (áνεύρυσμα), meaning widening or dilation, was introduced. Next, the prolific Greek physician Galen of Pergamon (129/131–c. 200/216), a most prominent figure in medicine and arguably the most accomplished of all medical researchers in antiquity, upon physical examination described “localized pulsatile swellings” and a ruptured aneurysm: “when an aneurysm is wounded, the blood is spouted out with so much violence that it can scarcely be arrested” [4]. Galen also recognized distinct differences between venous (dark) and arterial (bright) blood.

© Springer-Verlag GmbH Austria, part of Springer Nature 2019 O. H. Stanger et al. (eds.), Surgical Management of Aortic Pathology, https://doi.org/10.1007/978-3-7091-4874-7_1

1

2

Although his anatomical experiments on animal models led him to a more advanced understanding of the circulatory system, his work contained scientific errors, i.e., the belief that the circulatory system consists of two separate one-way systems of distribution, rather than a single unified system of circulation. He thought venous blood was generated in the liver, whence it was distributed and consumed by all organs of the body. Similarly, he postulated that arterial blood originated in the heart, whence it was distributed and consumed by all organs of the body. The blood was then regenerated in either the liver or the heart, completing the cycle [5]. Galen’s understanding of anatomy was importantly influenced by Hippocrates’ then prevailing humoral theory, and his reports were the only anatomical reference for many centuries, although mainly derived from animal (monkeys, pigs) vivisections. Particularly Galen’s erroneous conception that venous blood passes through tiny pores in the heart’s septum, moves from the right to the left chambers, and is mixed with inhaled air from the lungs long inhibited any new thought. Galen’s doctrines dominated medical thinking for many centuries, and particularly his writings on anatomy became the mainstay of the medieval physician’s academic curriculum. They remained uncontested until printed descriptions and illustrations of human dissections were published in the seminal work De humani corporis fabrica by Andreas Vesalius (1514–1564) in 1543 and when William Harvey (1578–1657) published his treatise entitled De motu cordis in 1628, in which he established that the pumping heart drives blood circulation [6, 7]. The works of Antyllos, a Greek surgeon of the Roman period and contemporary of Galen (second century A.D.), have only survived in the writings of Or(e)ibasius from Pergamon (325–403), who collected most of the fragments of Antyllos’ works [8]. He himself wrote The Synagogue Medica and classified aneurysms as either due to dilatation of the arteries (with cylindrical form) or caused by rupture of the artery (with round form) emptying blood into tissues [9]. Antyllos allegedly belonged to the “pneumatist” medical school, described false traumatic and true aneurysms, and

O. H. Stanger

proposed that aneurysms were a consequence of clotting. He was the first to recommend surgical treatment of small ­peripheral aneurysms by proximal and distal arterial ligation followed by central incision of the sac and evacuation of the thrombotic material (Fig. 1.1) [10]. He did not resect the sac, considering it dangerous to do so, because pulsation puts violent tension on the ligatures, potentially displacing them and leading to fatal bleeding, and advised: “Those who tie the arteries, as I advise, at each extremity, but amputate the intervening dilated part, perform a dangerous operation. The violent tension of the arterial pneuma often displaces the ligatures.” He also opposed surgery for large aneurysms but did operate on peripheral aneurysms. Antyllos expressed the (still valid) dilemma in stating: “To decline treatment of any aneurysms is foolish, but it is also dangerous to operate on all of them” [8]. Antyllos’ detailed technique is the earliest record of therapy of aneurysms. Importantly, this was recommended as state of the art by Albucasis (Abu al-Qasim al-Zahrawi; 936–1013) 800 years later [11] and, in fact, was the best surgical treatment available until the end of the nineteenth century (Fig. 1.2) [12].

Fig. 1.1  Surgical treatment of aneurysms as described by Antyllos. Proximal and distal ligation followed by central incision of the sac and removal of thrombotic material (from Ref. [10], with permission)

1  A Brief History of Aortic Pathology and Surgery

3

Fig. 1.2  Technique by Antyllos is still the state of the art in the eighteenth century [12]

1.2

 he Great Leap Forward T from Antiquity

Advancements in the vascular field were negligible until the sixteenth century, with a new breed of scientists whose own studies and observations made them question the received wisdom from antiquity. Flemish physician Andreas Vesalius laid the overdue foundation of exact modern human anatomy and was the first author to rely solely on his own observations of actual human anatomical dissections (Fig.  1.3a–d) [13]. He could thereby identify many errors passed down from Galen and others that derived from conflicting information from animal vivisection. Vesalius identified aneurysms of the thoracic and abdominal aorta (and considered them untreatable) and diagnosed a traumatic aneurysm in a rider who fell off his horse [14]. His colleague and friend, French surgeon Ambroise Paré (1510–1590) of Paris, is best known for reintroducing (and establishing) vascular ligature as treatment of choice for injured blood vessels (Fig.  1.4a–f) [15]. He warned of opening an aneurysm due to the inevitable fatal bleeding and considered aneurysms of internal parts to be incurable: “we cannot cure large aneurysms of the armpit or groin, for on cutting into them so large quantity of the blood and vital spirits escapes that the patient dies” [16]. He thought of vascular calcifications as “a gift from God” to prevent rupture of the aneurysm. His was one of the earliest accounts of aneurysms presumably caused by the spirochete that causes syphilis, then called the “French disease”: “The aneurismaes which happen in the internall parts are uncurable. Such as frequently happen to those who have often had the unction

and sweat of the cure of the French disease, because the blood being so attenuated and heated therewith that it cannot be contained in the receptacles of the Artery, it distends it to that largenesse as to hold a man’s fist; Which I have observed in the dead body of a certain Taylor, who by an Aneurisma of the Arterious veine suddenly whiles hee was playing at Tennis fell downe dead, the vessel being broken; his body being opened I found a great quantity of blood powred forth into the capacity of the chest, but the body of the Artery was dilated to that largeness I formerly mentioned, and the inner Coate thereof was bony. For which cause within a while after I shewed it to the great admiration of the beholders in the Physitions Schole whilest I publiquely dissected a body there; the whilst he lived said he felt a beating and a great heate over all his body by the force of the pulsation of all the Arteryes, by occasion whereof he often swounded” [17]. German physician Daniel Sennert(us) (1572– 1637) dealt with aortic dissections as separation of the aortic wall layers in a broader context [18], and German barber surgeon Matheus Gottfried Purmann (1648–1711) operated on a series of antecubital aneurysms in the 1680s, putting ligatures above and below the aneurysm and opening and removing the sac (Fig.  1.5) [19, 20]. Giovanni Maria Lancisi (1654–1720) of Rome published De motu cordis et aneurysmatibus in 1728, providing descriptions of the etiology and pathology of aneurysms, including case studies (Fig. 1.6a, b) [21]. The founder of modern pathology, Giovanni Battista Morgagni (1682–1771), is best known for introducing the concept that every disease

O. H. Stanger

4

a

b

c

d

Fig. 1.3 (a–d) De humani corporis fabrica (Andreas Vesalius, 1543). (a, b) Title page. Vesal looks to us while performing an autopsy, (c) depiction of the main arterial

blood vessels of the human body, (d) descending aorta (arteria magna) with side branches [13]

1  A Brief History of Aortic Pathology and Surgery

a

5

b

d c

Fig. 1.4 (a–f) Opera chirurgia (Ambroise Paré, 1594). (a) Title page, (b) abdominal anatomy, (c) most arterial injuries were traumatic in origin, (d–f) surgical instruments suggested by Paré [15]

O. H. Stanger

6

e

f

Fig. 1.4 (continued)

Fig. 1.5  Chirurgia curiosa (Gottfried Purmann, 1699). Antecubital aneurysm [19]

originates in a distinct anatomical location, as indicated by the title of his most important work (Fig.  1.7a–c). He left a study on dissecting aneurysms in 1761 [22]. He reported several ­ cases in which blood forced its way through the wall “coming out under the external coat of the artery,” and on a patient with acute aortic dissections, “A man … was taken by pain in the right arm and shortly thereafter in the left arm, … soon

to be followed by a tumor on the upper part of the sternum. … He was instructed to think serious and humble of his departure from life, which was inevitable and very soon to occur” [22]. To Morgagni it was beyond doubt that the syphilitic toxin corrodes the vessel wall, causing dilatation. The previous year, King George II of England (1683–1760) had died suddenly at Kensington Palace from pericardial tamponade caused by a ruptured aortic dissection. The case was very accurately described in the autopsy report by Frank Nicholls (1699–1778), the King’s personal physician: “… the pericardium was found distended with a quantity of coagulated blood, nearly a pint…; the whole ­ heart was so compressed as to prevent any blood contained in the veins from being forced into the auricles; therefore the ventricles were found absolutely void of blood…; and in the trunk of the aorta we found a transverse fissure on its inner side, about an inch and a half long, through which some blood had recently passed under its external coat and formed an elevated ecchymosis” [23, 24]. The Italian surgeon-anatomist Antonio Scarpa (1752–1832), secretary to Morgagni, quoted the writings of Sennert in his own book and thought of atherosclerosis as the main driver for aneurysms (Fig. 1.8) [25]. Arterial injuries were common when phlebotomy and bloodletting were applied for a wide range of diseases. In the armpit, the needle frequently ­lacerated the brachial artery instead of

1  A Brief History of Aortic Pathology and Surgery

a

7

b

Fig. 1.6 (a, b) De motu cordis et aneurysmatibus (Giovanni Maria Lancisi, 1728). (a) Title page, (b) thoracic anatomy [21]

a

b

c

Fig. 1.7 (a–c) De sedibus et causis morborum per anatomen indagatis (Giovanni Battista Morgagni, 1761). (a) Portrait of Giovanni Battista Morgagni. (b) Title page. (c) Thoracic anatomy [22]

the veins, with false aneurysms, arteriovenous fistula, and potentially fatal rupture as sequelae. In the eighteenth century, riding boots often

caused painful aneurysms of the popliteal artery, which, unsurprisingly, were particularly common in coachmen.

8

Fig. 1.8  A treatise on the anatomy, pathology, and surgical treatment of aneurysms (Antonio Scarpa, 1808). Illustration of thoracic anatomy [25]

English surgeon John Hunter (1728–1793) is well known for his famed ligation of the popliteal artery. He acquired a critical attitude toward traditional medical practice under the influence of his teacher William Cheselden (1688–1752), who was fundamental in establishing surgery as medical science, breaking away from the Company of Barbers to form the Company of Surgeons in its own right, which later became the Royal College of Surgeons (of London, 1800; of England, 1843). Accordingly, Hunter’s work represents the emergence of surgery as a scientific discipline based on anatomy and physiology. He thought that an aneurysm develops when the arterial wall loses elasticity and becomes too weak to withstand the force of blood, which, however, does not say much about the cause. From experiments in peeling off the outer part of the carotid artery in dogs without aneurysm formation, he concluded that trauma was not an important cause [26]. For the treatment of femoral aneurysms, he assumed that ligation above and below the aneurysm would suffice, and perfusion to the lower limb would find its way through smaller collateral side vessels. To prove his theory, Hunter

O. H. Stanger

conducted a series of experimental femoral artery occlusions in animals [27]. Upon sacrifice a few weeks later, he injected colored resin into the artery, demonstrating collaterals ensuring sufficient perfusion. Based on this finding, he concluded that with sufficient collaterals, an artery could be safely ligated and that ligation should be done at a distance from the diseased part of the aneurysm to avoid erosion and ­rupture [28]. Opportunity for proof of concept came with the treatment of a popliteal aneurysm in a 45-year-old coachman, Samuel Smart, on December 12, 1785. Hunter put a ligature on the superficial femoral artery high in the thigh in the area now known as Hunter’s canal (Canalis adductorius). The patient survived for 15  months, the aneurysm having shrunk to a hard knot and the limb surviving. Afterward Hunter was able to buy the leg from Smart’s widow and found “a completely thrombosed aneurysm, somewhat larger than a hen’s egg” [29]. This postmortem specimen can still be seen in the Hunterian Museum of the Royal College of Surgeons in Lincoln’s Inn Fields (Fig. 1.9). Opposition to Hunter’s method was soon to come, and the surgical establishment including Percival Pott (1714–1788) intensively defended the common treatment of symptomatic popliteal aneurysm, which meant limb amputation [30]. Although, with reference to Antyllos, Hunter was not the first to treat peripheral aneurysms by ligature, he pioneered the idea of justifying application of surgical techniques on the basis of experimental evidence and clinical success. Anatomist William Hunter (1718–1783), John’s elder brother, had produced a manuscript in 1757 [31], in which he discriminated between true and false aneurysms, describing them as dilated and pulsatile vessels, and also named a third type of aneurysm “that was formed partially by a wound or rupture of some

1  A Brief History of Aortic Pathology and Surgery

Fig. 1.9  Popliteal aneurysm surgically treated by John Hunter (1785) (original specimen). Hunterian Museum of the Royal College of Surgeons in Lincoln’s Inn Fields

of the coats of the artery, and partly by a dilatation of the rest” [32]. He was the first to describe arteriovenous fistulae, predominantly the result of phlebotomy and injuries to the brachial artery, along with the hissing noise heard on auscultation. Another type of aneurysm was caused by infections, most frequently as a complication of syphilis, although their infectious nature was not yet known and they were rather attributed to a dissolute lifestyle, particularly among soldiers.

9

Post-venesection brachial artery pseudoaneurysms were treated surgically with proximal ligature by French surgeons Dominique Anel (1678–ca. 1730) [33] and Pierre-Joseph Desault (1744–1795) in 1785 [34]. John Hunter had taught that ligation could be used for aneurysms of the subclavian, carotid, and femoral arteries. One of his students was Sir Astley Cooper (1768–1841), who after experimentation developed retroperitoneal access to the aorta in a cadaver model. In 1805 he performed one of the earliest ligations of the right common carotid artery in a human. And in 1817, in analogy to Hunter’s concept, he ligated the distal aorta to control a large ruptured left-sided external iliofemoral aneurysm in a 38-year-old porter, expecting thrombosis and obliteration of the lesion. He managed to get his finger around the aorta through a small transperitoneal incision, passed a single heavy silk ligature around with a needle, and tied the knot. In consequence, however, one leg became ischemic, and the patient survived for barely 48  h. The lesson was that Hunterian ligation was appropriate for aneurysms of small- and intermediate-sized vessels but proved universally fatal in patients with aortic aneurysms. The postmortem specimen is preserved and remains on display in the Gordon Museum of Pathology at King’s College London (Fig. 1.10a, b). Jean-Nicolas Corvisart des Marets (1755– 1821), personal physician to Emperor Napoléon I (1769–1821), was prominent in medical circles and had an interest in cardiology. He published works on diseases and organic lesions of the heart and the great vessels [35]. Besides describing dilatative cardiomyopathy and congestive heart failure, he provided a detailed evolution of aneurysms of the aorta [36]. In an ascending aortic aneurysm, he described the thrill and retromanubrial dullness to percussion. French surgeon René Théophile Laennec (1781–1826) not only invented the stethoscope but reported several cases of chronic aortic dissection diagnosed using that instrument [37]. He

O. H. Stanger

10 a

b

c

Fig. 1.10 (a and b) Ligature of abdominal aortic aneurysm performed by Sir Astley Cooper 1817 (original specimen). Gordon Museum of Pathology at King’s College London

was the first to use the term “dissecting aneurysm” (“L’anévrysme disséquant”) (Fig.  1.11a, b) [37], although, a few years earlier in 1802, Jean Pierre Maunoir (1775–1830) had described the pathological process more precisely: “Elles se rompent dans un point, et la tunique externe ou celluleuse, fait poche et s’oppose seule à l’effusion du sang qui passe par la déchirure des tuniques internes” [38], but this went largely unnoticed. Single cases of fatal ascending aortic aneurysm ruptures were reported by Scottish surgeons Allan Burns (1781–1813) [39] and Joseph Hodgson (1788–1869) (Fig.  1.12a, b) [40]. In fact, these two works were the first and most important textbooks on heart disease and cardiovascular pathology in English, notably written by two surgeons. Burns had then dissected 14 cases of aortic aneurysm and gave an excellent account of the symptoms. While praising Scarpa’s work on aneurysm, he disagreed with the view that it always resulted from a localized rupture of the inner coat and described diffuse cylindrical dilatation of the aorta with intact coats, a condition later described by Hodgson [41]. In 1824 Adolph Wilhelm Otto (1786–1845) provided probably the first description of coarctation of the aorta complicated by

aortic dissection and the presence of a bicuspid aortic valve (BAV) (Fig. 1.13) [42]. In Vienna, Joseph Škoda (1805–1881), an expert in physical diagnosis and representative of the legendary Second Vienna Medical School, applied auscultation and percussion with unheard-of precision. He was once called for consultation to Pierre Louis Jean Casimir de Blacas d’Aulps (1771–1839), a French nobleman and former French Minister to Austria, who s­ uffered from unexplained abdominal pain. Whereas three other famous fellow authorities diagnosed liver disease, Škoda found a leaking abdominal aneurysm instead based on auscultation and percussion and predicted imminent death. His diagnosis was confirmed by necropsy soon afterward, including the precise dimensions of the aneurysm [43, 44]. His diagnostic skills were precise enough to allow the first pericardial puncture by Viennese surgeon Franz Schuh (1804–1865) in 1840 [45]. Descriptions of splitting of the aortic tunica media in cases of chronic dissection were presented by Scottish pathologist William ­ Henderson (1810–1872) [46] and the ­presence of a distal reentry with a false aortic lumen by Thomas Bevill Peacock (1812–1882) [47, 48].

1  A Brief History of Aortic Pathology and Surgery

a

b

Fig. 1.11 (a, b) De l’Auscultation Médiate, ou Traité du Diagnostic des Maladies des Poumons et du Coeur, Fondé Principalement sur ce Nouveau Moyen d’Exploration. (a) Title page. (b) One of the earliest mentions of “Anévrysme dissécant de l’aorte” (René Théophile Laennec, 1819) [37]

11

1.3

Concepts and Theories of Pathological Mechanisms

In 1839, Viennese pathologist Carl von Rokitansky (1804–1878) explained the difference between dissection and spontaneous rupture on the basis of two types of degenerative changes: either through delamination when the adventitia loses its supporting function for the inner walls or by longitudinal rents due to brittleness and breakdown in the intima and media [49–51]. For dissection, Rokitansky favored causal inflammation, while Karl Köster (1843–1904) suggested that mesarteritis extending along the vasa vasorum weakens the media [52]. Rudolf Virchow (1821–1902) in Berlin primarily thought of atheromatous ulceration as the cause of dissection [53]. Another German pathologist, Friedrich von Recklinghausen (1833– 1910), in 1883 explained dissection as a consequence of inflammation [54], with “molecular changes of the elastic structures or subcellular events” along with stress from elevated blood pressure occurring in the aortic wall [55]. Then in 1875 a relatively high prevalence of aortic aneurysms was observed in army soldiers as compared to sailors [56]. British army surgeon Francis Henry Welsh (1839–1910) had studied the records of 53 men who had died from ruptured aortic aneurysm and noted that twothirds had a documented history of syphilis [57]. Welsh felt this frequency was greater than would be expected in the general population and wondered whether syphilis could be the cause of the aortic aneurysm. It was not until 1905 that the German zoologist Fritz Schaudinn (1871– 1906) together with dermatologist Erich Hoffmann (1868–1959) in Berlin first observed, in autopsy specimens of aorta, the causative spirochete that later became known as Treponema pallidum [58]. Sir William Osler (1849–1919) coined the term “mycotic aneurysm” in 1885, and in 1909 he argued for syphilis as an important cause of aneurysm [59]. In 1920 Friedrich Ernst Krukenberg (1871–1946) first suggested rupture of the aortic vasa vasorum to be responsible for aortic dissection [60].

O. H. Stanger

12

a

b

Fig. 1.12 (a, b) Tables depicting aortic saccular and atherosclerotic aneurysms (Joseph Hodgson, 1815) [40]

Fig. 1.13  First description of aortic coarctation complicated by aortic dissection in the presence of bicuspid aortic valve (BAV) (Adolph Wilhelm Otto, 1824) [42]

Again, it was Thomas Peacock who reported on 19 cases of aortic dissection in 1843 [61], recognized the importance of the intimal tear, and hypothesized that dissection was the result of disruption of the “internal coats of the vessel” [61]. Peacock even described experiments in which fluids were injected between the adventitia and media of the aorta simulating dissection and observed a tendency to see the canal reopen into

the original vessel; he wrote that this might be seen as “an imperfect natural cure of the disease” [62]. He also noted the difference in prognosis between dissections originating in the ascending aorta and those in the descending aorta. Having continued to collect cases, he published a series of 80 cases of dissection, dividing the process into three stages: (1) rupture of the internal aortic coats, (2) dissection and possible external rupture, and (3) recanalization [61]. Peacock made great contributions toward the understanding of aortic dissection with his experiments and observations. In fact his data on mortality, gender and age distribution, location of dissection, and symptoms are in close agreement with current literature [24]. Thus Peacock [47, 48, 63], Baltic German pathologist Eugen Bostroem (1850– 1928) [64], and Franz Schede (1882–1976) [65] all proposed penetration of the aortic wall by blood entering from the lumen as the primary event of aortic dissection. In contrast, Victor Babes (1854–1926) and Teodor Mironescu (1876–1954) reported a case of dissecting

1  A Brief History of Aortic Pathology and Surgery

­mesaortitis [66] and had also observed cases of dissection without tears and thus questioned the theory of primary penetration of the aortic wall as proposed by Peacock [67] and others. They rather thought of primary cleavage of the media as the triggering event of aortic dissection. Intramural hematoma (IMH) was subsequently described by Austrian pathologist-anatomist Hans Eppinger (1848–1916) [68, 69]. Pathologist Eduard v. Rindfleisch (1836– 1908) described the breakthrough of blood into the vessel wall due to wear and tear at certain predilected sites [70]. He assumed that pathologically reduced elasticity and resistance of the wall produced this tendency to rupture. In 1934, pathologist Theodore Shennan (1869–1948) from Aberdeen published the data of the largest necropsy series (300 cases) collected at the time and proposed four separate causal theories for aortic dissection: mechanical, inflammatory, degenerative, and congenital [71]. He noted that primary degenerative changes in the media with subsequent loss of elasticity were an important factor underlying the dissection process. The series by pathologist Albert E. Hirst (1915–) in 1958 even included 508 such cases [72]. Both reports were fundamental in providing important clinical information and essential for the understanding of the etiology and pathogenesis. The classic form of dissection is defined as entry of blood into the wall of the aorta with subsequent separation of the mural layers. French pediatrician Antoine Marfan (1858– 1942) studied the symptoms of the syndrome that would later bear his name and also reported the first case of arachnodactyly in 1896 [73], but it was only in 1943 that Helen Taussig (1898– 1986) pointed out an association between Marfan disease and aortic medionecrosis [74]. Also in 1943, the association between Marfan syndrome and aortic dissection was first noted by Lewis E.  Etter (1901–1979) and Lewis Pellman Glover (1900–1953) [75]. Swiss physician Otto Gsell (1902–1990) reported an aortic wall pathology in 1928 characterized as cystic medionecrosis with focus on degeneration of the muscle elements in the media [76]. Nearly identical “idiopathic aortic medio-

13

necrosis,” later known as cystic medial degeneration, was described in detail by Austrian pathologist Jakob Erdheim (1874–1937) [77]. This pathology, characterized by vacuolization of the media with noninflammatory loss of muscle cells and elastic fibers in the arterial wall, was subsequently accepted as the underlying cause for aortic dissection and rupture. This, however, came to be questioned as experience with larger case series increased and difficulties surfaced due to lack of adequate control series as the histological appearance of the aorta varies considerably with age and even at different levels within the same aorta [78, 79]. From the time of Laennec, numerous theories of causality have developed, namely, that dissecting aneurysms are due to trauma, chronic high blood pressure, infection, degenerative changes, inflammation, or disease of the vessel wall (of either the intima or the media or both). However, the sequence of events in the course of dissection was (and still is) a matter of debate, suggesting multiple and possibly interacting etiological factors.

1.4

Endovascular Treatment with Foreign Bodies

The standard treatment for thoracic aneurysms in the eighteenth and nineteenth century was complete rest, systemic administration of potassium iodide, and starvation diet regulated according to Mr. Tufnell’s method [80]. Other recommendations included vinegar, iron perchloride, alcohol, zinc chloride, gelatin, sodium chloride, or ergot salts [81], albeit with inconsistent success and rather reflecting lack of better options. A new (indirect) approach to prevent rupture was the introduction of foreign material into the aneurysmatic lesion to induce clotting.

1.4.1 Needles This concept followed the idea that foreign bodies, i.e., needles, would induce irritation and inflammation followed by clotting, thus reduc-

14

ing flow into the aneurysm sac and stabilizing the aneurysm through subsequent obliteration of the artery. Surgeon Benjamin Philipps (1772–1838) caused clot formation in the femoral and carotid arteries of dogs by inserting needles (1832), later to be supplemented by electrical current (one needle attached to the copper, the other to the zinc pole of a galvanic battery) in an effort to increase clot formation [82]. After sacrificing his experimental animals, he found coagula around the needles and adherent to the vessel wall, which stimulated enthusiasm for the procedure. Simultaneously in Paris, Alfred-Armand Velpeau (1795–1867) conducted similar coagulation experiments [83]. Even aneurysms of the ascending aorta were treated with puncture for the sake of scratching the inner layer of the vessel and promoting thrombus, though without success [84]. Results with simple needle puncture were unpredictable, and the technique was ultimately abandoned, only to give way to wires and current.

1.4.2 Wires To further enhance clot formation, English surgeon Charles H. Moore (1821–1870) and British physician Charles Murchinson (1830–1879) began to pack aneurysms with wire. They were inspired by a fibrin-covered bullet recovered from an autopsy on a sailor who had died of a gunshot wound to the chest. The metallic bullet that Moore found within the ascending aorta was embedded in fibrin, and it was concluded that a foreign body would attract fibrin, support the mass entangling it, and lead to the eventual filling of the cavity of the aneurysm. A foreign body that would be less irritating and superior to simple needle insertion was thought to be a wire that might be passed in through a small cannula [85]. Moore and Murchinson were the first to attempt percutaneous endovascular aneurysm repair by causing sac thrombosis through direct needle cannulation and wire packing. In 1864 they inserted 26 yards (!) of wire coils into a large thoracic aneurysm via direct aneurysm puncture (Figs.  1.14 and 1.15) [84–87]. They recounted

O. H. Stanger

that they observed a declining pulse, reduction in the size of the aneurysm, and overall clinical improvement. One might call it a success because the aneurysm had indeed partially thrombosed; however, the patient succumbed to sepsis and distal embolism. At autopsy, the coils of wire were filled with “fibrinous coagulum” and were “firmly adherent” [85]. Other investigators amended Moore’s techniques by developing coils from different materials. Richard Levis (1827–1890) from Philadelphia and John Henry Bryant (1867– 1906) from London used horsehair to treat subclavian and popliteal aneurysms, respectively, but both patients suffered rupture shortly after treatment [86]. Other surgeons used iron wire, steel wire, silvered copper wire, gold wire, coil, and metal watch springs, but with none or very limited success (Fig. 1.16) [88]. Instead, the full range of complications became obvious, such as hemorrhage from subtotal packing and distal embolization of wire or thrombus [81]. Sir D’Arcy Power (1855–1941) attributed the disappointing and often variable results to underestimation of the aneurysm size and hence incomplete wire insertion. He therefore used the “Colt apparatus” invented by George Herbert

Fig. 1.14  Description of a case in which an aneurysm of the ascending aorta was treated with insertion of wire (1864) [85]

1  A Brief History of Aortic Pathology and Surgery

A

15

B

C POLYETHYLENE BAND

–

+

MATAS-1913 CINISELLI-1856

CLATWORTHY-1950

E

D

CANNULA FOR INTRODUCING WIRE UMBRELLA

WIRE UMBRELLA

MOORE-1860

F

COLT-1921

G

+

–

H COATED WIRE

I

FASCIAL PLUS

LUCITE TUBE

PEARSE-1928

BLAKEMORE-1950

REID-1924

HEP SEALY-1949

Fig. 1.15  Endovascular treatment with foreign bodies. Endovascular treatment concepts for blood vessel constriction in aneurysms [87]

Colt (1878–1957) of Aberdeen, which consisted of “a trocar and cannula, a ramrod, a tube and a wisp that contained fine steel wires that expanded to form a miniature umbrella” (Figs.  1.15 and 1.17a–c) [87, 89], and reported “a case of aneurysm of the abdominal aorta treated by the introduction of silver wire” in 1903 [90]. The Colt device was remarkably advanced for the time because it opened into a three-dimensional shape. Although ultimate results were poor, an author concluded that “wiring is a good method of relieving pain, … , but one that may not alter the natural history of the disease” [91].

1.4.3 Electrothrombosis (Galvanopuncture) The next step was galvanopuncture, or the attempt to induce inflammation and clot formation with electrical current around the electrodes. Others theorized that coagulation resulted from the oxidation of blood cells and proteins or from the chemical deposition of albumin or through decomposition of salts in the blood through the acid produced at the positive pole [81]. Whatever the exact mechanism is, it was thought that current could potentially lead to occlusion of aneurysms.

16

Fig. 1.16  Wire introduced into an aneurysm to promote coagulation (from Ref. [88], with permission)

In one of the earliest contributions in 1824, Charles Scudamore (1779–1849) passed a galvanic current through blood, which then formed a dense black coagulum at the positive electrode [92]. In 1832, German physiologist Johannes Müller (1801–1858) published investigations describing the effects of galvanism on blood and egg white, among others [93]. Peripheral aneurysms were sometimes treated then by electrocoagulation, but thoracic aneurysms were tackled in similar fashion only in 1846, first by Luigi Ciniselli (1803–1878), who is credited with having popularized the technique of galvanic puncture. In 1856 he published data from 50 cases involving electropuncture of aortic aneurysms with a mortality rate of only 14% (as compared with 33% for ligature) and a 50% success rate overall in comparison (Fig.  1.15) [81, 87, 94], stimulating interest in the technique. He also reported a case of aneurysm of the descending aorta that he treated by galvanopuncture across the chest. In 1870 he published 23 cases that he had collected over time, whereby 6 patients recovered (though 3 of them relapsed a few months later) [95]. The experiments by surgeon John Duncan (1839–1899) and Sir Thomas Richard Fraser (1841–1920) using egg albumin and canine arteries supported the concept of albumin decomposition as the mechanism of current-related coagulation [96]. Duncan used to introduce both poles (steel needles) into an aneurysm and pass a

O. H. Stanger

current for 20 min [97]. The positive needle was covered with gutta-percha (alternatively vulcanite), the negative with glass, and both were inserted into the aneurysm through the skin and thorax (Fig. 1.18a–c) [98, 99]. The electric circuit was closed with a battery of Bunsen cells. Typically, blood clotted and the puncture sites bled. Duncan concluded that the operation delayed death only slightly, if at all. Other surgeons also had fatal outcomes. Complications included distal migration of the wires, formation of emboli, end limb ischemia, sepsis, and formation of distal aneurysms from the altered hemodynamics [100]. Rather than giving up on this technique, Alfonso Corradi (1833–1892) of Bologna added electrical current in 1879 to Moore’s original wire work, with silver and copper wire coils causing (electro)thrombosis. The passage of current through the coil was intended to encourage thrombosis, and it was suggested that an e­ lectrical current applied to a permanently inserted metallic coil would combine the dual benefits of wire insertion and electrothrombosis. The method became widely referred to as “Moore-Corradi method” and was widely used for years [101]. Surgeon Joseph Ransohoff (1853–1923) of Cincinnati also used this procedure by passing electric current through wire to enhance coagulation [102]. Guy LeRoy Hunner (1869–1957) in 1900 compiled 28 cases of aneurysms of the aorta treated by wiring according to the Moore-­ Corradi method [101]. Although most cases at the time generally died less than 1 year after the procedure due to rupture and sepsis, one case in this particular series managed to survive for at least 38 years [103]. Several accounts of galvanopuncture in thoracic (arch) aneurysms were given, some with encouraging results [104], where it was felt that the walls of the sac have had become stronger, thus lessening the risk of external rupture. Others were less fortunate, with fatal outcomes [105]. Arthur Blakemore (1897–1970) of New York explained the unsatisfying results with underestimation of the aneurysm size and incomplete wire insertion. Subsequently he proposed a novel method of determining the amount of wire

1  A Brief History of Aortic Pathology and Surgery

17

a 1 – Sc 2

b

A

F

D E

G

B

C

c A

D C F

E

B

Fig. 1.17 (a–c) Colt’s apparatus for wiring aneurysms. (a) Early instrument with cages for wiring aneurysms; (b) diagrammatic section through Colt’s instrument III in situ. Sac of aneurysm (A), expanding cage (B), compressed cage (C), solder at the center of the cage (D), collar on the

cannula (E), cartridge (F), ramrod (G); (c) flag-labelled side view of Colt’s instrument No. 2. A. Fixed handle (A), moveable handle (B), reel (C), milling tool (D), coil of wire (E), stud (F) [90]

O. H. Stanger

18

b

a

Fig. 1.18 (a–c) Electrolysis. (a) Galvanopuncture needles. (a) Insulated by vulcanite. (b) Uninsulated and multiple. The size varies with the case. (b) Aortic aneurysm

before operation (left) and the same 2 months after the operation by electrolysis (right) [99]

required to achieve thrombosis. Instead of the size of the aneurysm, he used blood velocity as a guide to the amount of wire required by heating the wire to 80 °C and estimating wire length from the difference of the diminished current required for reheating it in a second step. The rationale was that the rate of cooling of the first segment of wire inserted was relative to the velocity of blood flow (Fig. 1.19a–c) [107]. It was still in 1938 that Blakemore rediscovered and applied the previously described method of wire and application of an electrical current to induce thrombosis of the aortic aneurysm sac and reported 11 so treated cases with thoracic or abdominal aortic ­aneurysms [108]. But in general, results from wire insertion in aneurysms were poor, and the method was ultimately abandoned. Summarizing the merits of electrolysis (electrothrombosis), English surgeon Timothy Holmes (1825–1907) noted that “the circumstances which are favorable to a perfect success occur very rarely in practice” [109]. Words of warning also came from David Agnew (1818–1892), Ransohoff, and Rudolph Matas (1860–1957) who cautioned against the wires [81]. Matas even described wire insertion as “semisurgical” or “quasimedical” and regarded galvanopunctures as technique that “appeal to us

more as placebos than as real remedies” [110]. Power believed that “electrolysis seemed reminiscent of a time when little was known of the physiological processes connected with the clotting of blood” [91], and Ransohoff stated that “electrolysis fails, as a rule.” Instead, he recommended total extirpation of superficial aneurysms [102]. Very clearly, new concepts were needed.

1.5

Dawn of a New Era

1.5.1 Endoaneurysmorrhaphy When Rudolph Matas of New Orleans reported an internal repair technique known as “endoaneurysmorrhaphy,” it represented a major step forward in the surgical treatment of aneurysms. He thought that aneurysms could be cured by a radical operation that would replace ligature and first performed his procedure in May 1888 on a patient with a large brachial artery aneurysm of the left arm [111]. After ligation of the proximal and distal arteries, an incision was made into the aneurysm and the clot removed. The orifices of the blood vessels that entered the sac were then sutured from within, which preserved the collateral blood supply to the extremity.

1  A Brief History of Aortic Pathology and Surgery

a

19

b

c

Fig. 1.19 (a–c) Progressive constrictive occlusion of the aorta with wiring and electrothermic coagulation. (a) A roentgen ray of the abdomen taken after wiring and electrothermic coagulation of a very large arteriosclerotic aneurysm. The lesion has been stabilized now in excess of 8 years since operation. Note concentration of wire at the upper aortic-aneurysm junction for its impedance effect. (b) The electrical equipment employed in electrothermic coagulation of aneurysms. The equipment used to convert

AC current to ungrounded DC current is illustrated on the left. Mounted on the portable table is an ohmmeter, ammeter, voltmeter, and ratiometer. The latter is calibrated to show the average temperature of wire imbedded within an aneurysm during heating. (c) Lilly capacitance manometric tracings taken simultaneously from the brachial artery and the femoral artery via fine plastic catheters. Note the rise in brachial artery pressure upon gradual occlusion of the aorta distal to the renal arteries [106, 107]

Matas subsequently described using obliterative, restorative, and reconstructive techniques of endoaneurysmorrhaphy (Fig.  1.20a–c) [112]. In the obliterative form (used mainly in fusiform aneurysms), sutures were placed from within the sac aneurysm so as to occlude the proximal and distal artery; the walls were sewn together to obliterate the sac [113]. The other two techniques were modifications preserving arterial patency and were used preferably in sacciform aneurysms. This could be achieved by placing a cath-

eter in the main arteries and obliterating the aneurysm sac around the catheter with sutures. With regard to the etiology of vascular aneurysms, Matas wrote that “the sins, vices, luxuries and worries of civilization clog the arteries with the rust of premature senility, known as arteriosclerosis or atheroma, which is the chief in the production of aneurysms” [114]. There was another major leap forward in aortic surgery when in 1923 Matas successfully ligated the infrarenal aorta proximal to a large leaking luetic

O. H. Stanger

20

a

a

b

a

b b c

c

c

a

b

Fig. 1.20 (a–c) Endoaneurysmorrhaphy as described by Matas. (a) Technique of obliterative endoaneurysmorrhaphy. The aneurysms opened to expose orifices of the parent artery and a large branch artery (a). These orifices are closed by sutures (b), and the aneurysmal cavity is obliterated by bringing the walls together (c). (b) Technique of restorative endoaneurysmorrhaphy. The aneurysm is opened to expose the communication with the parent artery (a). This opening is closed with a continuous suture

(b). The aneurysmal cavity is then obliterated (c). (c) Technique of reconstructive endoaneurysmorrhaphy. With the aneurysm opened widely, the communication with the parent artery is closed by suture, a portion of the aneurysmal wall being used to prevent narrowing of the arterial lumen (a and b). A segment of rubber tubing is used as a guide in placing the sutures (from Ref. [112], with permission)

abdominal aortic aneurysm [115]. The patient survived nearly 18  months, and this original report, of the first successful ligation of the abdominal aorta since Cooper in 1817, later was supplemented with very detailed pictures [116]. Over the years Matas acquired vast expertise in treating aneurysms and reported his personal experience of 620 cases in 1940 [117]. Of these 101 were endoaneurysmorrhaphies, and he further emphasized the importance of testing the

collateral potential before proximally ligating an aneurysm. The reconstructive endoaneurysmorrhaphy that involved removing the diseased part and reconstructing the tunnel through the remaining healthy part was used until direct repair with graft replacement was introduced in the 1950s. In the same year that Matas performed his first ligation on an abdominal aneurysm, French Surgeon René Leriche (1879–1855) stated that, “the ideal treatment of arterial thrombosis is the

1  A Brief History of Aortic Pathology and Surgery

replacement of the obstructed segment with a vascular graft” [118]. Later, in 1936, he advocated bilateral sympathectomy for treatment of aortic occlusive disease, but it was abandoned in due course. In 1948, he coined the term “Leriche syndrome” for occlusive disease of the terminal abdominal aorta [119]. Jose Goyanes (1876–1964) of Madrid performed the first successful replacement of a human artery (with an interposition graft from the popliteal vein) in 1906 to bridge an excised popliteal aneurysm [120]. Other surgeons such as James Hogarth Pringle (1863–1941) of Glasgow, Bertram Bernheim (1880–1958) of Baltimore, and Erich Lexer (1867–1937) soon followed, using saphenous vein grafts to bridge defects in popliteal and axillary arteries [121, 122]. Surgical treatment was hampered by difficult imaging and diagnosis; in fact, most cases of aortic dissection were postmortem findings [72]. Substantial advance was made with the introduction of clinical angiography with sodium iodide contrast medium by Barney Brooks (1884– 1952), at Vanderbilt University, in 1923 [123]. António Egas Moniz (1874–1955) of Lisbon performed the first cerebral arteriography in 1927 (although Moniz was nominated twice for the Nobel Prize for his groundbreaking work in cerebral imaging, it was his work in psychosurgery that won him the Prize in 1949), and fellow Portuguese Reynaldo dos Santos (1880–1970) used translumbar aortography in 1929 (Fig. 1.21a, b). These pioneering achievements preceded today’s imaging methods and remained the only clinical tools for early diagnosis at the time.

1.5.2 Fenestration In an effort aimed to relieve acute arterial ischemia in the lower extremities in patients with aortic dissection, David Gurin of Great Neck (1904–1992), James W.  Bulmer (1892–1975), Richard Derby (1881–1963, husband of President Theodore Roosevelt’s daughter Ethel), and colleagues performed the first fenestration through localized reentry in the right external artery in 1935 [125]. Upon opening the vessel

21

through the non-dissected anterior wall, they found the true lumen narrowed by the dissection. They then incised the intima and media from within the vessel, creating an opening into the false lumen with flow into the lower extremities after removal of the clamp; closing the vessel restored pulsation in the extremity. A minor modification of fenestration was proposed by Robert S. Shaw (1920–2003), who opened the abdominal aneurysm sac and extracted a soft clot from its lumen, so permitting free bleeding from above, and then created a small window into the true aortic lumen [126]. Shaw also coined the term “fenestration.” Whereas others like Matas operated on true aneurysms, the fenestration by Gurin was the first attempt to tackle acute aortic dissection. Nevertheless, flap fenestration was soon ­recognized to be palliative as it failed to restore the mural integrity of the ascending aorta and arch.

1.5.3 (Cellophane) Wrapping Cellophane film was invented by the Swiss textile engineer Jacques E.  Brandenberger (1872– 1954) in 1908. It was produced as a polymer of cellulose and subsequently became an invaluable material for waterproofing products. The ability of cellophane to constrict blood vessels was first demonstrated by physiologist Irvine Page (1901–1991) who, besides discovering the serotonin and the renin-angiotensin system (RAS), created an experimental model of hypertension by wrapping cellophane around dog’s kidneys, as first described in 1939 [127]. In a subsequent necropsy study of the wrapped kidneys, they were found to be shrunken and encased in a dense fibroblastic and collagenous layer 4  mm thick [128]. The development of polyethylene cellophane was an important breakthrough as it produced a more intense fibrotic reaction than other types of the polymer. Based on these observations, cellophane wrapping was further investigated by Herman E. Pearse (1899–1983) of Rochester using ordinary alcohol-soaked DuPont cellophane No. 300  T [129]. He demonstrated that cellophane

O. H. Stanger

22

angle of 12th rib and midline four fingerbreadths from midline

Fig. 1.21  Translumbar aortography. Left: Site of needle injection for translumbar aortography. Patient is prone. Right: Approximate area of aorta which is needled (from Ref. [124], with permission)

could gradually obliterate the lumen of blood vessels, such as the internal carotid artery or aorta in dogs. The reaction produced a partly hyalinized fibro-collagenous layer with progressive constriction and obliteration of the lumen. Cellophane was first used clinically for aneurysms in 1943 by Paul W. Harrison (1883–1962) and Jacob Chandy (1910–2007) who successfully treated two arteriovenous aneurysms of the subclavian arteries with cellophane, resulting in their gradual elimination (Fig. 1.22) [130]. First attempts to palliate the aneurysmal dilatation of a chronic dissection of the descending aorta with cellophane wrapping were reported by Osler Abbott (1912– 1976) [131] and James Edgar Paullin (1881– 1951), both at Emory [132]. W.  Dean Warren from Charlottesville (1924–1989) tried Orlon fabric (Fig.  1.23a, b) [133]; others used fascia lata (Fig.  1.24) [134], polyvinyl sponge, and dermal wrapping, but these were soon abandoned because the aneurysms grew relentlessly. Several reports indicated that pure polyethylene cellophane was nonreactive, whereas the standard

“impure” material obtained from the primary manufacturer, E.I.  DuPont Nemours Company of Wilmington, Delaware, proved highly reactive, according to John K.  Poppe of Portland (1911– 2012), who reported excellent results with the compound in treating syphilitic aneurysms [135–137]. Arguably the most prominent patient to receive cellophane wrapping for treatment of an abdominal aneurysm was physicist and Nobel laureate Albert Einstein (1879–1955). In December 1948, surgeon Rudolf Nissen (1896–1981) treated his “grapefruit-sized” abdominal aneurysm by wrapping it to induce a “foreign body reaction” potentially leading to scarring and reinforcement of the aortic wall, so limiting expansion. Einstein recovered and left the hospital to return home and continue his physics work symptom-free until he died from complications after the inevitable rupture more than 5 years later. Later, Michael E.  DeBakey (1908–2008) found polyethylene wrapping unsuitable as a treatment for aortic aneurysms and rejected the technique [138].

1  A Brief History of Aortic Pathology and Surgery

A

23

So far, the various procedures that have been proposed and used in the surgical treatment of aneurysms of the aorta have been classified into three major categories: (1) those designed to promote thrombosis and fibrotic organization by partial, complete, or gradual occlusion or ligation of the aorta, by the introduction of foreign material, or by the stimulation of periarterial fibroblastic reaction (cellophane), (2) endoaneurysmorrhaphy, and (3) extirpation of the lesion.

B

C D

E Fig. 1.22 Subclavian aneurysm cured by cellophane wrapping and fibrosis. Illustration of changes in aneurysm. Preoperative impression (A), immediately postoperative with applied cellophane (B), condition at time of first follow-up examination after 2  months (C), marked shrinkage of the aneurysm after 7  months (D), and last observation 11  months post operation (E) (from Ref. [130], with permission)

a

b

1.6

Working Toward Definitive Surgical Solutions

A new era in the treatment of aneurysms began in the 1950s with a shift from indirect (palliative) treatments to direct repair. The groundwork was

INNOMINATE A.

SUP. VENA CAVA DISSECTING ANEURYSM

ORLON FABRIC BAND

PUL. A.

RT. CORONARY A.

Fig. 1.23 (a, b) Aortic wrapping with Orlon fabric. Left: Photograph of the opened aorta at autopsy. Note (1) false aneurysm just below the subclavian artery, (2) transverse tear of proximal internal opening, (3) minor involvement of renal arteries, (4) Orlon prosthesis with

PERICARDIUM

surrounding fibrous sheath, and small thrombus at aortic suture line. Right: A band of Orlon is applied to the intrapericardial aorta (from Ref. [133], with permission)

24

Fig. 1.24  Wrapping of the aorta. Wrapping with fascia to prevent aneurismal expansion [134]

laid by French surgeon Alexis Carrel (1873– 1944) and Charles Guthrie (1880–1963), who experimented with homograft aortic substitutes and vascular anastomosis techniques in the early 1900s. In 1912, the Nobel Prize for Physiology and Medicine was awarded to Alexis Carrel “in recognition of his work on vascular suture and the transplantation of blood vessels and organs.” He had worked together with Guthrie in refining vascular anastomotic techniques for vein grafts in the arterial system, demonstrating that arterial suturing and reconstruction with xenografts were feasible. Carrel used grafts of vena cava to replace segments of the thoracic aorta in experimental animal models [139]. Recognizing the dangers of spinal cord ischemia, he used paraffin tubes as shunts for distal blood flow. Taken together, this was important and fundamental work for what was yet to come, but aortic replacement was far from reality. It took a long time until October 19, 1944, when Clarence Crafoord (1899–1984) of Sweden pioneered the resection of coarctation

O. H. Stanger

with the first successful end-to-end reanastomosis and restoration of continuity of the aorta [140], followed by Robert Gross (1905–1988), of Boston, on July 6, 1945 (Fig. 1.25a) [141], and Harris B.  Shumacker (1908–2009) [142]. Gross was also the first (in 1948) to successfully replace a longer segment of a resected coarctation with a preserved arterial homograft (Fig. 1.25b) [143, 144]. Before the introduction of extracorporeal circulation in 1953, direct excisional repair of the thoracic aorta was limited to cases where side clamping was possible. The first were direct excisions of aneurysms of the subclavian artery, the innominate artery, and the aortic arch. Denton A.  Cooley (1920–2016) did three spectacular operations as early as the 1940s, one in 1945 with clamping of the ascending aorta, excision of an eroded part, and oversewing of it. In 1949, as a resident working with Alfred Blalock (1899– 1964), he operated on a patient who had just recently undergone coarctation resection but then developed a massive and paper-thin false aneurysm of the right subclavian artery. Blalock at the time was away, and Cooley excised the aneurysm successfully. On his return to Baltimore, Blalock remarked that, “if you are confronted with a serious surgical problem that has no proven solution, take a trip to Hawaii and your resident will handle it” [145]. And then in 1951, just having joined DeBakey in Houston, he had the opportunity to resect an aneurysm of the aortic arch with the same tangential clamp-and-resection technique and oversewing of the defect. In fact, this is believed to have been the first aneurysm repair of its kind (Fig. 1.26a, b) [146] and became the preferred technique for sacciform aneurysms of the thoracic aorta. Henry Bahnson (1920–2003) in Pittsburgh reported the first successful excision of a saccular aneurysm of the ascending aorta in 1953 [147]. First repair attempts at dissections of the descending aorta by DeBakey et  al. were excision of the dilated part, reunion of dissected wall layers, and restoration of continuity using end-to-­ end anastomosis [148]. The same was attempted in the dissected ascending aorta with excision of the entry, followed by reunion of the dissected

1  A Brief History of Aortic Pathology and Surgery

a

25

b

Fig. 1.25 (a, b) Treatment of aortic coarctation with homologous grafts. Left: Ideal form of therapy for coarctation of the aorta; above, thoracic aorta with a rather short zone of constriction; below, removal of the narrowed area and reconstruction of a full-sized aortic pathway by end-­

to-­end, everting anastomosis using interrupted mattress stitches of silk. Right: above, the findings at operation; below, complete removal of involved segment and replacement by a graft (from Ref. [144], with permission)

layers both proximally and distally and ultimately creation of an end-to-end anastomosis, as done by Charles Hufnagel (1916–1989) and Peter W.  Conrad (1927–2013) [149] of Georgetown University (Washington, D.C.) and George C.  Morris (1924–1996) in Houston [150]. Alternatively, patch reconstruction after resection of the false channel was used [151].

Homograft replacement of the aorta had initially been used in children with congenital heart disease after resection of aortic coarctation, first by Henry Swan in Colorado (1913–1996) [155], Russel Brock (1903–1980) in London [156], Gross (Fig. 1.25b) [144], and Paul W. Schafer (1915–) in Kansas [157]. Just after it had become clear that the 3-year survival rate for patients with untreated abdominal aortic aneurysms (AAAs) was only 50%, with two-thirds of deaths resulting from aneurysmal rupture (Fig.  1.28) [158], several surgeons independently performed successful abdominal aortic aneurysm reconstruction within just 1 month’s time. On November 14, 1950, Jacques Oudot (1913–1953) performed the first homograft replacement of an obstructed (thrombosed) aortic bifurcation, followed by the first crossover bypass

1.6.1 (Homo)graft Interposition Another shift from direct excisional repair to graft replacement began with the use of cadaveric allografts (homograft) as no synthetic material was yet available. Based on early work by Carrel and Guthrie [152, 153] and Gross (Fig.  1.27) [154], homograft preservation was perfected, and artery banks were established in the 1940s and 1950s.

O. H. Stanger

26

a

Fig. 1.26 (a, b) Aneurysmectomy (innominate artery and adjacent aorta). Left: Incision for thoracocervical approach (inset). Aneurysm arising in the innominate artery at its origin from the aortic arch. A Crawford clamp has been applied tangentially across the superior border of the aortic arch to occlude the origin of the innominate artery and aneurysm. Distal control of circulation in the

Fig. 1.27  Early homograft preparation and implantation experiments. Graft of abdominal aorta (dog to dog). The graft section had been stored in 10% homologous serum and balanced salt solution for 6 days and then had been implanted into a recipient animal which was kept for 6  months before sacrifice (from Ref. [154], with permission)

in the same patient with insertion of a graft between the two external iliac arteries [159]. Autopsy of the patient 3  years later revealed a thrombosed homograft [160]. On February 26, 1951, Norman Freeman (1903–1975) and Frank Leeds (1914–2003) in San Francisco successfully treated an aortic aneurysm with a vein inlay autograft from the left common iliac vein

b

aneurysm is obtained by temporary occlusion with tape around the right common and subclavian arteries. Right: Lateral aortorrhaphy following excision of aneurysm. The supraclavicular extension of the aneurysm through the eroded manubrium has been inverted and the thrombus evacuated (from Ref. [146], with permission)

sutured into the abdominal aorta and iliac arteries and then wrapped the aneurysmal sac around the reconstruction for external support of the vein graft [161, 162]. And on March 2, 1951, Paul W. Schafer and Creighton A.  Hardin (1918–2013) in Kansas resected an abdominal aneurysm with an indwelling polyethylene bypass shunt after clamping the aorta and replacing it with a human homograft. The patient died after 29 days from a leak in the native aortic wall [157]. Freeman reported reestablishing circulation in the legs with a splenoiliac anastomosis as an extra-anatomic bypass technique [163]. But likely most memorably, on March 29, 1951, French surgeon Charles Dubost (1914–1991) successfully resected an abdominal (infrarenal) aortic aneurysm with a 15-cm-long homograft replacement via a left extraperitoneal approach (Fig. 1.29) [160, 164–166]. The patient survived for 8  years. The report from Paris in 1951 that an abdominal aortic aneurysm had been successfully resected greatly influenced surgeons throughout the world who, until then, had

1  A Brief History of Aortic Pathology and Surgery Survival rates Abdominal aortic aneurysm

100

87.9

90

79.1

80 70 Per cent

Fig. 1.28  Survival rates for patients with abdominal aortic aneurysm. Survival rates for traced patients who had abdominal aortic aneurysm as compared to the survival rates of the normal population of age 65 years (from Ref. [158], with permission)

27

Normal population

67.0

65.1 58.1

60

49.2

50 40 30

Abdominal aortic aneurysm

26.9

18.9

20

10.0

10 0 0

1

2

3

4

5

6

7

8

9

Years

Fig. 1.29  Charles Dubost. Diagram of Dubost’s first aortic aneurysm replacement with homograft (March 29, 1951) (from Ref. [160], with permission)

regarded such an operation as being beyond the limits of surgery. Although the first procedure by Schafer and Hardin resulted in the patient’s death after 29 days due to hemorrhage from a leak [157], the following operations by Dubost [164], Ormand Julian (1913–1987) in Chicago [167], Brock [168], DeBakey and Cooley [169], and Bahnson [147] were successful. Likewise, ruptured abdominal aneurysms were successfully treated

between March 1953 and December 1954 by Bahnson [170], Frank Gerbode in Stanford (1907–1984) [171], Cooley and DeBakey [172], and Hushang Javid in Chicago (1921–) [173]. The first replacement of a descending thoracic aortic aneurysm with resection and homograft replacement was performed on April 2, 1951 by Conrad Lam (1905–1990) in Detroit, and the patient survived for 3 months before succumbing to infection [174]. During the procedure, blood

O. H. Stanger

28

flow distal to the operative site was maintained through a polyethylene tube inserted into the lumen of the vessel above and below the aneurysm (Fig. 1.30a, b). Lam concluded that leaving the aneurysmal sac intact as Matas suggested predisposes to infection, and full resection would be the preferable technique. DeBakey and Cooley had removed abdominal aneurysms completely in their first patients in 1952 [169]. Removal was also strongly advocated by Bahnson in 1953 [147], and it became clear that resection and complete replacement of the diseased aorta would eventually be the ultimate treatment of choice. DeBakey and ­ Cooley had been developing techniques for complex aneurysm repair and spinal cord protection during thoracic surgery for some years prior, and they performed a successful resection and segFig. 1.30 (a, b) Resection of the descending aorta and replacement with homograft. (a) Lucite tube used to conduct blood through the graft during the suturing. (b) The operative procedure for resection and replacement of the descending aorta (from Ref. [174], with permission)

mental graft replacement for fusiform aneurysms of the descending aortic aneurysm on January 5, 1953 [175], followed by Shumacker and Harris in 1956 [176]. The next major breakthrough took place in 1954, when the Houston team performed a series of successful surgical treatments of dissecting thoracic aortic aneurysms (Fig.  1.31) [177]. DeBakey and his associates went on to accumulate vast clinical and surgical experience in the management of AD patients, reporting a 20-year follow-up of 527 surgically treated patients as early as 1980 [178]. Ironically, Michael DeBakey himself underwent and survived open surgery for type A aortic dissection at the age of 97, arguably the oldest patient in history to do so. Ascending aortic replacement required the development of cardiopulmonary bypass and was

a

b

SUBCLAVIAN A

GRAFT

LUCITE TUBE

DIAPHRAGM

1  A Brief History of Aortic Pathology and Surgery

29

a

d

false passage

proximal opening

b true aortic lumen

c

re-entry into aortic lumen obliterated false passage

Fig. 1.31  Surgical treatment of dissecting thoracic aortic aneurysm. Illustration showing the site of origin and extent of the dissecting process in the thoracic aorta (a). The aorta has been divided (b), the false lumen has been

obliterated distally (c), and proximally a segment of the inner layer is being excised to create a reentry passage. The anastomosis is completed (d) (from Ref. [177], with permission)

first performed in 1956, again by Cooley and DeBakey [179], successfully replacing the ascending aorta with a homograft. Replacement of the aortic arch, with its inherent risk of cerebral ischemia, was understandably more challenging. Schafer and Hardin in 1951 [157] and Cooley, Mahaffey, and DeBakey in 1955 [180] failed in performing arch replacements using bypass shunts and hypothermia only. It was only with cardiopulmonary bypass that DeBakey and colleagues were first able to successfully replace the aortic arch as reported in 1957 [181]. By now, every section of the thoracic aorta from the arch to the diaphragm had been resected successfully and replaced by homografts [182]. Enthusiasm for homografts had swelled, and use was widely accepted in the early 1950s but then waned because of short supply and difficulty with the harvesting and banking of the grafts but foremost because of frequent structural degeneration and late complications of the grafts [182]. In fact, short-term results up to 3 years had been gratifying, but long-term outcome with homografts was poor, and aorta banks began to be closed [144]. At this point it was obvious that further progress would not be possible without a suitable flexible

conduit to replace resected segments of the aorta, and a search was begun for a more stable, long-­ term, synthetic conduit material.

1.6.2 Synthetic Grafts The use of prosthetic grafts leads to a new standard of care, starting with Arthur Voorhees (1921–1992), who made his momentous contribution in 1952 using a vinyon-N cloth as a plastic arterial substitute, and ending with Michael DeBakey and Denton Cooley who refined the design of the Dacron graft in 1954. Arthur Blakemore was known for performing portacaval shunts in patients with portal hypertension and developed the Sengstaken-­ Blakemore tube for management of hemorrhaging esophageal varices [183]. He was used to depressing cases with copious bleeding (venous and arterial) and is quoted as having said, “The only time I worry about bleeding is when I can hear it” [184]. When dealing with arterial occlusive disease, he tried injections of vasodilator drugs and performed lumbar sympathectomies but was frustrated over abdominal aortic

O. H. Stanger

30

a­ neurysms, having tried wires in the 1930s [108], also using the Colt apparatus. Now in the 1950s, Blakemore was involved in discovering a suitable graft material for aortic surgeries. Working in his animal laboratory in 1947, Voorhees incidentally discovered that a silk suture inadvertently left in a ventricular cavity of an animal was covered with a slick layer resembling natural endocardial tissue cells and speculated that “a piece of cloth might react in a similar way” [185]. He was unaware of Julius Dörfler (1872–1952) [186] and Herbert W.  Carson (1870–1930), who had observed earlier that silk sutures left in the lumen of an artery become encapsulated by a fine veil-like coating [187], and of Guthrie, who 30  years earlier had suggested that an implant need serve only as scaffolding for ingrowth of host tissue [188]. But as these findings went largely unnoticed, the important step was made when Voorhees proposed the concept of a fabric tube that “had to be strong, inert, stable, of the right porosity, supple, and yet easily transversed by a fine needle” [185]. The idea was that a fine mesh cloth could be used as an arterial graft, with fibrin plugs forming to stop leakage of blood through the walls of the prosthesis [189]. His first artificial artery was fashioned from a silk handkerchief. Next he turned to a bolt of

vinyon-N cloth that worked even better. According to another source, an orthopedic resident, James Wallace Blunt (1918–2003), offered Voorhees the vinyon-N cloth after it had failed as a tendon replacement. It had originally been designed as sail or parachute cloth but proved too inert to hold dye. Voorhees constructed a tube resembling the silk model and began using it as an aortic prosthesis in dogs in demanding and tedious procedures. By the end of 1950, 30 dogs had received implants with satisfying early patency, and in 1951 he had enough material to publish an optimistic preliminary report [189]. Pore size turned out to be critical for ingrowth of fibroblasts, and without the latter, neo-­ endothelium could not form. In February 1953, Blakemore at Columbia Presbyterian Medical Center used a vinyon-N graft from his lab to replace a ruptured abdominal aneurysm, only because the local homograft bank was unable to supply material. It became the first synthetic graft ever used to replace the human aorta [184]. The outcome encouraged further implantations, and over the following year, 16 additional aneurysms were similarly treated with a 56% survival rate (Fig. 1.32) [190]. Nevertheless, vinyon-N rapidly gave way to competitive fibers with more favorable physical Crump clamp

Vinyon ‘N’ cloth

Vinyon ‘N’ tube

Aneurysm of aorta

a

b

c

d

Pattern of cloth

Fig. 1.32  Aortic tubes constructed from vinyon-“N” cloth. Use of reinforcing cuffs over the proximal aortic segment and reinforcing strips about the line of anastomosis (from Ref. [190], with permission)

1  A Brief History of Aortic Pathology and Surgery

properties, including Orlon, Teflon, Ivalon, Nylon, and finally Dacron. Norman Shumway (1923–2008) at Stanford University experimented with rolled sheets of polyvinyl sponge (Ivalon) [191], and Shumacker used layered Nylon, incorporating a thin polyethylene film for hemostasis [192], whereas the Houston group used braided Nylon tubes experimentally (Fig. 1.33) [193]. To properly respect these early achievements, one must realize that prostheses were far from being delivered perfectly and ­manufactured in all sizes. Instead, “tubes were cut and sewn in scrub rooms, … unsophisticated and often cranky” [194]. With all these fabrics, durability remained a problem. Some fibers deteriorated rapidly, while others failed to form a strong bond with surrounding tissues. While Crawford in Houston worked on a technique for freeze-drying human arteries taken from autopsies, in 1954  W.  Sterling Edwards (1920–2004) was inspired by Voorhees’ enthusiasm for synthetic cloth for arterial grafts. Telling one of his patients, who happened to be an execu-

31

tive at Chemstrand Corporation, about his ­difficulties with creating easy-to-sew and wrinkle-free nylon grafts, the patient helped set up a collaboration with a physical chemist at the company [195]. They soon introduced the concept of crimping cylindrical grafts to allow greater flexibility without kinking and to provide better handling characteristics [196]. The Edwards-Tapp braided nylon graft was manufactured by US Catheter and Instrument Corp., until Edwards switched his preference from nylon, with its disappointing durability and degeneration in the phase of body fluids, to Teflon because of its superior tensile strength profile. Teflon prostheses remained commercially available until 1979. However, it was the discovery and introduction of Dacron that opened a new chapter.

1.6.3 Dacron Tubes of various plastic materials were employed, but all were found to have certain

a

b

Fig. 1.33  Braided Nylon tube for implantation into the thoracic aorta. Left: Braided Nylon tube for implantation into the thoracic aorta. Right: Braided Nylon prosthesis 196 after implantation into the thoracic aorta (dog). (a) The outer connective tissue sheath has been peeled away

from prosthesis. (b) Longitudinal section of prosthesis and adjacent aorta showing loosely adherent, outer fibrous connective tissue sheath and the smooth adherent inner lining (from Ref. [193], with permission)

32

disadvantages owing to design and physical characteristics that limited their adaptability and hindered their practical application. Beginning in 1954, DeBakey and his group began to experiment with Dacron. This material was a polyester polymer that was developed around 1939 and had been introduced in the USA by E.  I. DuPont de Nemours and Company, Inc., in 1946. According to legend, DeBakey discovered the material in a department store more or less by accident when he was actually looking for nylon, but it was sold out and the clerk suggested Dacron instead. Eventually DeBakey came to prefer this textile and used it to create the first artificial arterial patches and tubes using his wife’s sewing machine (Fig. 1.34). After 2 years of testing on animals, DeBakey was satisfied with Dacron tubes that were easier to sew than vinyon [197, 198]. Cooperation with the Philadelphia College of Textiles and Science led to the development of a knitting machine capable of producing seamless knitted (instead of braided) Dacron grafts in various sizes and with bifurcations, made flexible by proper ­crimping [139, 199]. The new material was now widely used by the group around DeBakey, Cooley, Morris, Oscar Creech (1916–1967), and Crawford. In fact, clinical experience was so highly gratifying that DeBakey employed this graft exclusively. Within less than 4 years, the group had implanted more than 1000 synthetic grafts with a 90% success

Fig. 1.34  Michael DeBakey at home sewing a Dacron vascular graft (c. 1955)

O. H. Stanger

rate, and this new arterial substitute was introduced to the medical community in 1958 with the landmark paper reporting their highly satisfactory results [200]. DeBakey et al. had collected 803 cases of occlusive disease of the aorta and iliac and femoral arteries including 448 cases with aortoiliac (complete and incomplete) occlusion. At first, the group had also routinely performed lumbar sympathectomy as was standard at the time as a supplemental procedure but gave up on it because of the high incidence of distressing post-sympathectomy neuralgia. Ultimately, flexible knitted Dacron tubes were judged to be the best arterial substitute available and came into wide use [200]. A review of chemical and physical data as well as in  vivo experiments on a wide range of fabrics in 1955 concluded that “Dacron appeared to have the most desirable qualities in the overall evaluation” and was thus the best material for aortic substitution [198]. In 1956, vinyon-N was no longer commercially available, and both nylon and Orlon exhibited significant loss of tensile strength over time [201]. The Meadox Weaving Corp., an upholstery and drapery fabric manufacturer in New Jersey, collaborated with Ormond Julian and Ralph Deterling (1917–1992) of New  York to design and fabricate grafts. Beginning in 1954, they produced the first woven grafts, and in 1961, Meadox Medical Inc. teamed up with Cooley to produce a graft line carrying his name [202]. Bleeding control remained an issue, particularly in the fully heparinized patient undergoing CPB.  Grafts needed to be tightly woven with low porosity but at the cost of less desirable handling and suture characteristics. Cooley introduced the method of autoclaving a porous graft soaked with autologous plasma, which renders it completely impervious to blood. Better sealing later became available, including impregnation with bovine collagen or albumin. Since Cooley and DeBakey’s first successful replacement of the ascending aorta with a tube graft [179], this has become the standard procedure for dealing with dissecting aneurysms and chronic nondissecting dilatations.

1  A Brief History of Aortic Pathology and Surgery

1.7

The Aftermath

The ground had now been laid by the availability of a reliable substitute and the concept of complete removal of the diseased segment. Since then, too many technical advances have been made to cover them all, and to mention the names of the many who have contributed to our current understanding of aortic disease and our management concepts would be beyond the scope of this survey.

a

33

Clearly, the most fundamental advancement was the development and introduction of the cardiopulmonary bypass by John H.  Gibbon (1903–1973) in 1953 (Figs. 1.35 and 1.36) [203– 209]. DeBakey’s contribution, while still in medical school in 1932, had been to assemble a hand-cranked roller pump, first used to transfuse blood directly from a donor to a patient and later adapted for use in the heart-lung machine. In 1957 Cooley introduced the left heart bypass to replace the descending aorta [210].

c

b

Fig. 1.35 (a–c) John Gibbon’s heart-lung machine. (a) Equipment used by John Gibbon in early laboratory experiments in extracorporeal circulation. (b) Heart-lung

machine Gibbon Model I (1949). The first oxygenator built by IBM. (c) Heart-lung machine Gibbon Model II (1951) (from Ref. [207], with permission)

O. H. Stanger

34

Innominate a.

L.atrial cannula Teflon prosthesis

Caval cannula

Pump oxygenator

Femoral a.cannula

Venous pressure

Fig. 1.36  Cardiopulmonary bypass. Plan of cardiopulmonary bypass used for prosthetic replacement of the ascending aorta (from Ref. [208], with permission)

Routine use of cardiopulmonary bypass greatly simplified aortic surgery allowing, among other things, controlled (deep) hypothermia and brain perfusion. Wilfred G.  Bigelow (1912–2005) of Toronto developed the idea of reducing a patient’s body temperature before an operation to lower metabolism and oxygen need [211]. After basic research in animal models, Floyd John Lewis (1916–1993) at the University of Minnesota performed the first successful human open-heart operation (September 2, 1952), closing an atrial septum defect in a child, after inducing hypothermia by wrapping the child in cooling blankets [212]. Notably, this and subsequent operations, also by Henry Swan, who carried on with this technique, were performed without cardiopulmonary bypass in large patient series. As mentioned

before, Cooley failed in replacing an aortic arch using bypass shunts and hypothermia only [180], and a solution for organ protection, particularly the brain, remained paramount. After cardiopulmonary bypass became widely available, C.  Walton Lillehei (1918–1999), who had participated in Lewis’s historic operation, and John W.  Kirklin (1917–2004) in Rochester observed spontaneous cooling in patients undergoing surgery with beneficial consequences. Rather than relying on this “side effect,” Will C. Sealy (1912– 2001) at Duke University introduced the heat exchanger to the DeWall oxygenator for controlled induction of hypothermia and rewarming [213], progressively allowing more complex and time-consuming procedures. Donald Ross (1922–2014) and Brock in London advocated the use of deep hypothermia induced with the heartlung machine [214] and so stimulated wider interest in further study of this technique. The breakthrough for wide acceptance came with the work of Randall B. Griepp (1940–) and colleagues at Mount Sinai in New  York. The introduction of deep hypothermic circulatory arrest (DHCA) in the mid-1970s dramatically reduced the incidence of neurological damage following aortic surgery [215]. Griepp, however, also stressed the limits of hypothermic circulatory arrest for cerebral protection [216]. Subsequently, the Crawford [217, 218] and Cooley [219, 220] groups used deep hypothermia for arch interventions and also suggested using moderate hypothermia for open repair of proximal aortic arch anastomoses [221]. Nicholas Kouchoukos (1937–) in St. Louis pioneered the use of profound hypothermic circulatory arrest for repair of descending thoracic and thoracoabdominal aneurysms [222, 223]. The combination of cardiopulmonary bypass and total circulatory hypothermic arrest provided a major advance that greatly enhanced the safety of distal aortic procedures. The use of cerebral perfusion was reconsidered and then revived by William H.  Frist (1952–) and colleagues [224]. Subsequent advances focused on improving brain protection by defining an approximately 30-min time limit for circulatory arrest [225], which could be

1  A Brief History of Aortic Pathology and Surgery

extended with cerebral perfusion techniques such as uni- and bilateral retrograde (RCP) and antegrade cerebral perfusion (ACP) [226]. To simplify clinical management of aortic dissections, many classification systems were suggested, but only the Stanford and DeBakey nomenclatures have prevailed over the time. The “Stanford classification” differentiates among aortic dissections based on whether the ascending aorta is involved, regardless of the site of tear and irrespective of the distal extent of dissection [227]. The “DeBakey classification” [228], which was modified in 1982 to more closely resemble the Stanford classification [178], classifies dissections not involving the ascending aorta as type III; those limited to the ascending aorta are DeBakey type II, and dissections involving the ascending, arch, and descending aorta are classified as type I. Treatment of aortic dissection was greatly influenced by Myron W.  Wheat (1924–2012) and others who evaluated the merits of open versus pharmacological management against the background of persistently high operative mortality. In contrast to aortic dissection with involvement of the ascending aorta, the majority of patients with uncomplicated type B aortic dissection treated medically were found to survive the acute phase, thus giving rise to medical therapy rather than surgery [229, 230]. Management of patients with type B dissection was fundamentally modified and later came to include interventional treatment modalities. Wheat in 1965 also emphasized the role of blood pressure control in the medical management of acute aortic dissection [229], still the mainstay for aortic dissections in absence of complications. Treatment of proximal aortic dissection with concomitant valve insufficiency was managed by narrowing of the annulus and valve bicuspidalization [133] and with the concept of commissural resuspension and attenuation of the sinotubular junction [149, 231]. Ross, in 1962, and Sir Brian Barratt-Boyes (1924–2006) in 1964 successfully implanted the aortic homograft in the orthotopic position [232, 233]. Albert Starr (1926–) in 1963 excised the incompetent aortic valve in aortic root aneurysm, replacing it with a Starr-Edwards valve and

35

replacing the aneurysmal ascending aorta with a graft [234]. In 1964 Wheat reported the first successful replacement of the entire ascending aorta including the valve with a separate Starr-Edwards valve and a woven Teflon aortic prosthesis; a flap of aortic tissue around the coronary ostia was left to incorporate into the graft [235]. Some patients, however, required replacement of the aortic root as well. Subsequently, combined operations were introduced that replaced the ascending aneurysm in conjunction with replacement of the aortic valve and reimplantation of the coronary arteries. In 1968, Hugh Bentall (1920–2012) at Hammersmith Hospital and Anthony De Bono reported their technique for complete replacement of the ascending aorta, using a composite mechanical valve and a Dacron conduit with reimplantation of the coronary ostia (Fig. 1.37) [236]. In cases with ascending aortic aneurysms with associated functional aortic insufficiency (but otherwise normal cusps), “aortic valve sparing operations” were developed with the aim of preserving the native aortic valve. These procedures are known as the “reimplantation” technique as introduced by Tirone E. David (1944–) of Toronto [237, 238] and the “remodeling” technique as described by Sir Magdi Yacoub (1935–) in London [239, 240] in the early 1990s. Surgery on the aorta, except for its arch portion, had become well established in the 1960s, including introduction of the “island technique” of brachiocephalic vessel reattachment, which simplified the procedure and reduced the number of anastomoses required [241]. However, the risk of multiple-stage operations required for the frequently encountered aneurysms extending distally from the aortic arch remained a problem. In 1983, Hans Georg Borst (1927–) in Hannover introduced the two-stage elephant trunk principle to simplify the second stage by leaving an extended vascular graft free within the descending aorta during the first-stage operation [242, 243]. The technique was then refined [244] and later complemented with the “frozen elephant trunk” technique to allow repair of concomitant aortic arch and proximal descending aortic aneurysms in a single-stage procedures with a

O. H. Stanger

36 Teflon 5

1

2

Starr valve

Coronary perfusion 3

4 Aortic ring

Fig. 1.37 Complete replacement of ascending aorta (Bentall and De Bono). Left: Starr valve has been sutured to aortic prosthesis: sutures have been placed in aortic ring before fixing the combined prostheses. Right: Combined prostheses in situ. Insets 1–4 show details of holes fash-

“hybrid” vascular graft [245]. One of the most promising recent innovations in aortic arch repair is the “trifurcated graft” technique (Fig.  1.38) [246, 247], along with a large number of debranching hybrid repair concepts using concomitant endovascular stent grafts [248, 249]. Arguably the last major development to date in treating aortic disease has been the evolution of endovascular stent grafting. That new era of treatment started in 1986 when an alternative to surgically placed grafts emerged. Harrison Lazarus (1939–) of Salt Lake City had conceived and essentially completed the design of an endovascular graft for abdominal aneurysm ­ repair by the mid-1980s and filed for a US patent in 1986 (awarded in 1988) [250, 251]. The pioneering clinical work is first and foremost ­ associated with Nikolai Volodos (1934–) in Kiev, Ukraine [252–255], and Juan Parodi (1940–) of Argentina [256]. Volodos reportedly performed the ­first-­ever aortic repair with a stent graft in the 1980s (in a patient suffering from a post-traumatic aneurysm of the distal descending thoracic aorta) (Fig. 1.39) [255], but this pioneering work

ioned in the side wall of the Teflon tube to reincorporate the coronary ostia within the lumen of the new ascending aorta. Inset 5 shows the vertical slit in the prosthesis (from Ref. [140], with permission)

Fig. 1.38 Aortic arch replacement with a trifurcated graft. With the main limb of the trifurcated graft clamped, antegrade selective cerebral perfusion is initiated through the axillary artery. The elephant trunk technique is used to reconstruct the arch, and the graft is anastomosed to the proximal repair. The trifurcated graft is then anastomosed to the reconstructed aorta (from Ref. [247], with permission)

has become widely known in the Western world only since the mid-1990s. Argentinian Julio Palmaz (1945–) of San Antonio invented and patented the balloon-expandable stent, which was later approved for use in peripheral arteries in 1991. Human endovascular abdominal

1  A Brief History of Aortic Pathology and Surgery

Fig. 1.39  Pioneer of vascular stent graft design. Nikolai L. Volodos (from Ref. [255], with permission)

Fig. 1.40  Thoracic aortograms obtained before and after stent graft placement over the primary entry tear in aortic dissection. Left: Before stent graft deployment showing flow of contrast medium from the true lumen (T) across

37

a­neurysm repair (EVAR) was performed by Parodi and associates in Buenos Aires and Palmaz on September 6, 1990, following extensive experiments with stainless steel stents hand sewn to thin-walled Dacron tube grafts [257, 258]. They launched the modern endovascular treatment revolution that led to profound transformations that have changed everything in vascular surgery. Treatment modalities for thoracic aneurysms followed successful repairs of abdominal aortic aneurysms. Michael D.  Dake at Stanford reported the first endovascular thoracic (descending) aortic repair with a homemade endograft in 1994 [259, 260], soon followed in 1999 by a report of earliest clinical experience with endovascular stent graft intervention for acute type B aortic dissection (Fig.  1.40) [261]. As such, TEVAR intervention has added an entirely new dimension to the management of AD and aortic disease (Fig. 1.41) [262–264]. Rapid diagnosis and appropriate management decisions were greatly advanced with improved

the entry tear (arrow) into the false lumen (F). Right: After stent graft placement. Only the true lumen is evident (from Ref. [261], with permission)

O. H. Stanger

38

a

b

c

Fig. 1.41 Stent graft with side branches. Left: A. Branched stent graft with one free-flow stent, one internal sealing stent, followed by reducing stents, from which emerge the proximal side branches. B. Detail of the emergence of the side branches from the main stent graft module. C. Iliac side branch device. Right: The

three-dimensional CT angiography reconstruction shows a thoracoabdominal stent graft with four branches, in adjunct to left internal iliac artery revascularization with iliac side branch device and contralateral embolization of internal iliac artery (from Ref. [262], with permission)

imaging modalities, such as CTA, CT, TEE, TTE, and MRI. Today’s clinical management is further greatly enhanced by recognition of “aortic syndromes” and distinct pathologies such as penetrating atherosclerotic ulcers (PAU), IMH, aortic ruptures, and dissection, apart from congenital syndromes. The author is well aware that this brief review of aortic surgery cannot be complete, especially with

regard to all the individuals who have contributed to the field over the centuries: “history, as it lies at the root of all science” can be understood as “the essence of innumerable biographies” [265]. This textbook looks deep into the past and, from there, toward the future with all its promising new directions, and as such it is to be hoped that it will help to take the history of aortic management one step further.

1  A Brief History of Aortic Pathology and Surgery Acknowledgment I wish to thank Genie Lamont (Graz, Austria), William Edwards (Curator, Gordon Museum, King’s College, London), Matt Nicholson (Hunterian Museum, The Royal College of Surgeons of England, London), Susan Y. Green at Baylor College of Medicine (Houston, Texas) and the librarians of the Josephina Library, Institute for the History of Medicine (Medical University of Vienna, Austria) for their valuable support.

References 1. Ghalioungui P.  The Ebers papyrus: a new English translation, commentaries and glossaries. Cairo: Academy of Scientific Research and Technology; 1987. 2. Bryan CP.  The papyrus Ebers. Translated from the German version. Chapter XX, the heart and circulatory system. London; 1930. p. 129. 3. Oser-Grote CM.  Aristoteles und das Corpus Hippocraticum. Stuttgart: Steiner; 2004. 4. Galen J. Observations on aneurysms. Translated by John Eric Erichsen. London; 1986. 5. Coxe JR.  The writings of Hippocrates and Galen. Epitomised from the original Latin translations. Philadelphia: Lindsay and Blakiston; 1846. 6. Furley DJ, Wilkie JS.  Galen on respiration and the arteries. Princeton, NJ: Princeton University Press; 1984. 7. Bylebyl JJ. William Harvey and his age. Baltimore: Johns Hopkins University Press; 1979. 8. Raeder J.  Oribasii collectionum medicorum reliquiae. Vol. III, Libri XLV, Cap. 24 (Corpus Medicorum Graecorum). Berlin; 1931 9. Daremberg C. Oevres d’Oribase. IV. Paris; 1867. 10. Sachs M.  Die Methoden der Blutstillung in ihrer historischen Entwicklung. Hamostaseologie. 2000;20:83–9. 11. Abu-al-Quasim: On surgery and instruments. English translation and comments by MS Spink and GL Lewis. London: The Wellcome Institute; 1973. 12. Masotti D. Dissertazione sull’aneurisma del poplite. Florence; 1772. 13. Vesalii A. De humani corporis fabrica libri septem. Padua; 1543. 14. Bonetus T.  Sepulchretum anatomia practica ex cadaveribus morbo denatis. Book 3, Sect 2. Lugduni (Lyon); 1700. p. 290. 15. Paré A. Opera chirurgia. Frankfurt; 1594. 16. Malgaigne JF.  Ouvres d’Ambroise Paré. Livre V. Paris; 1840. 17. The collected works of Ambroise Paré. Translated out of Latin by Thomas Johnson from the first English edition. London; 1634. p. 287. 18. Sennertus Daniel. Opera. Op Omn Lib 5,1, Cap 42, p.306. Lugduni (Lyon); 1650. 19. Purmann G. Chirurgia curiosa. Frankfurt; 1699.

39 20. Thompson JE. Early history of aortic surgery. J Vasc Surg. 1998;28:746–52. 21. Lancisi, GM.  De motu cordis et aneurysmatibus, opus posthumum in duas partes divisum. Rome; 1728. 22. Battista MG.  De sedibus et causis morborum per anatomen indagatis [the seats and causes of diseases investigated by anatomy; in five books, containing a great variety of dissections, with remark]. Ep 26, Art 17 (Ep 48) F 50. Venice; 1761. 23. Nicholls L. Observations concerning the body of his late majesty. Phil Trans (London). 1761;52:265–74. 24. Leonard JC.  Thomas Bevill peacock and the early history of dissecting aneurysm. Br Med J. 1979;2:260–2. 25. Scarpa A. A treatise on the anatomy, pathology and surgical treatment of aneurysms. Translated by John Henry Wishart. Edinburgh; 1808. 26. Hunter J.  Treatise on venereal disease 1786. In: Oevres complètes de John Hunter. Traduit par Gustav Richelot. Paris; 1843. 27. George Q. John Hunter 1728–1793. London; 1981. p. 109ff. 28. Home E.  An account on Mr Hunter’s method of performing the operation for the popliteal aneurism. Communicated in a letter to Simmons by Mr Eduard home, surgeon. Med J (London). 1786;7:391–406. 29. Schechter DC, Bergan JJ. Popliteal aneurysm: a celebration of the bicentennial of John Hunter’s operation. Ann Vasc Surg. 1986;1:118–26. 30. Pott P. Remarks on the necessity and property of the operation of amputation in certain cases. Chirurgical works. London; 1779. 31. Hunter W. A history of aneurysms of the aorta, with some remarks on aneurysms in general. Trans Med Obstet Soc Phys. 1757;1:323–57. 32. Chitwood WR.  John and William hunter on aneurysms. Arch Surg. 1977;112:829–36. 33. Sloffer CA, Lanzino G.  Historical vignette. Dominique Anel: father of the Hunterian ligation? J Neurosurg. 2006;104:626–9. 34. Stimson LA. An inquiry into the use of the ligature in the treatment of aneurysm. Ann Surg. 1885;1:13–25. 35. Corvisart JN.  Essai sur les maladies et les lésions organiques du cœur et des gros vaisseaux. Paris; 1806. 36. Karamanou M, Vlachopoulos C, Stefanadis C, Androutsos G. Professor Jean-Nicolas Corvisart des Marets (1755–1821): founder of modern cardiology. Hell J Cardiol. 2010;51:290–3. 37. Laennec RT. De l’Auscultations Médiate, ou Traité du Diagnostic des Maladies des Poumons et du Coeur, Fondé Principalemente sur ce Nouveau Moyen d’Esploration. Paris; 1819. 38. Maunoir JP. Mémoires Physiologiques et Practiques sur l’Aneurysme et la Ligature des Artères. Géneva; 1802. 39. Burns A. Observations on some of the most frequent and important diseases of the heart. Edinburgh; 1809. p. 231.

40 40. Hodgson J. A treatise on the diseases of arteries and veins, containing the pathology of aneurisms and wounded arteries. London; 1815. p. 63. 41. Bedford DE.  The surgeon cardiologist of the 19th century. Br Heart J. 1967;29:461–8. 42. Otto AW.  Neue seltene Beobachtungen zur Anatomie, Physiologie und Pathologie gehörig. Berlin; 1824. S. 66. 43. Sakula A.  Joseph Skoda 1805–81: a centenary tribute to a pioneer of thoracic medicine. Thorax. 1981;36:404–11. 44. Fejfar Z, Hlavačkova K. Joseph Skoda. Clin Cardiol. 1997;20:740–1. 45. Stanger O.  Development of modern physical diagnosis and its application in cardiothoracic surgery. Chirurg. 2001;72:853–60. 46. Henderson: on organised healing of dissections. Edinb Mon J. 1843;3:613. 47. Peacock TB. Cases of dissecting aneurysm, or that form of aneurysmal affection in which the sac is situated between the coats of the vessel. Edinb Med Surg J. 1843;60:276. 48. Peacock TB.  Partial rupture of the aorta and separation of the coats by the extravasated blood: “dissecting aneurysm”. Trans Path Soc London. 1947;25:59. 49. Rokitansky VK. Über spontane Zerreissungen der Aorta. Schmidts Jahrbuch. 1839;24:30–3. 50. Rokitansky VK. Über einige der wichtigsten Krankheiten der Arterien. Wien; 1852. 51. Rokitansky C.  Handbuch der pathologischen Anatomie, Vol 3. Wien; 1842–1846. 52. Köster K. Über die Entstehung der spontanen Aneurysmen und die chronische Mesarteriitis. Berl Klin Wschr. 1876;12:322–6. 53. Virchow R. Handbuch der speciellen Pathologie und Therapie. Erlangen; 1854. 54. Recklinghausen FD.  Auserlesene pathologisch-­ anatomische Betrachtungen. Virchows Arch Pathol Anat Physiol. 1864;30:372–87. 55. Recklinghausen FD.  Handbuch der allgemeinen Pathologie des Kreislaufs und der Ernährung. Stuttgart; 1883. 56. Myers ABR.  Aortic aneurysm. Probable cause of its great prevalence in the army compared with the navy and male population, and consequent remedy. Lancet. 1869;1:250–5. 57. Welch FH. On aortic aneurysm in the royal army and the conditions associated with it. Med Chir Trans. 1876;59:59–78. 58. Schaudinn F, Richard; Hoffmann E.  Vorläufiger Bericht über das Vorkommen von Spiochaeten in syphilitischen Krankheitsprodukten und bei Papillomen. In: Arbeiten aus dem Kaiserlichen Gesundheitsamte. Berlin: Springer; 1905. p. 527–34. 59. Osler W. On syphilis and aneurysm. The Schorstein Lecture. Br Med J. 1909;2(2552):1509–14. 60. Krukenberg E.  Beiträge zur Frage des Aneurysma dissecans. Beitr Path Anat Allg Path. 1920;67:329. 61. Peacock TB.  Report on cases of dissecting aneurysms. Trans Path Soc Lond. 1863;14:87.

O. H. Stanger 62. Peacock TB. An account of some experiments illustrative of the mode of formation of the dissecting aneurysm. Monthly J Med Sci. 1843;3:871–8. 63. Peacock TB. Intimal rupture in dissection. Edinburgh Med Surg J. 1843;60:279. 64. Bostroem E.  Das geheilte Aneurysma dissecans. Deut Arch Klin Med. 1887;42:1. 65. Schede F. Beiträge zu Ätiologie, Verlauf und Heilung des Aneurysma dissecans der Aorta. Virchows Arch Pathol. 1908;192:52. 66. Whitman RG, Stein HB. A contribution to the pathogenesis of dissecting aneurysms: a case of dissecting Mesaortitis (babes and Mironescu), without dissecting aneurysm. J Med Res. 1924;44:579–92. 67. Babes V, Mironescu T. Über dissezierende Arteriitis und Aneurysma dissecans. Beitr Pathol Anat Allg Pathol. 1910;48:221. 68. Eppinger H.  Bekanntes über die Pathogenesen der Aneurysmen. Langenbecks Arch Chir. 1885;35:17–42. 69. Eppinger H.  Pathogenesis, Histogenesis und Aetiologie der Aneurysmen, einschließlich des Aneurysma equi verminosum: Pathologisch-­ anatomische Studien. Berlin; 1887. 70. Rindfleisch E.  Zur Entstehung und Heilung des Aneurysma dissecans aortae. Virchows Arch Pathol Anat. 1893;131:374–8. 71. Shennan T.  Dissecting aneurysm. In: Medical research council special report, Series No. 193. London: His Majesty’s Stationary Office; 1934. 72. Hirst AE, Johns VJ, Kime SW.  Dissecting aneurysms of the aorta: a review of 505 cases. Medicine. 1958;37:217–79. 73. Marfan AB. Un cas de déformation congénitale des quatre members, plus prononcée aux extrémités, caracterisée par l’ allongement des os avec un certain degré d’ amincissement. Bull Mem Soc Med Hop Paris. 1896;13:220–6. 74. Baer RW, Taussig HB, Oppenheimer EH. Congenital aneurysmal dilatation of the aorta associated with arachnodactyly. Bull Johns Hopk Hosp. 1943;72:309–32. 75. Etter LE, Glover LP.  Arachnodactyly complicated by dislocated lens and death from rupture of dissected aneurysm of the aorta. J Am Med Assoc. 1943;123:88–9. 76. Gsell O.  Wandnekrosen der Aorta als selbständige Erkrankung und ihre Beziehung zur Spontanruptur. Virchows Arch. 1928;270:1. 77. Erdheim J.  Medionecrosis idiopathica cystica. Virchows Arch Path Anat. 1929;273:454–79. 78. Manley G. Histology of the aortic media in dissecting aneurysms. J Clin Pathol. 1964;17:220–4. 79. Hurley JV.  Dissecting aneurysm of the aorta. Histological appearances and an hypothesis of pathogenesis. Aust Ann Med. 1959;8:297–306. 80. Schechter DC.  Flashbacks: electrical treatment of aneurysms. Pacing Clin Electrophysiol. 1979;2:234–45. 81. Siddique K, Alvernia J, Fraser K, Lanzino G. Treatment of aneurysms with wires and electricity: a historical overview. J Neurosurg. 2003;99:1102–7.

1  A Brief History of Aortic Pathology and Surgery 82. Philipps B.  A series of experiments performed for the purpose of showing that arteries may be obliterated without ligature, compression or the knife. London; 1832. 83. Velpeau A. Mémoire sur la piqûre ou l’acupuncture des artères dans le traitement das anévrismes. Gaz Med Paris. 1831;2:1–4. 84. Weir RF, Page ED. Aneurysm of the ascending aorta treated by Macewen’s needling method for inducing white thrombus. NY Med J. 1892;55:510–1. 85. Moore CH, Murchinson C.  On a new method of procuring the consolidation of fibrin in certain incurable aneurisms: with the report of a case in which an aneurism of the ascending aorta was treated by the insertion of wire. Med Chir Trans (London). 1864;47:129–49. 86. Ransohoff J. A case of aortic aneurism treated by the insertion of wire. J Am Med Assoc. 1886;7:481–5. 87. Jacobson JH, McAllister FF. A method for the controlled occlusion of larger blood vessels. Ann Surg. 1957;154:334–43. 88. Cooley DA.  Aaneurysm operations: past, present, and future. Ann Thorac Surg. 1999;67:1959–62. 89. Colt GH, Marshall CJ, Wakeley CPG. Three cases of aneurysm of the aorta treated by wiring. Br J Surg. 1925;13:109–19. 90. Power D, Colt GH.  A case of aneurysm of the abdominal aorta treated by the introduction of silver wire, with a description of instruments invented and constructed by Mr G H Colt to facilitate the introduction of wire into aneurysms. Med Chir Trans. 1903;86:363–76. 91. Power D.  The palliative treatment of aneurysm by “wiring” with Colt’s apparatus. Br J Surg. 1921;9:27–36. 92. Scudamore C. An essay on the blood comprehending the chief circumstances which influence its coagulation; the nature of the buffy coat; with a concise medical view of the state of the blood in disease and an account of the powers of a saturated solution of alum as a styptic remedy in hemorrhages. London; 1824. 93. Duncan J. The treatment of aneurism by electrolysis. Edinburgh; 1867. 94. Ciniselli L.  Sulla elettro-puntura nella cura degli aneurismi: studi ed osservazioni. Tip. Vescovile Feraboli; 1856. 95. Ciniselli L. Sugli aneurismi dell’aorta toracica finora trattati colla elettro-puntura. Annali Universali di Med. 1870;214:292. 96. Duncan J, Fraser TR.  On the treatment of aneurism by electrolysis: with an account of an investigation into the action of galvanism on blood and on albuminous fluids. Med Chir Soc Edinb Med J. 1867;13:101–20. 97. Duncan J. Treatment of aneurism by electrolysis. St. Thomas’s Hosp Rep. 1880;10:107. 98. Duncan J.  Lectures on electrolysis. Lecture I.  Br Med J. 1876;1(803):619–22. 99. Duncan J.  Lectures on electrolysis. Lecture II.  Br Med J. 1876;1:715–7.

41 100. Finney JM.  The wiring of otherwise inoperable aneurisms: with report of cases. Ann Surg. 1912;55:661–81. 101. Hunner GL. Aneurism of the aorta treated by insertion of a permanent wire and galvanism (Moore-­ Corradi-­method) with a report of five cases. Bull Johns Hopkins Hosp. 1900;11:263–79. 102. Ransohoff J.  The extirpation of aneurisms. Ann Surg. 1894;19:78–84. 103. Altman P, Voorhees AB.  Aneurysm of the aorta treated by wiring: case report of a 38-year survival. Ann Surg. 1976;184:738–44. 104. Dr. Simpson showing a case of aortic aneurysm, where galvano-puncture had been performed five times. Br Med J. 1881;1:276. 105. Smith WJ.  Case of thoracic aneurism treated by galvano-puncture. Fatal result. Br Med J. 1877;2:295–313. 106. Blakemore AH.  Progressive constrictive occlusion of the abdominal aorta with wiring and electrothermic coagulation. A one-stage operation for arteriosclerotic aneurysms of the abdominal aorta. Ann Surg. 1951;133:447–62. 107. Blakemore AH.  Progressive constrictive occlusion of the aorta with wiring and electrothermic coagulation for the treatment of arteriosclerotic aneurysms of the abdominal aorta. Ann Surg. 1953;137:760–74. 108. Blakemore AH, King BG.  Electrothermic coagulation of aortic aneurysms. J Am Med Assoc. 1938;111:1821–7. 109. Holmes T. The surgical treatment of aneurism in it various forms. Lancet. 1873;2:583–6. 110. Matas R.  The treatment of abdominal aneurism by a preliminary exploratory celiotomy and peritoneal exclusion of the sac, followed at a later sitting by wiring and electrolysis, with the report of two hitherto unpublished cases. Trans South Surg Gynec Assoc. 1901;13:272–330. 111. Matas R.  Traumatic aneurysm of the left brachial artery. Incision and partial excision of the sac-­ recovery. Med New N Y. 1888;53:462–6. 112. Creech O. Endo-aneurysmorrhaphy and treatment of aortic aneurysm. Ann Surg. 1966;164:935–46. 113. Matas R.  An operation for the radical cure of aneurysm based upon arteriorrhaphy. Ann Surg. 1903;37:161–96. 114. Matas R. Surgery of the vascular system. AMA Arch Surg. 1956;72:1–19. 115. Matas R. Ligation of the abdominal aorta: report of the ultimate result, one year, five months and nine days after ligation of the abdominal aorta for aneurism at the bifurcation. Ann Surg. 1925;81:457–64. 116. Matas R.  Aneurysm of the abdominal aorta at its bifurcation into the common iliac arteries: a pictorial supplement illustrating the history of Corinne D., previously reported as the first recorded instance of cure of an aneurysm of the abdominal aorta by ligation. Ann Surg. 1940;112:909–22. 117. Matas R. Personal experiences in vascular surgery: a statistical synopsis. Ann Surg. 1940;112:802–39.

42 118. Leriche R. Des obliterations arterielles hautes comme causes des insuffisances circulatoires des membres inferieurs. Bull Mem Soc Chir. 1923;49:1404–6. 119. Leriche R, Morel A.  The syndrome of thrombotic obliteration of the aortic bifurcation. Ann Surg. 1948;127:193–206. 120. Goyanes J. Nuevos trabajos de cirugia vascular, substitucion plastic de las arteries por las venas o arterioplastia venosa. El Siglo Med. 1906;53:546–61. 121. Bernheim BM.  The ideal operation for aneurysm of the extremity. Report of one case. Bull Johns Hopkins Hosp. 1916;27:93. 122. Lexer E.  Die ideale Operation des arteriellen und des arteriell-venösen Aneurysma. Arch Klin Chir. 1907;83:459. 123. Brooks B.  Intra-arterial injection of sodium iodide. Preliminary report. J Am Med Assoc. 1924;82:1016–9. 124. Goodwin WE, Scardino PL, Scott WW. Translumbar aortic puncture and retrograde catheterization of the aorta in aortography and renal arteriography. Ann Surg. 1950;132:944–58. 125. Gurin D, Bulmer JW, Derby R. Dissecting aneurysm of the aorta: diagnosis and operative relief of acute arterial obstruction due to this cause. N Y J Med. 1935;35:1200–2. 126. Shaw RS. Acute dissecting aortic aneurysms: treatment by fenestration of the internal wall of the aneurysm. N Engl J Med. 1955;253:331–3. 127. Page IH. The production of persistent arterial hypertension by cellophane perinephritis. J Am Med Assoc. 1939;113:2046–8. 128. Page IH.  A method for persistent hypertension by cellophane. Science. 1939;89:273–4. 129. Pearse HE. Experimental studies of the gradual occlusion of large arteries. Ann Surg. 1940;112:923–37. 130. Harrison PW, Chandy J.  A subclavian aneurysm cured by cellophane fibrosis. Ann Surg. 1943;118:478–81. 131. Abbott OA.  Clinical experiences with the application of polythene cellophane upon aneurysms of the thoracic vessels. J Thorac Surg. 1949;18:435–61. 132. Paullin JE, James DF. Dissecting aneurysms of the aorta. Postgrad Med. 1948;4:291–9. 133. Warren WD, Beckwith J, Muller WH.  Problems in the surgical management of acute dissecting aneurysms of the aorta. Ann Surg. 1956;144:530–48. 134. Kirschner M, Schubert A.  Lehrbuch der Chirurgie. Allgemeine u. spezielle chirurgische Operationslehre. Berlin; 1927. 135. Poppe JK. Cellophane treatment of syphilitic aneurysms with report of results in six cases. Am Heart J. 1948;36:252–6. 136. Poppe JK.  Treatment of aortic aneurysms by wrapping with foreign body. Dis Chest. 1949;15: 726–33. 137. Poppe JK.  Reinforcement of aortic aneurysm by wrapping. J Thorac Surg. 1954;27:36–43. 138. DeBakey ME, Creech O, Cooley DA, Halpert B. Failure of polyethylene wrapping in treatment of aortic aneurysms. AMA Arch Surg. 1955;70:65–78. 139. Smith RB.  Presidential address: the foundations of modern aortic surgery. J Vasc Surg. 1998;27:7–15.

O. H. Stanger 140. Crafoord C, Nylin G.  Congenital coarctation of the aorta and its surgical treatment. J Thorac Surg. 1945;14:347–61. 141. Gross RE. Surgical correction for coarctation of the aorta. Surgery. 1945;18:673–8. 142. Shumacker HB.  Coarctation and aneurysm of the aorta; report of a case treated by excision and end-­ to-­end suture of aorta. Ann Surg. 1948;127:655–65. 143. Gross RE, Hurwitt ES, Bill AH, Peirce EC.  Preliminary observations on the use of human arterial grafts in the treatment of certain cardiovascular defects. N Engl J Med. 1948;239:578–9. 144. Gross RE.  Treatment of certain aortic coarctations by homologous grafts: a report of nineteen cases. Ann Surg. 1951;134:753–68. 145. Westaby S.  Landmarks in cardiac surgery. Oxford: Oxford University Press; 1997. p. 234. 146. Cooley DA, DeBakey ME.  Surgical considerations of intrathoracic aneurysms of the aorta and great vessels. Ann Surg. 1952;135:660–80. 147. Bahnson HT. Considerations in the excision of aortic aneurysms. Ann Surg. 1953;138:377–86. 148. DeBakey ME, Cooley DA, Creech O. Surgical considerations of dissecting aneurysms of the aorta. Ann Surg. 1955;142:586–610. 149. Hufnagel CA, Conrad PW.  Dissecting aneurysm of the ascending aorta: direct approach to repair. Surgery. 1962;51:84–9. 150. Morris GC, Henly WS, DeBakey ME.  Correction of acute dissecting aneurysm of aorta with valvular insufficiency. J Am Med Assoc. 1963;184:185–6. 151. DeBakey ME, Beall AC, Cooley DA, Crawford ES, Morris GC Jr, Garrett HE, Howell JF.  Dissecting aortic aneurysms of the aorta. Surg Clin North Am. 1966;46:1045–55. 152. Carrel A, Guthrie CC.  Uniterminal and biterminal venous transplantations. Surg Gynecol Obstet. 1906;2:266–86. 153. Carrel A.  Results of the transplantation of blood vessels, organs and limbs. J Am Med Assoc. 1908;51:1662–7. 154. Peirce EC, Gross RE, Bill AH Jr, Merrill K. Tissue-­ culture evaluation of the viability of blood vessels stored by refrigeration. Ann Surg. 1949;129:333–48. 155. Swan H, Maaske C, Johnson M, Grover R. Arterial homografts. II Resection thoracic aorta aneurysm using a stored human arterial transplant. Arch Surg. 1950;61:732–7. 156. Brock RC, Graham JP.  Resection of a combined aneurysm and coarctation of the aorta with insertion of a homograft. Guys Hosp Rep. 1952;101:207–15. 157. Schafer PW, Hardin CA.  The use of temporary polythene shunts to permit occlusion, resection, and frozen homologus graft replacement of vital vessel segments; a laboratory and clinical study. Surgery. 1952;31:186–99. 158. Estes JE Jr. Abdominal aortic aneurysm: a study of 102 cases. Circulation. 1950;2:258–64. 159. Oudot J.  La greffe vasculaire dans les thromboses du Carrefour aortique [vascular grafting in thromboses of the aortic bifurcation]. Presse Med. 1951;59:234–6.

1  A Brief History of Aortic Pathology and Surgery 160. Friedman SG.  The 50th anniversary of abdominal aortic reconstruction. J Vasc Surg. 2001;33:895–8. 161. Freeman ME, Leeds FH. Vein inlay graft in the treatment of aneurysms and thrombosis of the abdominal aorta; a preliminary communication with report of 3 cases. Angiology. 1951;2:579–87. 162. Connolly JE. Norman freeman: the “first” American specialist in vascular surgery. J Vasc Surg. 1995;22:188–94. 163. Freeman NE, Leeds FH. Resection of aneurysms of the abdominal aorta with anastomosis of the splenic to the left iliac artery. Surgery. 1953;34:1021–31. 164. Dubost C, Allary M, Oeconomos N.  Resection of an aneurysm of the abdominal aorta: reestablishment of the continuity by a preserved human arterial graft, with result after five months. AMA Arch Surg. 1952;64:405–8. 165. Dubost C, Allary M, Oeconomos N.  Aneurysm of the abdominal aorta treated by resection and graft. Arch Mal Coeur Vaiss. 1951;44:848–51. 166. Dubost C.  First successful resection of an aneurysm of the abdominal aorta with restoration of the continuity by a human arterial graft. World J Surg. 1982;6:256–7. 167. Julian OC, Grove WJ, Dye WS, Olwin J, Sadove MS.  Direct surgery of arteriosclerosis. Ann Surg. 1953;138:387–403. 168. Brock RC. Discussion on reconstructive arterial surgery. Proc R Soc Med. 1953;46:115–30. 169. DeBakey ME, Cooley DA.  Surgical treatment of aneurysm of abdominal aorta by resection and restoration of continuity with homograft. Surg Gynecol Obstet. 1953;97:257–66. 170. Bahnson HT.  Treatment of abdominal aortic aneurysms by excision and replacement with homograft. Circulation. 1954;9:494–503. 171. Gerbode F.  Ruptured aortic aneurysm–a surgical emergency. Surg Gynecol Obstet. 1954;98:579. 172. Cooley DA, DeBakey ME. Ruptured aneurysms of abdominal aorta: excision and homograft replacement. Postgrad Med. 1954;16:334–42. 173. Javid H, Dye WS, Grove WJ, Julian OC. Resection of ruptured aneurysm of the abdominal aorta. Ann Surg. 1955;142:613–23. 174. Lam CR, Aram HH.  Resection of the descending aorta for aneurysm; a report of the use of a homograft in a case and an experimental study. Ann Surg. 1951;134:743–52. 175. Cooley DA, DeBakey ME.  Successful resection of aneurysm of thoracic aorta and replacement by graft. J Am Med Assoc. 1953;152:673–6. 176. Shumacker HG, Harris J.  Cure of dissecting aneurysm of the thoracic aorta by total excision. Arch Surg. 1956;73:991–4. 177. DeBakey ME, Cooley DA, Creech O Jr. Surgical considerations of dissecting aneurysm of the aorta. Ann Surg. 1955;142:586–612. 178. DeBakey ME, McCollum CH, Crawford ES, Morris GC Jr, Howell J, Noon GP, Lawrie G. Dissection and dissecting aneurysms of the aorta: twenty-year follow-­up of five hundred twenty-seven patients treated surgically. Surgery. 1982;92:1118–34.

43 179. Cooley DA, DeBakey ME.  Resection of entire ascending aorta in fusiform aneurysm using cardiac bypass. J Am Med Assoc. 1956;162:1158–9. 180. Cooley DA, Mahaffey DE, DeBakey ME. Total excision of the aortic arch for aneurysm. Surg Gynecol Obstet. 1955;101:667–72. 181. DeBakey ME, Crawford ES, Cooley DA, Morris GC.  Successful resection of fusiform aneurysm of aortic arch with replacement by homograft. Surg Gynecol Obstet. 1957;105:657–64. 182. DeBakey ME, Creech O, Cooley DA, Halpert B.  Structural changes in human aortic homografts. Arch Surg. 1954;69:472–82. 183. Sengstaken RW, Blakemore AH. Balloon tamponade for the control of hemorrhage from esophageal varices. Ann Surg. 1950;131:781–9. 184. Levin SM.  Breakthrough: Arthur Blakemore and Arthur Voorhees Jr. J Vasc Surg. 2012;55:1829–31. 185. Voorhees AB.  The origin of the permeable arterial prosthesis: a personal reminiscence. Surg Rounds. 1988;2:79–84. 186. Doerfler J. Über Arteriennaht. Beitr Klin Chir. 1899;26:781–825. 187. Carson HW.  Modern operative surgery. London: Waverley; 1924. p. 378–80. 188. Guthrie CC. End-results of arterial restitution with devitalized tissue. J Am Med Assoc. 1919;73: 186–7. 189. Voorhees AB, Jaretzki A, Blakemore AH. The use of tubes constructed from Vinyon “N” cloth in bridging arterial defects. Ann Surg. 1952;135:332–6. 190. Blakemore AH, Voorhees AB.  The use of tubes constructed from Vinyon “N” cloth in bridging arterial defects  – experimental and clinical. Ann Surg. 1954;140:324–34. 191. Shumway NE, Gliedman NL, Lewis FJ. An experimental study of the use of polyvinyl alcohol for aortic grafts. Surg Gynecol Obstet. 1955;100:703–6. 192. Shumacker HB, Harris EJ, Siderys H.  Pliable plastic tubes as aortic substitutes. Surgery. 1955;37:80–93. 193. Self MM, Cooley DA, DeBakey ME, Creech O. The use of braided nylon tubes for aortic replacement. Ann Surg. 1955;142:836–43. 194. Voorhees AB.  How it all began. In: Sawyer PN, Kaplitt MJ, editors. Vascular grafts. New  York: Appleton-Century-Crofts; 1978. p. 3. 195. Edwards WS.  Arterial grafts. Arch Surg. 1978;113:1225–33. 196. Edwards WS, Tapp JS. Peripheral artery replacement with chemically treated nylon tubes. Surg Gynecol Obstet. 1956;102:443–9. 197. DeBakey ME, Jordan GL, Abbott JP, Halpert B, O’Neal RM.  The fate of Dacron vascular grafts. Arch Surg. 1964;89:757–82. 198. Deterling RA, Bhonslay SB. An evaluation of synthetic materials and fabrics suitable for blood vessel replacement. Surgery. 1955;38:71–91. 199. DeBakey ME, Cooley DA, Crawford ES, Morris GC Jr. Clinical application of a new flexible knitted Dacron arterial substitute. Am Surg. 1958;24: 862–9.

44 200. DeBakey ME, Crawford ES, Cooley DA, Morris GC Jr. Surgical considerations of occlusive disease of the abdominal aorta and iliac and femoral arteries; analysis of 803 cases. Ann Surg. 1958;148:306–24. 201. Creech O, Deterling RA, Edwards S, et al. Vascular prostheses: report of the Committee for the study of vascular prostheses of the Society for Vascular Surgery. Surgery. 1957;41:62–80. 202. Dale WA. In: Johnson G, DeWeese JA, editors. Band of brothers: creators of modern vascular surgery. Pittsboro; 1996. p. 289. 203. Gibbon JH.  Application of a mechanical heart and lung apparatus to cardiac surgery. Minn Med. 1954;37:171–85. 204. Gibbon JH. Maintenance of cardiorespiratory functions by extracorporeal circulation. Circulation. 1959;19:646–56. 205. Gibbon JH. Development of the artificial heart and lung extracorporeal blood circuit. J Am Med Assoc. 1968;206:1983–6. 206. Romaine-Davis A.  John Gibbon and his heart-­ lung machine. Philadelphia, PA: University of Pennsylvania Press; 1991. 207. Fou AA, Gibbon JJ. The first 20 years of the heart-­ lung machine. Tex Heart Inst J. 1997;24:1–8. 208. Bahnson HT, Spencer FC.  Excision of aneurysm of the ascending aorta with prosthetic replacement during cardiopulmonary bypass. Ann Surg. 1960;151:879–88. 209. Schmidt PJ.  Transfusion medicine history illustrated. Technology transfer: the DeBakey roller pump. Transfusion. 2007;47:953–4. 210. Cooley DA, Belmonte BA, DeBakey ME, Latson JR.  Temporary extracorporeal circulation in the treatment of cardiac and aortic disease; report of 98 cases. Ann Surg. 1957;145:898–912. 211. Bigelow WG.  Cold hearts. In: The story of hypothermia and the pacemaker in heart surgery. Toronto: McClelland and Stewart; 1984. 212. Lewis FJ, Taufic M. Closure of atrial septal defects with the aid of hypothermia; experimental accomplishments and the report of one successful case. Surgery. 1953;33:52–9. 213. Sealy WC, Brown IW Jr, Young WG Jr. A report on the use of both extracorporeal circulation and hypothermia for open heart surgery. Ann Surg. 1958;147:603–13. 214. Brock R, Ross DN.  Hypothermia. III.  The clinical application of hypothermic techniques. Guys Hosp Rep. 1955;104:99–113. 215. Griepp RB, Stinson EB, Hollingsworth JF, Buehler D. Prosthetic replacement of the aortic arch. J Thorac Cardiovasc Surg. 1975;70:1051–63. 216. Griepp EB, Griepp RB.  Cerebral consequences of hypothermic circulatory arrest in adults. J Card Surg. 1992;7:134–55. 217. Crawford ES, Saleh SA, Schwessler JS.  Treatment of aneurysms of the transverse aortic arch. J Thorac Cardiovasc Surg. 1979;78:383–93. 218. Crawford ES, Saleh SA.  Transverse aortic arch aneurysm: improved results of treatment employ-

O. H. Stanger ing new modifications of aortic reconstruction and hypothermic cerebral circulatory arrest. Ann Surg. 1981;194:180–8. 219. Ott DA, Frazier OH, Cooley DA.  Resection of the aortic arch using deep hypothermia and temporary circulatory arrest. Circulation. 1978;58:S277–31. 220. Cooley DA, Ott DA, Frazier WWE.  Surgical treatment of aneurysms of the transverse aortic arch: experience with 25 patients using hypothermic techniques. Ann Thorac Surg. 1981;32:260–72. 221. Cooley DA, Livesay JJ. Technique of “open” distal anastomosis for ascending and transverse arch resection. Cardiovasc Dis. 1981;8:421–6. 222. Kouchoukos NT, Wareing TH, Izumoto H, Klausing W, Abboud N.  Elective hypothermic cardiopulmonary bypass and circulatory arrest for spinal cord protection during operations on the thoracoabdominal aorta. J Thorac Cardiovasc Surg. 1990;99:659–64. 223. Rokkas CK, Kouchoukos NT.  Profound hypothermia for spinal cord protection in operations on the descending thoracic and thoracoabdominal aorta. Semin Thorac Cardiovasc Surg. 1998;10:57–60. 224. Frist WH, Baldwin JC, Starnes VA, Stinson EB, Oyer PE, Miller DC, Jamieson SW, Mitchell RS, Shumway NE. A reconsideration of cerebral perfusion in aortic arch replacement. Ann Thorac Surg. 1986;42:273–81. 225. Svensson LG, Crawford ES, Hess KR, Coselli JS, Raskin S, Shenaq SA, Safi HJ.  Deep hypothermia with circulatory arrest. Determinants of stroke and early mortality in 656 patients. J Thorac Cardiovasc Surg. 1993;106:19–31. 226. Stamou SC, Rausch LA, Kouchoukos NT, Lobdell KW, Khabbaz K, Murphy E, Hagberg RC. Comparison between antegrade and retrograde cerebral perfusion or profound hypothermia as brain protection strategies during repair of type A aortic dissection. Ann Cardiothorac Surg. 2016;5:328–35. 227. Daily PO, Trueblood HW, Stinson EB, et  al. Management of acute aortic dissections. Ann Thorac Surg. 1970;10:237–47. 228. DeBakey ME, Henly WS, Cooley DA, Morris GC, Crawford ES, Beall AC.  Surgical management of dissecting aneurysms of the aorta. J Thorac Cardiovasc Surg. 1965;49:130–49. 229. Wheat MW.  Acute dissecting aneurysms of the aorta: diagnosis and treatment. Am Heart J. 1980;99:373–87. 230. Wheat MW, Palmer RF, Bartley TD, Seelman RC.  Treatment of dissecting aneurysms of the aorta without surgery. J Thorac Cardiovasc Surg. 1965;50:364–73. 231. Spencer FC, Blake H. A report of the successful surgical treatment of aortic regurgitation from dissecting aortic aneurysm in a patient with the Marfan symdrome. J Thorac Cardiovasc Surg. 1962;44:238–45. 232. Ross DN.  Homograft replacement of the aortic valve. Lancet. 1962;2:487. 233. Barratt-Boyes BG. Homograft aortic valve replacement in aortic incompetence and stenosis. Thorax. 1964;19:131–50.

1  A Brief History of Aortic Pathology and Surgery 234. Herr RH, Starr A, Pierie WR, Wood JA, Bigelow JC.  Aortic valve replacement. Ann Thorac Surg. 1968;6:199–217. 235. Wheat MW, Wilson JR, Bartley TD.  Successful replacement of the entire ascending aorta and aortic valve. J Am Med Assoc. 1964;188:717–9. 236. Bentall H, De Bono A.  A technique for complete replacement of the ascending aorta. Thorax. 1968;23:338–9. 237. David TE, Feindel CM.  An aortic valve-sparing operation for patients with aortic incompetence and aneurysm of the ascending aorta. J Thorac Cardiovasc Surg. 1992;103:617–21. 238. David TE, Feindel CM, David CM, Manlhiot C. A quarter of a century of experience with aortic valve-­ sparing operations. J Thorac Cardiovasc Surg. 2014;148:872–9. 239. Sarsam MA, Yacoub M.  Remodeling of the aortic valve anulus. J Thorac Cardiovasc Surg. 1993;105:435–8. 240. David TE. Current readings: aortic valve-­sparing operations. Semin Thorac Cardiovasc Surg. 2014;26:231–8. 241. Bloodwell RD, Hallman GL, Cooley DA.  Total replacement of the aortic arch and the “subclavian steal” phenomenon. Ann Thorac Surg. 1968;5:236–45. 242. Borst HG, Walterbusch G, Schaps D. Extensive aortic replacement using the “elephant trunk” prosthesis. Thorac Cardiovasc Surg. 1983;31:37–40. 243. Borst HG. The birth of the elephant trunk technique. J Thorac Cardiovasc Surg. 2013;145:44. 244. Svensson LG. Rationale and technique for replacement of the ascending aorta, arch, and distal aorta using a modified elephant trunk procedure. J Card Surg. 1992;7:301–12. 245. Karck M, Chavan A, Hagl C, Friedrich H, Galanski M, Haverich A. The frozen elephant trunk technique: a new treatment for thoracic aortic aneurysms. J Thorac Cardiovasc Surg. 2003;125:1550–3. 246. Kazui T, Bashar AH. Aortic arch replacement using a trifurcated graft. Ann Thorac Surg. 2006;81:1552. 247. Spielvogel D, Etz CD, Silovitz D, Lansman SL, Griepp RB.  Aortic arch replacement with a trifurcated graft. Ann Thorac Surg. 2007;83:S791–5. 248. Zhou W, Reardon ME, Peden EK, Lin PH, Bush RL, Lumsden AB.  Endovascular repair of a proximal aortic arch aneurysm: a novel approach of supra-­ aortic debranching with antegrade endograft deployment via an anterior thoracotomy approach. J Vasc Surg. 2006;43:1045–8. 249. Benrashid E, Wang H, Keenan JE, Andersen ND, Meza JM, McCann RL, Hughes GC. Evolving practice pattern changes and outcomes in the era of hybrid aortic arch repair. J Vasc Surg. 2016;63:323–31. 250. Lazarus HM.  Intraluminal graft device, system and method. United States Patent number 4,787,899; approved November 29, 1988 (filed December 10, 1986).

45 251. Lazarus HM.  Endovascular grafting for the treatment of abdominal aortic aneurysms. Surg Clin North Am. 1992;72:959–68. 252. Volodos NL, Shekhanin VE, Karpovich IP.  A self-­ fixing synthetic blood vessel endoprosthesis. [in Russian. Vestn Khir Im II Grek. 1986;137:123–5. 253. Volodos NL.  Historical perspective: the first steps in endovascular aortic repair: how it all began. J Endovasc Ther. 2013;20:S3–23. 254. Volodos NL. The 30th anniversary of the first clinical application of endovascular stent-grafting. Eur J Vasc Endovasc Surg. 2015;49:495–7. 255. Svetlikov AV. Unknown pages of the history of vascular stent grafting. J Vasc Surg. 2014;59:865–8. 256. Criado FJ.  EVAR at 20: the unfolding of a revolutionary new technique that changed everything. J Endovasc Ther. 2010;17:789–96. 257. Parodi JC, Palmaz JC, Barone HD.  Transfemoral intraluminal graft implantation for abdominal aortic aneurysms. Ann Vasc Surg. 1991;5:491–9. 258. Parodi J, Criado F, Barone H, Schoenholz C, Queral LA.  Endoluminal aortic aneurysm repair using a balloon-expandable stent-graft device: a progress report. Ann Vasc Surg. 1994;8:523–9. 259. Dake MD, Miller DC, Semba CP, Mitchell RS, Walker PJ, Liddell RP.  Transluminal placement of endovascular stent-grafts for the treatment of descending thoracic aortic aneurysms. N Engl J Med. 1994;331:1729–34. 260. Dake MD, Miller DC, Mitchell RS, Semba CP, Moore KA, Sakai T. The “first generation” of endovascular stent-grafts for patients with aneurysms of the descending thoracic aorta. J Thorac Cardiovasc Surg. 1998;116:689–703. 261. Dake MD, Kato N, Mitchell RS, Semba CP, Razavi MK, Shimono T, Hirano T, Takeda K, Yada I, Miller C. Endovascular stent-graft placement for the treatment of acute aortic dissection. N Engl J Med. 1999;340:1546–52. 262. Ferreira M, Lanziotti L, Monteiro M.  Branched devices for thoracoabdominal aneurysm repair: early experience. J Vasc Surg. 2008;48:S30–6. 263. Nienaber CA, Fattori R, Lund G, Dieckmann C, Wolf W, von Kodolitsch Y, et al. Nonsurgical reconstruction of thoracic aortic dissection by stent-graft placement. N Engl J Med. 1999;340:1539–45. 264. Evangelista A, Mukherjee D, Mehta RH, O’Gara PT, Fattori R, Cooper JV, et  al. Acute intramural hematoma of the aorta: a mystery in evolution. Circulation. 2005;111:1063–70. 265. Carlyle T.  On history. In: Vanden Bossche CR, editor. Historical essays. Berkeley: University of California Press; 1830. p. 2002.

Part I Anatomy

2

The Surgical Anatomy of the Aortic Root Robert H. Anderson, Diane E. Spicer, and Shumpei Mori

2.1

Introduction

Multiple accounts have already been provided of the necessary morphological evidence, not least by ourselves [1–4]. It remains a fact, nonetheless, that a recent questionnaire analysed by the consortium of German surgeons involved in treating diseases of the aortic valve revealed the presence of a “Tower of Babel” regarding the words used in description of the aortic root [5]. In part, this reflects the usage of words in fashions that would not be anticipated from their dictionary definitions. It also reflects the ongoing penchant of anatomists and cardiologists to describe the heart in “Valentine” orientation, rather than as seen within the body during life [6, 7]. In this chapter, we seek to remedy these deficiencies, providing an account of the aortic root as observed in attitudinally appropriate orientation. We use images of the salient anatomic features as seen during life, comparing them to those obtained in the anatomic laboratory, so as to focus on the features of surgical significance. R. H. Anderson (*) Institute of Genetic Medicine, Newcastle University, Newcastle-upon-Tyne, UK e-mail: [email protected] D. E. Spicer Division of Pediatric Cardiology, University of Florida, Gainesville, FL, USA S. Mori Department of Internal Medicine, Division of Cardiovascular Medicine, Kobe University Graduate School of Medicine, Kobe, Japan

2.2

What Is the Aortic Root?

The aortic root is the central part of the left ventricular outflow tract. During its development, the outflow tract is initially an exclusively muscular tube with a solitary lumen, which extends from the outflow of the developing right ventricle to the margins of the pericardial cavity. As it is divided into its aortic and pulmonary components, the walls of its distal part acquire an arterial phenotype as they are converted into the intrapericardial arterial trunks. Shortly after arterialisation of the distal third of the developing outflow tract, half of the proximal part is transferred to become supported by the left, rather than the right, ventricle. It is the transfer of the posterior proximal component to the left ventricle that permits closure of the persisting embryonic interventricular communication. Even after this transfer, however, the intermediate component, now recognisable as the aortic root, remains mostly enclosed within myocardial walls (Fig. 2.1). It is the subsequent regression of the distal myocardial boundary, concomitant with formation of the valvar sinuses, that permits the interleaflet triangles to separate the left ventricular cavity from extracardiac space, as we will describe below. As can be seen from Fig.  2.1, nonetheless, the leaflets of the aortic valve are, from the outset, located within the middle part of the systemic component of the developing outflow tract, with the proximal part forming the subvalvar left ventricular outflow tract and the

© Springer-Verlag GmbH Austria, part of Springer Nature 2019 O. H. Stanger et al. (eds.), Surgical Management of Aortic Pathology, https://doi.org/10.1007/978-3-7091-4874-7_2

49

R. H. Anderson et al.

50

Intrapericardial aorta

Aortic root Distal

Subvalvar outflow tract

Intermediate

Proximal

Developing mitral valve

Fig. 2.1  The image is prepared from an episcopic dataset from a mouse embryo sacrificed at embryonic day 13.5. The dorsal part of the divided outflow tract has been transferred to the left ventricle, and the embryonic interventricular communication has been closed. It can now be seen that the distal part of the outflow tract is forming the intrapericardial aorta, the intermediate part is developing into the aortic root and the proximal part has become the subvalvar outflow tract of the left ventricle. Note, however, that as yet the valvar sinuses have still to form, so that the intermediate component still has predominantly myocardial walls, while the muscular inner heart curvature still separates the developing leaflets of the mitral and aortic valves (white star with red borders), so that, at this stage, the developing aortic valve possesses a completely muscular infundibulum. The authors retain the copyright in the original image from which this figure was prepared

distal part the intrapericardial component of the tubular aorta. It is the middle part of the outflow tract, housing the valvar leaflets, which will form the definitive aortic root.

2.3

 hat Are the Boundaries W of the Aortic Root?

During its development, the aortic root is separated by the cushions within the outflow tract from the developing pulmonary root. The cushions, therefore, form an embryonic septum. In postnatal life, in contrast, the root has its own discrete walls, with no septal components intervening between it and the right-sided components

(Fig. 2.2). The root, which is the channel between the subvalvar left ventricular outflow tract and the tubular aorta, then extends from a virtual plane created by joining together the basal margins of the valvar leaflets to a more obvious boundary at the sinutubular junction (Fig.  2.2). The ventricular surfaces of the leaflets, when viewed from the aspect of the left ventricle in their closed position, resemble the surfaces of a molar or premolar tooth (Fig. 2.3). We presume that it is this appearance of the leaflets, when viewed from their ventricular aspect, that stimulated their alternative naming as “cusps”. In our opinion, however, it is the use of “cusp” by some investigators to account for these moving parts of the aortic root that, in no small way, is responsible for the “Tower of Babel” identified by the German surgeons following the response to their questionnaire [8]. This is because, aside from its inappropriate use when compared to its definition in dictionaries, the term “cusp” is also used on many occasions to describe the valvar sinuses. We prefer, therefore, to avoid its use when describing the components of the aortic root.

2.4

How Should We Describe the Components of the Aortic Root?

As we have shown in Fig.  2.2, the aortic root extends from the virtual basal plane to the sinutubular junction. In essence, the root is the area of the outflow tract that encloses the semilunar attachments of its moving components. As is shown in Fig. 2.4a, the rising parts of the semilunar hinges meet together at the sinutubular junction. Using vernacular terminology, such a conjunction would be described as a cusp, the word being defined as “a pointed end where two curves meet” [9]. The word is currently used interchangeably, however, to describe either the moving parts of the root or the walls that support them. So as to avoid this potential confusion, our preference is to describe the moving components as the valvar leaflets, using the term to describe the moving parts of both the arterial and the ­atrioventricular valves. In this respect, we use

2  The Surgical Anatomy of the Aortic Root

51

Right coronary aortic sinus

Right coronary aortic sinus

Intrapericardial aorta

Intrapericardial aorta

Infundibulum Infundibulum

Mitral valve Mitral valve

a

Non-adjacent aortic sinus

b

Non-adjacent aortic sinus

Fig. 2.2  The left image (panel a) shows a normal heart sectioned to replicate the parasternal long-axis echocardiographic cut. It is compared, in the right-hand panel (b), with a virtual dissection made from a computed tomographic dataset prepared from a patient undergoing investigation for coronary arterial disease. The images show how the aortic root extends from the sinutubular junction distally (red dotted line) to a proximal virtual plane (double-­headed solid red arrow) created by joining together the proximal attachments of the valvar leaflets

(white stars with red borders in panel a). Note that these planes are not necessarily parallel. The virtual dissection, seen in panel b, also shows the level of the anatomic ventriculo-­arterial junction (double-headed dotted red arrow). In panel a, the white dotted line shows the extracardiac area which interposes between the infundibulum of the right ventricle and the crest of the muscular ventricular septum. The authors retain the copyright in the original images from which these figures were prepared

“valvar” as the adjective, rather than “valvular”, on the basis that “valvular” would surely imply the presence of a little valve or valvule. We describe the walls enclosing the leaflets as the valvar sinuses. In Fig.  2.4b, we then show the view of the closed valve as seen from its arterial aspect. The thickened areas where the distal parts of the leaflets come together at the sinutubular junction are universally described as “commissures”. We use “sinutubular” in this instance, as opposed to the more popular “sinotubular”, since logically a “sinotubular” junction would be one with China, whereas the junction is obviously between the valvar sinuses and the tubular aorta. When we then turn to the “commissure”, when defined anatomically the term describes a joint between two bones or, in the case of the eyelids, the zone over which they meet. By analogy, the commissures of the aortic valve, if defined anatomically, should be the zones of apposition of the closed leaflets, which meet together centrally at the valvar centroid. We recognise, nonetheless,

that the peripheral attachments at the sinutubular junction will continue to be described as the valvar commissures, irrespective of the less than perfect use of the term. Because of this, we describe additionally the zones of apposition between the valvar leaflets. It is then important to note that, as the leaflets close along their zones of apposition, the line of closure is no more than half way up the overall height of the root. This feature is emphasised by Schaefers and his colleagues as the effective height of the leaflets, the greatest length of the individual leaflets being taken to represent their geometric height (Fig. 2.5) [10]. Having distinguished between the leaflets and the valvar sinuses, we note that the leaflets are hinged in semilunar fashion so as to enclose the cavities of the sinuses. It is then important to note that, on the ventricular aspect of the sinuses, triangles of the wall extend from the base of the root to the level of the sinutubular junction. These are the interleaflet fibrous triangles [2]. Their pre-

R. H. Anderson et al.

52

Right coronary aortic leaflet

Left coronary aortic leaflet

incorporated at the base of this sinus. We describe the sinus as being nonadjacent, since it is opposite the sinuses that give rise to the coronary arteries, these sinuses being adjacent to the subpulmonary infundibulum (Fig.  2.6b). In most instances, the nonadjacent sinus is also non-­ coronary. On rare occasions, however, it can give rise to a coronary artery [11], meaning that it would then be inappropriate to describe it as being “non-coronary”.

Membranous septum Mitral valve

Non-adjacent aortic leaflet

Fig. 2.3  The image shows the appearance of the closed aortic valve as seen from the apex of the left ventricle. The surfaces of the valvar leaflets (white stars with red borders) protrude into the ventricular cavity, resembling the surface of a molar or premolar tooth. It was, presumably, this appearance that stimulated anatomists to describe the leaflets as “cusps”. As we will explain, this word is poorly suited to the description of the leaflets as viewed in their entirety. Note that only the two leaflets guarding the sinuses that give rise to the coronary arteries are supported by ventricular myocardium. There is fibrous continuity between the nonadjacent valvar leaflet and the aortic leaflet of the mitral valve (red dashed line). The ends of the area of continuity are known as the fibrous trigones (white triangles with red borders).They anchor the aortic mitral unit to the roof of the left ventricle. The right fibrous trigone is also continuous with the membranous septum, forming the so-called central fibrous body. The authors retain the copyright in the original images from which these figures were prepared

cise relationships are of key surgical importance (see below). The semilunar hinges of the leaflets, furthermore, cross the anatomic junction between the arterial walls of the valvar sinuses and their supporting ventricular structures (Fig.  2.6a). Because of this arrangement, crescents of the ventricular tissue are incorporated at the base of each of the valvar sinuses. In the cases of the sinuses that give rise to the coronary arteries, these crescents are myocardial. The semilunar hinge of the nonadjacent valvar leaflet at its nadir, however, crosses the area of fibrous continuity with the aortic leaflet of the mitral valve. This means that an extensive area of fibrous tissue is

2.5

 ow Are the Aortic Sinuses H Located During Life?

We have emphasised that the other problem currently existing in providing accurate descriptions of the components of the aortic root is the time-­ honoured tradition of accounting for the heart as seen on its apex in Valentine position, rather than as it is located in the body during life. This problem can now be circumvented by using, as the gold standard, three-dimensional reconstructions of computed tomographic datasets. Such datasets are now routinely obtained during the investigation of patients with coronary arterial disease. The information contained within the datasets also shows the attitudinally correct orientation of the aortic root (Fig. 2.7). The images show well how the aortic root is positioned centrally both within the overall intrapericardial outflow tract from the left ventricle and also within the cardiac silhouette. When the sinuses themselves are reconstructed (Fig. 2.7b), it can be seen that the right coronary aortic sinus is located anteriorly relative to the other sinuses, which are positioned side by side, with the nonadjacent sinus in right-­ sided location.

2.6

 hat Is the Valvar W “Annulus”?

Perhaps the greatest problem in achieving consensus for the description of the components of the aortic root is in providing a universally acceptable definition of the valvar annulus. The answers provided to the questionnaire circulated

2  The Surgical Anatomy of the Aortic Root

53

Right coronary aortic sinus Non-adjacent aortic sinus

Interleaflet triangle

Membranous septum

a

b

Fig. 2.4  The images are chosen to emphasise the features that currently create problems in achieving a universally acceptable, and logical, nomenclature for description of the components of the aortic root. Panel a shows the root opened through the left coronary aortic valvar sinus. The red dotted lines show the point of conjunction of the two curves produced by the semilunar hinges of the right coronary and non-coronary valvar leaflets. In vernacular terminology, the point at which they join, shown by the white oval, would be described as a “cusp”. At the moment, however, the word “cusp” is used interchangeably to describe either the aortic valvar leaflets or the valvar sinuses. The image also shows the area of valvar continuity (red dashed line), which is thickened at its ends to produce the fibrous trigones

(white triangles with red borders). The right trigone is continuous with the transilluminated membranous septum, the two fibrous components together producing the central fibrous body. Panel b shows the arterial surfaces of the closed valve. The stars show the points of attachment of the adjacent valvar leaflets at the sinutubular junction. If employed in the vernacular usage, these points would be called the valvar “cusps”. In fact, they are universally described as the valvar “commissures”. It is the line of coaptation between adjacent structures, as shown by the dotted lines meeting at the centroid of the valvar orifice (white circle), however, which anatomists define as a “commissure”. The authors retain the copyright in the original images from which these figures were prepared

worldwide by the German surgeons [8] showed that there was division between defining the “annulus” on the basis of the semilunar attachment of the valvar leaflets and according to the virtual basal plane as measured by echocardiographers. There are potential problems with either of these definitions. If we reconstruct the semilunar attachments of the valvar leaflets, which are anatomic entities, then rather than producing a ring-like configuration, the resulting structure is crown-like (Fig. 2.8). It is true that the entirety of the aortic root could be isolated and slipped on the finger in the form of a ring. If asked to describe the reconstructed image shown in Fig.  2.8, however, most viewers would surely describe the entity as a coronet, rather than a ring. It is also the case that the hinges are only seen in

semilunar fashion when the leaflets themselves have been removed (Fig. 2.6a). The remnants of the hinges, of course, are of key surgical importance, since they are used as anchorages for sutures during replacement of the aortic valve. It remains questionable, nonetheless, whether they are best described as the “annulus”. The potential difficulties are highlighted when we know that, when measuring the dimensions of the root, it is the virtual plane forming its basal boundary that is considered to represent the “annulus” by the echocardiographer (Fig. 2.9). As we have already indicated, this definition is not without its own problems. In the first place, as shown in our illustrations, and unlike the hinge lines of the valvar leaflets, there is no anatomic entity representing the echocardiographic “annulus”. The dimen-

R. H. Anderson et al.

54

Sinutubular plane

Effective height

Midsinusal plane

Virtual basal plane

Geometric height

Fig. 2.5  The image is a section across the aortic root, taken through the nadirs of the right coronary and left coronary aortic leaflets and showing the configuration of the leaflets in their closed position. The leaflets close at about halfway within the height of the root as measured from the virtual basal plane to the sinutubular junction. This is more or less the midsinusal level of the root. This height is identified by the Homburg surgeons [11] as representing the effective height of the leaflets. The geometric height, representing the greatest length of the leaflet, is taken from the basal attachment to its tip along the free edge. The authors retain the copyright in the original image from which this figures was prepared

sions of the alleged “annulus”, furthermore, vary depending on whether the measurement is taken from basal hinge point to basal hinge point, which requires an oblique cut across the short axis of the root or from a cut bisecting the root (Fig. 2.9b). The problem is then magnified when it can be seen that the basal circumference of the root is ovoid rather than circular, with differences in its dimensions relating not only to whether cuts are taken from hinge point to hinge point as opposed to bisecting the root but also with regard to its obvious long and short axes (Figs. 2.10 and 2.11). So as to obtain full information regarding the dimensions of the root, it is necessary to measure its diameters at the level of the basal plane, at midsinusal level and at the level of the sinutubular junction, taking note additionally that the basal plane is far from circular in its outline (Fig. 2.10). Regarding definitions, since the echocardiographers will continue to measure the

dimensions of the basal plane, a case can be made for considering this level as representing the “annulus” [12]. As suggested more recently, however, and taking note of the fact that surgeons will be using the remnants of the semilunar hinges as the points of anchorage of their sutures, it is reasonable to distinguish between the surgical and echocardiographic annuluses [13].

2.7

 hat About the Interleaflet W Triangles?

The increasing emphasis on surgical repair of the diseases of aortic valves has focussed attention on the need for precise understanding of the relationships of the interleaflet triangles [14]. As we have already explained, the fibrous triangles occupy the spaces on the ventricular aspects of the semilunar hinges of the leaflets as they extend to the zeniths of their attachments at the sinutubular junction (Fig.  2.6a). As we have also explained, during the development of the valve, the cushions, which excavate to form the valvar leaflets [15], are enclosed within a turret of myocardium. With formation of the valvar sinuses, the myocardial border effectively regresses towards the ventricular base, with myocardium eventually persisting only at the bases of the two aortic sinuses that give rise to the coronary arteries (Fig. 2.6a). There is initially myocardium also at the base of the nonadjacent sinus (Fig. 2.1), but this attenuates with ongoing development to produce the area of aortic-to-mitral valvar continuity (Fig.  2.3). Subsequent to the myocardial regression, the apexes of the triangles between the hinges of the leaflets interpose between the distal extent of the left ventricular cavity and extracardiac space. In the case of the triangle between the left coronary aortic sinus and the nonadjacent sinus, the fibrous tissue separates the left ventricular cavity from the middle portion of the transverse sinus of the pericardium, being situated between the posterior aspect of the aortic root and the anterior component of the interatrial groove (Fig. 2.12). The triangle separating the non-coronary and right aortic coronary sinuses is the most complicated of the three. This is because the base of the

2  The Surgical Anatomy of the Aortic Root

55

Non-adjacent aortic sinus

Left coronary aortic sinus

Pulmonary trunk

Interleaflet triangles

Right coronary aortic sinus

Myocardium

a

Membranous septum

Fibrous tissue

Right fibrous trigone

b

Non-adjacent aortic sinus

Fig. 2.6  In panel a, we show the consequence of the crossing of the semilunar hinges of the aortic valve across the anatomic ventriculo-arterial junction. Because of the crossing, myocardial crescents are incorporated at the bases of the two valvar sinuses that give rise to the coronary arteries, and fibrous tissue is incorporated at the base of the third sinus. The ventricular aspects of the spaces between the hinges are then filled by fibrous triangles. Note that, in this heart, the membranous septum is sepa-

rated from the right fibrous trigone (compare with Fig. 2.4a). Panel b shows the relationship of the aortic root to the pulmonary trunk. By virtue of this relationship, two of the valvar sinuses, which give rise to the coronary arteries, are adjacent to the trunk and its supporting infundibulum; the third sinus is nonadjacent. In most instances it is also non-coronary but very rarely can give rise to a coronary artery. The authors retain the copyright in the original images from which these figures were prepared

triangle is continuous with the membranous septum. In most instances the membranous septum itself is continuous with the right fibrous trigone, these structures making up the so-called central fibrous body (Fig.  2.4a). There is individual variation in the relationship of these structures; nonetheless, since in the heart shown in Fig. 2.6a, the fibrous trigone is separate from the membranous septum. The membranous septum itself, however, continues to form the base of the fibrous triangle that ascends to the sinutubular junction. The membranous septum is additionally crossed on its right side by the hinge of the septal leaflet of the tricuspid valve. This divides the septum into its atrioventricular and interventricular components. The atrioventricular conduction axis penetrates through the atrioventricular component of the septum so that, as we will demonstrate, these relationships are also key to the identification of the location of the conduction tissues. It is the ventriculo-infundib-

ular fold that interposes between the membranous septum and the interleaflet triangle, leaving the triangle to separate the distal extent of the outflow tract from the rightward margin of the transverse sinus (Fig. 2.13). The third triangle is the smallest of the three. It is found at the apex of the space between the semilunar leaflets supported by the aortic valvar sinuses that give rise to the coronary arteries. Since these sinuses are themselves adjacent to the pulmonary root, it follows that this triangle separates the left ventricular outflow tract from the extracardiac tissue plane that runs between the aortic root and the free-standing muscular infundibular sleeve of the right ventricle. It is the presence of the infundibular sleeve that lifts the leaflets of the pulmonary valve away from the ventricular base, permitting the pulmonary root to be removed in its entirety and used as an autograft in the Ross procedure (Fig. 2.14) [16].

R. H. Anderson et al.

56

Intrapericardial aorta

Left coronary aortic sinus

Non-adjacent aortic sinus

Left coronary aortic sinus

Subvalvar outflow tract

Right coronary aortic sinus

a

Right coronary aortic sinus

b

Inferior interventricular artery

Fig. 2.7  The image in panel a is reconstructed from a computed tomographic dataset obtained from a patient undergoing assessment of coronary arterial disease. Seen in frontal projection, it shows the central location of the aortic root, both relative to the cardiac silhouette and the components of the outflow tract. The aortic valvar sinus giving rise to the right coronary artery is situated anteriorly. The reconstruction shown in panel b, from a computed tomographic dataset generated from a different patient, shows that the nonadjacent aortic sinus is right-­

sided and the sinus giving rise to the left coronary artery is left-sided. Both these sinuses are posterior to the sinus giving rise to the right coronary artery. Note also that the artery coursing through the inferior interventricular groove is coursing in almost horizontal fashion when viewed in attitudinally appropriate fashion. It is a mistake to describe this artery as being “posterior descending” [9]. The authors retain the copyright in the original images from which these figures were prepared

2.8

of the muscular ventricular septum, although in some instances the bundle runs below the septal crest. In most cases, nonetheless, the fascicles of the left bundle branch cascade down the left side of the septum from the septal crest, with the right bundle branch penetrating back through the ­muscular septum to surface within the right ventricle in relation to the medial papillary muscle of the tricuspid valve. The relationships between the left bundle branch and the aortic valvar leaflets cannot be demonstrated in any better fashion than by reverting to the illustration provided by Tawara when he provided the first description of the axis (Fig. 2.15) [17]. As the left bundle branch takes its origin from the axis on the crest of the ventricular septum, it is most closely related to the nadir of the hinge of the right coronary aortic leaflet. The conduction axis itself continues

 here Is the Atrioventricular W Conduction Axis?

As can be inferred from the real and virtual dissections shown in Fig.  2.13, the atrioventricular conduction axis is intimately related to the aortic root. It originates from the atrioventricular node, which is located at the apex of the triangle of Koch. The atrioventricular node becomes the penetrating atrioventricular bundle or the bundle of His, as it enters the atrioventricular component of the membranous septum, the fibrous tissue of the septum insulating the axis from the atrial myocardium. Having penetrated through the atrioventricular component of the membranous septum, the non-branching component of the axis is sandwiched between the interventricular component of the membranous septum and the crest

2  The Surgical Anatomy of the Aortic Root

Pulmonary root

Aortic root

57

of the interleaflet triangle between the two coronary aortic valvar leaflets. In the developing heart, the tract continues round the root and penetrates through the area of aortic-to-mitral valvar continuity [18]. In the postnatal heart, however, it is rarely possible to trace the tract beyond the triangle between the two coronary aortic leaflets [19].

Tricuspid ring

2.9

a

Membranous septum

Mitral ring

b

c Fig. 2.8  The hinges of the leaflets of the cardiac valves have been reconstructed from a computed tomographic dataset. The outlines of the atrioventricular valves are relatively annular, although the ring of the mitral valve is saddle-shaped. The hinges of the arterial valves, however, reconstruct in the form of three-pointed coronets. Note the location of the membranous septum, shown in green. The authors retain the copyright in the original image from which this figure was prepared

beyond the take-off of the right bundle branch as the so-called dead end tract. This component will be positioned directly inferior to the base

Conclusions

The anatomy of the aortic root is conditioned by the semilunar nature of the attachments of its leaflets. The root itself is the central component of the intrapericardial outflow tract from the left ventricle, being interposed between the subvalvar ventricular outflow tract and the intrapericardial component of the ascending aorta. As such, it possesses a discrete boundary with the distal component, the sinutubular junction, but its proximal boundary is virtual, being formed by the place that is created by joining together the basal attachments of the leaflets. The root is also centrally located within the heart. Its components should now be described in attitudinally appropriate fashion, the more so since all its parts can readily be demonstrated using virtual dissections of computed tomographic datasets. Controversies regarding the descriptions can now be remedied by describing the moving parts of the root as the leaflets and the components in which they are housed as the sinuses, thus avoiding the need to use the term “cusp”. The other contentious term, the “annulus”, is best dealt with by distinguishing between the semilunar hinges of the leaflets, which when reconstructed produce the surgical annulus, and the diameter of the virtual basal plane, which constitutes the echocardiographic annulus. It is attention to the locations and relations of the interleaflet triangles, separating the ventricular aspects of the valvar sinuses, which crystallises the understanding of the surgical anatomy.

R. H. Anderson et al.

58 Right coronary aortic sinus

Right coronary aortic sinus

Commissure Non-adjacent aortic sinus

Infundibulum

Zone of apposition

Aortic-mitral continuity

a

Transverse sinus

Non-adjacent aortic sinus

Fig. 2.9  The images, made with reconstructions and sections from a computed tomographic dataset, illustrate the problems with using the virtual basal plane as the echocardiographic “annulus”. Panel a shows how measurements would be made if taken from basal hinge point to basal hinge point of adjacent leaflets (red double-headed arrow). As shown in this figure, however, such measure-

Interleaflet triangle

b

Transverse sinus

ments cut obliquely across the aortic root. The cut bisecting the root passes from the basal hinge of one leaflet to the zone of apposition to the opposite leaflets, as shown in panel b (yellow double-headed arrow). The authors retain the copyright in the original images from which these figures were prepared

Membranous septum

Muscular ventricular septum

16.7 mm 18.9 mm

a Fig. 2.10  The images, from a computed tomographic dataset, show cross sections across the aortic valvar sinuses (panel a) and the virtual basal plane (panel b) of an aortic root. The cuts are viewed from the apex of the left ventricle. The arrows show the markedly different

Right fibrous trigone

b

Aortic-mitral continuity

dimensions when taken from basal attachment to basal attachment (double-headed red arrow) as opposed to bisecting the root (double-headed yellow arrows). The authors retain the copyright in the original images from which these figures were prepared

2  The Surgical Anatomy of the Aortic Root

a

b

Fig. 2.11  The images, from the same dataset as that used to provide Fig.  2.10 and taken to replicate parasternal echocardiographic cuts, show the differences in measurements obtained when taking the cut from nadir to nadir of adjacent leaflets (panel a) as opposed to bisecting the root

Left coronary sinus

59

(panel b). The double-headed arrows are in positions comparable to the planes shown in Fig. 2.10. The authors retain the copyright in the original images from which these figures were prepared

Apex of interleaflet triangle

Non-adjacent sinus Transverse sinus

Left atrium

Anterior interatrial groove Non-adjacent sinus

a Fig. 2.12  The images show the relations of the fibrous triangle interposing between the left coronary and the non-coronary aortic sinuses. The image shown in panel a is made by removing the apex of the triangle and photographing the aortic root from behind. The triangle separates the left ventricular outflow tract from the middle part

b of the transverse pericardial sinus. Panel b is a reconstruction from a computed tomographic dataset. The open red arrow shows the wall of the triangle interposed between the outflow tract and the transverse sinus. The authors retain the copyright in the original images from which these figures were prepared

R. H. Anderson et al.

60 Apex of interleaflet triangle

Apex of interleaflet triangle

Right coronary sinus Non-adjacent sinus Non-adjacent sinus Right coronary sinus

Triangle of Koch

a

Membranous septum

Ventriculo-infundibular fold

Fig. 2.13  The images show the location of the apex of the interleaflet triangle interposed between the nonadjacent and the right coronary aortic sinuses. Panel a is created by removing the apex of the triangle and is viewed from the right side. The ventriculo-infundibular fold has also been removed to show the interventricular component of the membranous septum. The virtual dissection in panel b, replicating the dissection made using the speci-

Apex of interleaflet triangle

Membranous septum

b

men, is prepared from a computed tomographic dataset. In each image, the star shows the location of the atrioventricular node, at the apex of the triangle of Koch, and the red dotted line shows the course of the conduction axis, with the right bundle branch emerging in the right ventricle in relation to the medial papillary muscle of the tricuspid valve. The authors retain the copyright in the original images from which these figures were prepared

Left coronary sinus

Right coronary sinus

Right coronary sinus Infundibulum

Infundibulum

Myocardium at sinusal base

a

Septal perforating artery

Fig. 2.14  The images show the relations of the interleaflet triangle separating the hinges of the valvar leaflets guarding the right and left coronary aortic sinuses. Panel a is a dissection made by removing first the pulmonary root, leaving behind the basal part of the free-standing infundibular sleeve, and then the apex of the interleaflet triangle. Panel b is a section through a reconstruction made from a computed tomographic dataset. The open red arrow shows how the fibrous triangle separates the distal

b

Septal perforating artery

part of the left ventricular outflow tract from the tissue plane between the aortic root and the subpulmonary infundibulum. The image in panel a shows how the septal perforating artery enters the crest of the muscular ventricular septum through this tissue plane. Note also, in panel b, the myocardium at the base of the right coronary aortic sinus. The authors retain the copyright in the original images from which these figures were prepared

2  The Surgical Anatomy of the Aortic Root Right coronary sinus

Non-adjacent sinus

Right coronary leaflet

Fig. 2.15  The figure is modified from the original prepared by Tawara in 1906 [17]. It shows how the left bundle branch, coloured in red, is closest to the nadir of the hinge of the right coronary aortic leaflet as it branches from the atrioventricular conduction axis and runs down the left ventricular aspect of the muscular ventricular septum. The authors retain the copyright in the original image from which this figure was prepared

References 1. Anderson RH, Devine WA, Ho SY, Smith A, McKay R.  The myth of the aortic annulus: the anatomy of the subaortic outflow tract. Ann Thorac Surg. 1991;52:640–6. 2. Sutton JPIII, Ho SY, Anderson RH.  The forgotten interleaflet triangles: a review of the surgical anatomy of the aortic valve. Ann Thorac Surg. 1995;59:419–27. 3. Anderson RH, Lal M, Ho SY. Anatomy of the aortic root with particular emphasis on options for its surgical enlargement. J Heart Valve Dis. 1996;5:S249–57. 4. Anderson RH.  Clinical anatomy of the aortic root. Heart. 2000;84:670–3. 5. Mori S, Spicer DE, Anderson RH. Revisiting the anatomy of the living heart. Circ J. 2016;80:24–33.

61 6. Cook AC, Anderson RH. Attitudinally correct nomenclature. Heart. 2002;87:503–6. 7. Anderson RH, Loukas M. The importance of attitudinally appropriate description of cardiac anatomy. Clin Anat. 2009;22:47–51. 8. Sievers HH, Hemmer G, Beyersdorf F, Moritz M, Moosdorf R, Lichtenberg A, et al. The everyday used nomenclature of the aortic root components: the tower of babel? Eur J Cardiothorac Surg. 2012;41:478–82. 9. http://www.oxforddictionaries.com/definition/ english/cusp 10. Bierbach BO, Aicher D, Issa OA, Bomberg H, Graber S, Glombitza P, Schafers HJ.  Aortic root and cusp configuration determine aortic valve function. Eur J Cardiothorac Surg. 2010;38:400–6. 11. Garg A, Ogilvie BC, McLeod AA.  Anomalous origin of the left coronary artery from the non-coronary sinus of Valsalva. Heart. 2000;84:136. 12. Frater RWM, Anderson RH.  How can we logically describe the components of the arterial valves? J Heart Valve Dis. 2010;19:438–40. 13. Loukas M, Aly I, Tubbs RS, Anderson RH. The naming game: a discrepancy among the medical community. Clin Anat. 2015;29:285. https://doi.org/10.1002/ ca.22666. 14. Khelil N, Sleilaty G, Palladino M, Fouda F, Escande R, Debauchez M, Di Centa I, Lansac E.  Surgical anatomy of the aortic annulus: landmarks for external annuloplasty in aortic valve repair. Ann Thorac Surg. 2015;99:1220–6. 15. Anderson RH, Mohun TJ, Spicer DE, Bamforth SD, Brown NA, Chaudhry B, Henderson DJ.  Myths and realities relating to development of the arterial valves. J Cardiovasc Dev Dis. 2014;1:177–200. 16. Merrick AF, Yacoub MH, Ho SY, Anderson RH.  Anatomy of the muscular subpulmonary infundibulum with regard to the Ross procedure. Ann Thorac Surg. 2000;69:556–61. 17. Tawara S.  Das Reizleitungssystem des Säugetierherzens. Jena: Gustav Fischer; 1906. 18. Anderson RH, Boyett MR, Dobrzynski H, Moorman AF. The anatomy of the conduction system: implications for the clinical cardiologist. J Cardiovasc Transl Res. 2013;6:187–96. 19. Kurosawa H, Becker AE. Dead-end tract of the conduction axis. Int J Cardiol. 1985;7:13–20.

3

Morphology and Surgical Anatomy of the Aorta and Pulmonary Trunk Siew Yen Ho

3.1

Introduction

In the normally structured heart, the great arteries exiting the ventricles have a distinctive crossover relationship with the right ventricular outflow tract passing anteriorly and cephalad to the left ventricular outflow tract (Fig.  3.1a). Moreover, the left ventricular outflow tract is directed rightward and cephalad. The aorta emerging from the left ventricle is sandwiched between the right ventricular outflow tract anteriorly and the atria posteriorly. Here, the anterior interatrial groove and the atrial septum lie behind the aorta, separated only by the transverse pericardial sinus. Knowing the central location of the aortic valve in the heart is crucial for better understanding not only of cardiac anatomy but also why interventions on the aortic valve can impact upon other chambers of the heart and, importantly, the atrioventricular conduction system. Whereas the pulmonary valve is the most superiorly situated of the cardiac valves, lying behind the third costal cartilages, the aortic valve is adjacent but located to the right and posterior in relation to the pulmonary valve. The two sets of semilunar valves are not at the same level. Instead, the plane of the aortic valve tilts inferiorly at an angle to the pulmonary valve (Fig. 3.1b). The nadirs of the aortic S. Y. Ho (*) Department of Cardiac Morphology, Royal Brompton Hospital, Imperial College London, London, UK e-mail: [email protected]

sinuses lie in a plane at an angle of 30° from the horizontal [1]. Thus, the arterial surface of the closed leaflets of the aortic valve is directed not only upward but also rightward at an angle of at least 45° to the median plane [2]. The aortic and pulmonary valves are similar in construction, both having semilunar leaflets but with important differences in terms of their attachments and relationships to surrounding structures. Neither valve has an annulus that is commonly perceived as a ring. According to Walmsley (1925) [3] writing in Quain’s Anatomy, Henle was the first to introduce the term “arterial root” to replace the term “arterial ring.” Describing the arterial roots, Walmsley [3] stated “at each of the arterial openings there is a short tubular zone formed of fibrous tissue, the proximal and distal borders of which, at its junctions with the ventricular muscle and with the typical arterial wall respectively, are uneven,” alluding to the lack of a ring or circle. The inappropriate use of the term annulus for the aortic valve was also emphasized by McAlpine [1] in 1975. Although the term annulus remains by convention in clinical practice, it is better to appreciate the fact that the “ring” formed by the attachment lines of the semilunar leaflets is in the shape of a coronet, whereas other rings can be drawn by joining the tops of the leaflet attachments at the sinutubular junction or just below the nadirs of the leaflet attachments (the basal ring) or the anatomic junction between ventricular tissues and the arterial walls (the

© Springer-Verlag GmbH Austria, part of Springer Nature 2019 O. H. Stanger et al. (eds.), Surgical Management of Aortic Pathology, https://doi.org/10.1007/978-3-7091-4874-7_3

63

S. Y. Ho

64

Anterior-posterior view a

b

Fig. 3.1 (a) The endocast demonstrates the crossover relationship between the left and right ventricular outflow tracts and the spiral relationship of the ascending aorta (Ao) with the pulmonary trunk (PT). The right pulmonary artery is hidden underneath the aortic arch, while the left pulmonary artery (LPA) passes posterolaterally. (b) Enlarged view of the ventricular outlets shows the imprints made by the semilunar leaflets of the pulmonary root and

those of the aortic root are at different levels. The dotted ovals mark the level of the sinutubular junction in each. The tricuspid valve (TV) orifice is indicated by the broken line. L, R left, right coronary aortic sinuses, LAA left atrial appendage, LPA left pulmonary artery, LV left ventricle, RAA right atrial appendage, RV right ventricle, SCV superior caval vein

a­ natomic ventriculo-arterial junction) (Fig.  3.2). Then it is necessary to be consistent regarding which ring is used when making measurements, e.g., on cross-sectional imaging when sizing for replacement valves. Of the so-called rings, the sinutubular junction is the narrowest at 10–15% smaller than the “basal ring” and approximately 75% of the maximal sinus diameter [4–6]. It is also the best defined “ring.” The precise location of the ring described by the anatomic ventriculo-­ arterial junction cannot be defined on imaging. By contrast, the coronet shape marks the hemodynamic ventriculo-arterial junction. In this chapter the arterial roots and their continuations into the great arteries are reviewed to provide an anatomical background for cardiac surgeons and interventionists. For consistency, all the illustrations are shown as close to anatomical orientation as possible instead of switching

between anatomical and rotated views mimicking surgical orientations.

3.2

The Aortic Root

Although leaflets are the crucial components of a valve, both the aortic and the pulmonary valves are more complex than just having leaflets and a so-called annulus. More useful is the term “aortic root” to refer to the aortic valve from its origin at the left ventricular outlet to its junction with the ascending portion of the aorta. Anatomically, this whole structure is the aortic valve which comprises of the semilunar leaflets with attachments (or hinge lines) to ventricular and aortic walls and to the “anterior” mitral leaflet, the aortic sinuses (of Valsalva), the interleaflet triangles, and the sinutubular junction (Fig. 3.2) [7].

3  Morphology and Surgical Anatomy of the Aorta and Pulmonary Trunk

Sinutubular junction

65

3.2.1 T  he Aortic Valve with Three Leaflets

The aortic valve apparatus sits at the origin of the aorta where it bulges into three aortic sinuses (of Anatomic ventriculoarterial junction Valsalva) (Fig. 3.3b). The sinuses are contained Basal annulus within the pericardial sac. Consequently, rupture of a sinus of Valsalva aneurysm due to separation of the intima from the media can lead to an adjacent cardiac chamber owing to the central location of the aortic valve or directly into the Interleaflet fibrous triangle pericardial space. The region of the aortic root adjoining the right and non-coronary aortic Sinutubular junction sinuses abuts the anterior right atrial wall causing RCA it to bulge into the atrial chamber forming the aortic mound (or torus aorticus) which can give the false impression of being part of the atrial LCA Anatomic ventriculo- septum when viewing the septum from the right arterial junction atrial aspect. The superior border of the sinuses is Mitral the sinutubular junction (also known as the supra-­ valve aortic ridge) (Fig. 3.4). On the outside, the sinutubular junction is where the tubular portion of the aorta joins the sinusal portion. Inside, there is Fig. 3.2  Diagrams depict the configuration of the aortic usually a slightly raised ridge of thickened aortic root and the so-called rings represented as oval-shaped wall. The sinutubular junction is not perfectly cirbroken lines. The upper panel shows the semilunar attachcular. It takes on the contour of the three sinuses, ments (hinge lines) of the leaflets that form a three-­ pronged coronet with interleaflet fibrous triangles (yellow giving it a mildly trefoil or scalloped outline shapes) filling in the gaps between the prongs. The lower (Fig.  3.3b). The semilunar lines of attachment panel shows the assembly of the aortic root with its proxi- (hinge lines) of the valvar leaflets demarcate the mal insertions into the ventricular myocardium on one inferior borders of the sinuses (Fig. 3.2). But, the side and the mitral valve on the other side and its distal border with the ascending aorta at the sinutubular junc- hinge lines cross the anatomic junction between tion. LCA left coronary artery, RCA right coronary artery ventricular tissues and aortic tissues—the anatomic ventriculo-arterial junction—thus incorporating ventricular tissues into the deepest parts of Guarding the left ventricular outflow tract, the the sinuses (Figs. 3.4 and 3.5). aortic root has an intimate relationship with the Naming of the sinuses can be contentious, but, ventricular septum and the mitral valve. The for simplicity, in the normally structured heart, tract is muscular at its border with the septum two of the aortic sinuses give rise to the main and fibrous at its opposite border where it adjoins coronary arteries; hence the sinuses are termed the anterior leaflet of the mitral valve (Fig. 3.3a). right and left coronary sinuses (Fig.  3.3b). The The root leans slightly rightward, over the ven- orifices of the coronary arteries are commonly tricular septum, to overly the right ventricle spa- found close to the level of the sinutubular junctially. In the elderly, the relationship between tion (Figs. 3.4 and 3.5) [8]. The third sinus is conseptal crest and aortic root changes to give a veniently termed the non-coronary aortic sinus. sigmoid-shaped ventricular septum. This normal In anatomic descriptions, however, the sinuses variation should be differentiated from hypertro- are named anterior (for right coronary), left posphic cardiomyopathy. terior (for left coronary), and right posterior (for

66

a b

Fig. 3.3 (a) View of the left ventricular outflow tract shows just over half of the proximal part of the aortic root (dotted arc) is bordered by fibrous tissues which comprise mainly the area of fibrous continuity between the aortic valve and anterior leaflet of the mitral valve that form the mitral-aortic curtain. The remaining part is bordered by ventricular myocardium. (b) View of the arterial aspect of the aortic root at the level of the sinutubular junction highlights its central location in the heart. The coronary aortic sinuses (L and R) “face” the pulmonary root, whereas the non-coronary aortic sinus (N) is nearest the atria, especially the atrial septum. The tissue plane between the aortic root and the pulmonary outflow tract has been dissected into and the pulmonary root pulled forward to reveal the musculature of the subpulmonary infundibulum which is not part of the ventricular septum. Note the first septal perforator artery (blue arrow) arising from the left anterior descending coronary artery (LAD) and the proximity of the left and right main coronary arteries (LCA, RCA) to the infundibulum. CX circumflex artery

S. Y. Ho

non-coronary). In attitudinal orientation, ­however, the sinuses are in anterior, left posterolateral, and right posterolateral positions, respectively [1]. In all hearts, irrespective of location of the aortic valve relative to the pulmonary valve, the sinuses that give origin to the coronary arteries (or solitary artery) are nearly always the aortic sinuses that are adjacent to the pulmonary valve. These are described as the “facing” aortic sinuses (Fig. 3.4). It is exceedingly rare to encounter origin of a coronary artery from the “non-facing” aortic sinus [9, 10]. In a study of 100 formalin-fixed hearts from adult patients with normally functioning aortic valves, Silver and Roberts observed that the luminal area of the aorta at the sinutubular junction increased with age and with heart weight where increased heart weight was attributed to systemic hypertension [4]. Volume-wise, the sinuses are largest when the valve closes, serving as reservoirs during ventricular diastole and allows filling of the coronary arteries. The right sinus is the largest as is its height, with the left sinus being the smallest on both counts [4, 11]. The differences in sinusal height result in the plane of the sinutubular junction not parallel to a plane joining the bases of the sinuses but having a tilt of 11 degrees [12]. If a line is drawn joining the nadirs of the sinuses/ leaflets, the so-called basal annulus immediately below, it will look more like an ellipse rather than a circle. Notably, cross-sectional imaging such as a transesophageal long-axis or parasternal longaxis plane through the aortic valve will not be transecting two nadirs in the same image, and measurements of the aortic valve diameter are made at this level (Figs. 3.4 and 3.5). If two nadirs are transected, the longitudinal plane likely will be tangential to the diameter of the root. Thus, 2D echocardiographic assessment of left ventricular outlet dimension approximates to the basal ring when the hinge points of opposing leaflets are deemed to be at maximal separation. Furthermore, 3D imaging techniques, including RT 3D TEE, have demonstrated a change in the shape of the basal annulus from an elliptical shape in diastole toward a more rounded shape in systole, resulting in an increase in the annulus area. In systole the anteroposterior diameter increases as the mitral-­ aortic curtain is pushed toward the mitral orifice.

3  Morphology and Surgical Anatomy of the Aorta and Pulmonary Trunk

a

b

c

d

Fig. 3.4 (a and b) are two halves of the same heart that has been cut longitudinally simulating a parasternal echocardiographic plane and a transesophageal plane, respectively. They show the spatial relationship between the right ventricular outflow tract (RVOT) and the aortic outflow tract and the location of the pulmonary valve higher than the aortic valve. (c and d) are enlargements of the aortic root. The series of blue arrows indicate the muscle wall of the subpulmonary infundibulum lying

67

immediately in front of the facing aortic sinuses. The coronary orifices (RCA, LCA) are close to the level of the sinutubular junction. The anatomic ventriculo-arterial junction is indicated by the blue broken line. The cut in this heart is through the nadir of the hinge line of the right coronary leaflet but not through the nadir of the non-coronary leaflet. The so-called basal annulus at the hinge points of the cut leaflets approximates to the red broken line

68

Normally, the valve has three leaflets that are nearly, but not perfectly, equal in size (Fig. 3.6) [11]. Each of the three leaflets of the normal aortic valve has a free margin, a fixed margin where it is attached in semilunar fashion to the aortic root, and a body (or belly). The three leaflets with their accompanying aortic sinuses then form three cups (not cusps) when the leaflets meet on valve closure. The maximal height of each leaflet is considerably less than that of its sinus on account of its scoop-shaped free margin (Figs. 3.4 and 3.5). When the valve opens, the sinutubular junction enlarges, straightening the free margin of the leaflets, and the leaflet bodies fall back into their sinuses without the potential of occluding any coronary orifice. The semilunar hinge lines

a b

Fig. 3.5  The left ventricular outflow tract of a normal heart from an infant has been incised longitudinally and displayed flat. (a) The open arrows indicate the thickening of the aortic wall at the sinutubular junction. The highest points of attachments of the leaflets are at the commissures. The anatomic ventriculo-arterial junction is traced by the broken line. The right and left fibrous trigones (diamonds) are at each end of the area of aortic-mitral fibrous continuity (dotted line). The membranous septum is the pale area marked by the asterisk. (b) The three aortic sinuses (R, N, L) are visible after removing the leaflets. The triangles mark the sites of the interleaflet fibrous triangles. The non-coronary sinus and the adjacent half of the left coronary sinus have no muscle in the nadirs

S. Y. Ho

of adjacent leaflets meet at the level of the ­sinutubular junction, forming the commissures (Fig. 3.5). The body of the leaflets is pliable and thin in the young but becomes thicker and stiffer with age. Minor degrees of calcification are common in the leaflets of elderly patients. In degenerative valvar stenosis commonly seen in patients over the age of 65 years, calcific nodules on the aortic surface of the leaflets and along the hinge lines render the leaflets stiffer and less mobile (Fig. 3.6). Normally, each leaflet has a somewhat crimpled surface facing the aorta and a smoother surface facing the ventricle. The leaflet is slightly thicker toward its free margin, and on its ventricular surface is the zone of apposition, known as the lunule, occupying the full width along the free margin and spanning approximately one third of the depth of the leaflet. This is where the leaflet meets the adjacent leaflets during valvar closure. At the midportion of the lunule, the ventricular surface is thickened to form the nodule (of Arantius). In planar view, when the valve is in closed position, the leaflets show triradiating lines of apposition between the lunules of adjacent leaflets reaching to the peripheral commissures and are encircled by the aortic wall (Fig. 3.6). The semilunar attachments (or hinge lines) of the three valvar leaflets put the aortic root partly in the ventricle and partly in the aorta since the attachments cross the anatomic ventriculo-­ arterial junction (Fig.  3.5a). This arrangement leaves three nearly triangular-shaped pieces of arterial wall in between the arcs [13]. These are the interleaflet fibrous triangles that project above the ventricular mass like three prongs of a coronet (Fig. 3.2). Although they are part of the arterial wall, these triangles are considered ventricular hemodynamically when the valve is closed. In potential communication with extracardiac space, they are potential sites of aneurysmal formation [1]. The triangle between the left coronary and non-coronary leaflets is along the area of aortic-mitral fibrous continuity, but its upper part abuts on the transverse pericardial sinus. The triangle between the right and non-­coronary leaflets adjoins the interventricular part of the membranous septum and is a guide for the location of

3  Morphology and Surgical Anatomy of the Aorta and Pulmonary Trunk

a

b

d

69

c

e

Fig. 3.6 (a–c) are a series of aortic valves showing progression in thickening of the normal valvar leaflets from infancy to adulthood. The nodules of Arantius are more obvious in the older valve (c). (d) A valve with severe ste-

nosis due to senile calcification. (e) Dilatation of the sinutubular junction has straightened the free margins of the leaflets making the valve incompetent

the atrioventricular conduction bundle (Fig. 3.7). The triangle between the two coronary leaflets sits immediately behind the muscular subpulmonary infundibulum. The nadirs of the semilunar hinge lines of the leaflets below the anatomic ventriculo-arterial junction also result in the depths of the right ­coronary, and the anterior half of the left coronary sinus contains ventricular myocardium (Figs. 3.4c, d and 3.5b) [7]. The extent of myocardial inclusion varies from heart to heart. When there is persistence of the embryonic left ventriculo-infundibular fold (inner heart curvature), myocardium can also be present in the remaining sinuses, but this is rarely found in the human heart. Usually, the fold disappears com-

pletely resulting in fibrous continuity between aortic and mitral valves that blurs the precise site of the anatomic ventriculo-arterial junction in this area. The area of valvar continuity is thickened at both ends to form the right and left fibrous trigones (Fig.  3.5): the right trigone contributing to the central fibrous body of the heart which is also the site of penetration of the atrioventricular bundle (of His). The conduction bundle emerges on the ventricular septum sandwiched between the membranous septum and the crest of the muscular septum (Fig. 3.8), usually taking a course toward the left side of the crest. The interleaflet triangle situated between the right and non-coronary leaflets adjoins the membranous septum and is a good guide to the location of the a­ trioventricular con-

S. Y. Ho

70 Sup Inf

RVOT R N

Left bundle

Membranous septum

branch AV bundle Mitral valve

Right fibrous trigone

Fig. 3.7  The fibrous tissue sheath (whitish color) surrounding the conduction bundle is clearly visible in this heart specimen that has been sectioned longitudinally in similar fashion to the heart in Fig. 3.4a. The atrioventricular (AV) bundle emerges from the central fibrous body and continues into the left bundle branch that descends superficially on the septal surface. The area of the membranous septum is related to the fibrous triangle between the right (R) and non-coronary (N) leaflets

duction bundle. From here, the bundle bifurcates into the left and right bundle branches. The proximal portion of the left bundle branch encased in its fibrous sheath is often visible in the subendocardium of the outflow tract in heart specimens (Fig. 3.7). The right bundle branch takes an intramyocardial course through the muscular septum to emerge in the subendocardium on the right side of the septum immediately below the attachment of the medial papillary muscle.

3.2.2 T  he Aortic Valve with Two Leaflets The bicuspid aortic valve reportedly has an incidence of 1–2% or 0.9–2.5% in the normal population making it a common variant of a normal valve [14, 15]. Although the majority are not symptomatic and have no clinical signs, they tend to have a higher incidence of sclerosis or calcification on the leaflets leading to aortic stenosis at

a younger age compared to patients with a ­three-leaflet valve. The morphology of the bicuspid aortic valve is variable, but it has the same essential elements that make up a valve with three leaflets, i.e., aortic sinuses, leaflets, sinutubular junction, ventriculo-arterial junction, basal “ring,” and interleaflet fibrous triangles, and it occupies the same central location in the heart [16]. The valve leaflets may be nearly equal in size, or one leaflet may be much larger than the other (Fig. 3.8). The larger leaflet very often has a raphe in the middle. Some bicuspid valves show a cleft in one leaflet suggesting incomplete separation into two leaflets during development, while others are the consequence of acquired fusion occurring late in life. These may be distinguished from true bicuspid valves by tracing the free edge of the pseudocommissure toward the sinutubular junction. In these valves, three interleaflet fibrous triangles and three sinuses may be detected. The triangle underneath the fusion line may be nearly as tall as the other triangles. By contrast, the triangle lying beneath the raphe of a bicuspid valve is reduced in height considerably, contributing to restricting leaflet mobility [16]. A minority have two nearly equal leaflets, two sinuses, and two interleaflet triangles [16]. In practice, however, clear distinction is not always possible. The leaflets of the bicuspid valve are arranged in either anteroposterior or left-right (lateral) orientations. Anteroposterior is more common, occurring in 79% of cases, and both coronary arteries take origin from the anterior sinus that has the raphe [17]. In the left-­right arrangement, a coronary orifice can be found in each sinus, with the raphe always in the right leaflet. The valvar orifice may also be restricted by the total length of the free edge of the leaflets. If the leaflets are redundant along the free edge, the orifice is less likely to be stenotic [14]. Calcification of a bicuspid valve begins first along the raphe and also on the aortic surface of the other leaflet (Fig. 3.9) [17]. In terms of relationship to the atrioventricular conduction bundle, the landmark of the central fibrous body with the membranous septum still applies. Owing to variations in leaflet configuration and the size and locations of interleaflet fibrous triangles, the

3  Morphology and Surgical Anatomy of the Aorta and Pulmonary Trunk

Fig. 3.8  This longitudinally sectioned heart is viewed from the anterior aspect. It shows the aortic root “leaning” over the ventricular septum and the atrioventricular component of the membranous septum. The red shape represents the location of the atrioventricular node and

bundle superimposed on the picture. The conduction bundle passes toward the left side of the ventricular septum. Note the aortic sinus bulging against the wall of the right atrium at the aortic mound

a

b

c

d

Fig. 3.9  Four bicuspid aortic valves. (a) Normal pliable leaflets with raphe in the anterior leaflet. (b) Sclerotic leaflets. (c) Disproportionate leaflets with redundancy of the

71

larger leaflet. Both leaflets have calcific nodules. (d) Severely stenotic and regurgitant valve with calcification along the raphe and both leaflets

S. Y. Ho

72

rightward (septal) margin of the area of fibrous continuity between the aortic and mitral valves that marks the right fibrous trigone may be a more consistent guide to the site of the conduction bundle. Irrespective of the number of leaflets in the aortic root, the aorta ascends from the sinutubular junction slightly forward and toward the right before curving toward the left and posteriorly.

3.3

 he Ascending Aorta T and Aortic Arch

Like the sinuses, the proximal part of the ascending aorta is enclosed within the pericardium. It measures approximately 5  cm in the adult and extends from a level at the lower border of the third costal cartilage and left half of the sternum to the upper border of the second right costal cartilage. Since the aortic root is to the right and posterior relative to the pulmonary root, the anterior aspect of the proximal ascending aorta is behind the tip of the right atrial appendage and the pulmonary outflow tract which passes cephalad and leftward. Higher up, no longer invested in fibrous pericardium, it ascends toward the right atrium and the superior caval vein. The ascending aorta is approximately 3  cm in diameter at its origin and widens slightly at its upper part where it turns to become the aortic arch. The arch curves cephalad and then toward the left and posteriorly so as to pass in front of the trachea before turning downward, on the left side of the fourth vertebra, to become the descending thoracic aorta. Owing to the spiral relationship between aortic and pulmonary outflow tracts, the aortic arch passes over the right side of the pulmonary bifurcation (Fig. 3.1). The arch is crossed on its left aspect by the left phrenic nerve, the left vagus nerve, and its cardiac branches. The arterial ligament (or patent arterial duct in the fetus) connects the underside of the arch to the bifurcation, a little toward the left pulmonary artery. In the concavity of the arch runs the left recurrent laryngeal nerve as it loops around the arch and also the superficial cardiac plexus, the left bronchus, and some lymph glands. There are usually three major arteries that arise from the aortic arch to supply the head,

neck, and arms. In sequence, they are the brachiocephalic, left common carotid, and left subclavian arteries. The brachiocephalic artery (also known as the innominate artery) is the largest, and it divides into the right subclavian and right common carotid arteries. Occasionally, only two arteries are found coming off the arch and described as the bovine pattern. In these cases, the left common carotid artery arises from the brachiocephalic trunk. Quite often, the arch gives rise to four arteries. The additional artery is usually the left vertebral artery that arises in between the origins of the left common carotid and subclavian arteries. The segment of the arch between the origin of the left subclavian artery and the insertion of the arterial duct/ligament is known as the aortic isthmus. If significant narrowing affects this segment, or more segments, then there is tubular or diffuse hypoplasia of the arch. Discrete narrowing commonly described as coarctation usually affects the part of the arch opposite the ductal insertion [18].

3.4

The Thoracic Aorta

This portion of the aorta, continuous with the arch, begins its descent on level with the lower border of the fourth thoracic vertebra, on the left side, and gradually shifts to the midline to lie immediately in front of the vertebral column at the lower border of the 12th vertebra where it terminates and passes through the aortic hiatus in the diaphragm to become the abdominal aorta. It is covered anteriorly by the pleura and root of the left lung and the pericardium. Along its course, the descending thoracic aorta gives rise to arteries to the viscera as well as the intercostal, subcostal, and superior phrenic arteries.

3.5

The Abdominal Aorta

Beginning at the aortic hiatus of the diaphragm, the abdominal aorta ends where it divides into the two common iliac arteries. It runs in front of the vertebral column to the level of the fourth lumbar vertebra. Along the way it gives off a series of

3  Morphology and Surgical Anatomy of the Aorta and Pulmonary Trunk

large branches: the inferior phrenic arteries, the celiac trunk, the superior mesenteric, the suprarenal, the renal, the spermatic or ovarian, the ­inferior mesenteric, the lumbar, and the middle sacral arteries. The diameter of the abdominal aorta diminishes considerably at its distal portion to nearly two thirds that of the thoracic aorta.

73

Like the aortic root, the pulmonary root extends from the bases of the sinuses to the sinutubular junction and is covered by the fibrous pericardium. Unlike the aortic root, the pulmonary sinuses are not prominent, and the sinutubular junction is less discrete. The wall of the pulmonary root, pulmonary trunk, and its branches is thinner than that for the aorta (Fig. 3.9). Different from the left ventricular outflow tract, the pulmonary outflow tract is an entirely muscular extension from the right ventricle. This cone-shaped muscular infundibulum raises the pulmonary root just above the uppermost part of

the ventricular septum (Fig.  3.10). This allows the pulmonary root to be harvested in its entirety, e.g., in the Ross procedure, without traversing into the left ventricle. Since the right ventricular outflow tract is directed cephalad and leftward, the infundibulum raises the pulmonary valve above and to the left of the tricuspid valve orifice. Two sinuses of the pulmonary root “face” the aortic root albeit at a slightly higher level. When intervening in the pulmonary infundibulum, it is important to realize that the posterior aspect of the infundibulum is not septal so any transgression here would exit the heart into the tissue plane immediately in front of the aortic root (Figs.  3.4 and 3.10). Moreover, the proximal courses of the right and the left anterior descending coronary arteries are in the vicinity (Figs. 3.3 and 3.10). The first septal perforating artery that supplies the right bundle branch, and sometimes also the atrioventricular conduction bundle, penetrates the anterior interventricular groove close by (Fig. 3.3) [19]. There is clear distinction between the muscular subpulmonary infundibulum and wall of the

Fig. 3.10  The right ventricular outflow tract has been sliced through to show its free-standing muscular infundibulum, part of which is the ventriculo-infundibular fold (VIF). The latter inserts into the septomarginal trabeculation (SMT) which is a septal structure. The major coro-

nary arteries are in the vicinity of the infundibulum. The red streak represents the right bundle branch which emerges from the ventricular septum at the base of the medial papillary muscle (asterisk) to run down the SMT in subendocardial location

3.6

The Pulmonary Trunk

S. Y. Ho

74

pulmonary trunk demarcating the anatomic ventriculo-arterial junction. The three semilunar leaflets of the pulmonary valve are of similar construction to those of the aortic valve but are thinner. Because the hinge lines cross the ventriculo-arterial junction, they enclose a small segment of ventricular myocardium in the depths of each sinus [20]. As with the aortic valve, there are three interleaflet fibrous triangles (Fig. 3.11).

3.7

The Pulmonary Trunk and Bifurcation

The pulmonary trunk (or main pulmonary artery) is about 5 cm long and 3 cm wide. It passes cephalad and posteriorly to pass initially anterior to the aortic root and then to the left of the ascending aorta. It lies within the pericardial sac where, together with the ascending aorta, it is invested in a tube of visceral pericardium. The fibrous pericardium fades out as the trunk bifurcates into the

a

right and left main pulmonary arteries. Slightly wider and longer than the left main, the right pulmonary artery passes underneath the concavity of the aortic arch, taking a nearly horizontal and slightly downward course to occupy the angle between the ascending aorta and superior caval vein in front and the azygos vein above before reaching the lung hilum. The left main pulmonary artery also runs nearly horizontally but posterolaterally to cross in front of the descending aorta. It is above the left main bronchus until it gives off its first lobar branch, and then it runs downward and behind the bronchus.

3.8

Conclusions

The intricate spatial relationship between the right and left ventricular outflow tracts and between the great arteries themselves needs to be appreciated in three dimensions because of

b

Fig. 3.11  Two hearts with the right ventricular outflow tract incised into and splayed open to show the muscular infundibulum (green broken line) supporting all three leaflets (a), pulmonary sinuses, the semilunar hinge lines, and the interleaflet fibrous triangles. Removal of the leaflets in

(b) reveals the clear demarcation between ventricular myocardium and arterial wall (anatomic ventriculo-­ arterial junction). The semilunar hinge lines crossing this junction enclose three segments of ventricular myocardium (short arrows) on the hemodynamically arterial side

3  Morphology and Surgical Anatomy of the Aorta and Pulmonary Trunk

the close relationships with other cardiac chambers and valves and important structures such as the conduction system and the coronary arteries. The aortic and pulmonary valves are best described as their respective roots so as to encompass all the structural elements that make up the functioning unit of a valve, not just the leaflets. Normal functioning of a semilunar valve is dependent on all its elements working together. Abnormal function often affects more than one element of the unit. Furthermore, the root does not function in isolation. The structures adjoining the unit must also be included so the entire apparatus works in a coordinated fashion for optimal “morphodynamics” [21]. Whether carrying out surgical repairs or replacing the aortic or pulmonary roots by suturing prostheses or using sutureless prostheses or utilizing transcatheter interventions, good understanding of the relationship of the roots to other cardiac structures is fundamental. Furthermore, so long as the term “annulus” remains in common usage, there must be clarity within each surgical or interventional team which “annulus” is used for implanting the prosthesis and which “annulus” is used for sizing.

References 1. McAlpine WA.  Heart and coronary arteries. Berlin: Springer-Verlag; 1975. p. 9–26. 2. Walmsley R.  Anatomy of left ventricular outflow tract. Br Heart J. 1979;41:263–7. 3. Walmsley T.  The heart. In: Sharpey-Schaffer T, Symington J, Bryce TH, editors. Quain’s elements of anatomy. London: Longmans, Green & Co.; 1925. p. 42–53. 4. Sliver MA, Roberts WC. Detailed anatomy of the normally functioning aortic valve in hearts of normal and increased weight. Am J Cardiol. 1985;55:454–61. 5. Kunzelman KS, Grande KJ, David TE, Cochran RP, Verrier ED.  Aortic root and valve relationships. Impact on surgical repair. J Thorac Cardiovasc Surg. 1994;107:162–70.

75

6. Thubrikar M, Nolan SP, Bosher LP, Deck JD.  The cyclic changes and structure of the base of the aortic valve. Am Heart J. 1980;99:217–24. 7. Ho SY. Structure and anatomy of the aortic root. Eur J Echocardiogr. 2009;10:3–10. 8. Muriago M, Sheppard MN, Ho SY, Anderson RH.  Location of the coronary arterial orifices in the normal heart. Clin Anat. 1997;10:297–302. 9. Liberman L, Pass RH, Kaufman S, Hordof AJ, Printz BF, Prakash A. Left coronary artery arising from the non-coronary sinus: a rare congenital coronary anomaly. Pediatr Cardiol. 2005;26:672–4. 10. Garg A, Ogilvie BC, McLeod AA.  Images in cardiology. Anomalous origin of the left coronary artery from the non-coronary sinus of Valsalva. Heart. 2000;84:136. 11. Vollebergh FE, Becker AE.  Minor congenital variations of cusp size in tricuspid aortic valves. Possible link with isolated aortic stenosis. Br Heart J. 1977;39:1006–11. 12. Choo SJ, McRae G, Olomon JP, St George G, Davis W, Burleson-Bowles CL, Pang D, Luo HH, Vavra D, Cheung DT, Oury JH, Duran CM. Aortic root geometry: pattern of differences between leaflets and sinuses of Valsalva. J Heart Valve Dis. 1999;8:407–15. 13. Sutton JP III, Ho SY, Anderson RH.  The forgotten interleaflet triangles: a review of the surgical anatomy of the aortic valve. Ann Thorac Surg. 1995;59:419–27. 14. Edwards JE.  The congenital bicuspid aortic valve. Circulation. 1961;23:485–8. 15. Roberts WC.  The congenitally bicuspid aortic valve. A study of 85 autopsy cases. Am J Cardiol. 1970;26:72–83. 16. Angelini A, Ho SY, Anderson RH, Devine WA, Zuberbuhler JR, Becker AE, Davies MJ.  The morphology of the normal aortic valve as compared with the aortic valve having two leaflets. J Thorac Cardiovasc Surg. 1989;98:362–7. 17. Davies MJ.  Pathology of cardiac valves. London: Butterworths & Co; 1980. p. 1–61. 18. Ho SY, Anderson RH. Coarctation, tubular hypoplasia, and the ductus arteriosus. Histological study of 35 specimens. Br Heart J. 1979;41:268–74. 19. Hosseinpour AR, Anderson RH, Ho SY. The anatomy of the perforating arteries in normal and congenitally malformed hearts. J Thorac Cardiovasc Surg. 2001;121:1046–52. 20. Ho SY. Anatomic insights for catheter ablation of ventricular tachycardia. Heart Rhythm. 2009;6:S77–80. 21. Yacoub MH, Kilner PJ, Birks EJ, Misfeld M. The aortic outflow and root: a tale of dynamism and crosstalk. Ann Thorac Surg. 1999;68:S37–43.

4

The Normal Aorta and Changes with Age Mary N. Sheppard

4.1

Anatomy of the Aorta

The aorta is the largest vessel in the body. It transports oxygenated blood from the left ventricle of the heart to every organ. The aorta arises from the aortic sinuses above the aortic valve and continues up and over the left hilum of the lung as the ascending and arch of the aorta. The aortic arch then descends as the thoracic and abdominal aorta in the midline posteriorly. The aorta starts in the heart with the aortic valve; immediately adjacent is the aortic root, followed by the ascending aorta, the transverse aorta (the aortic arch), the descending aorta and the thoracoabdominal aorta. The aorta ends in the abdomen after bifurcation of the abdominal aorta in the two common iliac arteries. The ventriculo-aortic junction is characterized by three sinuses (sinuses of Valsalva) which support the semilunar attachments of the aortic valve. Each sinus is the area of the aorta above the attachment of the semilunar leaflets, extending up to a ridge encircling the aorta at the commissures known as the supra-aortic ridge or sinotubular ridge (Fig. 4.1).

Two of the sinuses give rise to the coronary arteries. The right coronary artery arises from the right coronary sinus, and the left coronary artery arises from the left coronary sinus. Both ostia can be seen as rounded openings, but sometimes the opening is elliptical and so needs probing to see clearly (Fig. 4.1). In up to 75% of people, a small infundibular branch of the right coronary artery is located close to the main right coronary opening. The coronary arteries can arise at the sinotubular junction, above or below the junction or close to the commissures between the leaflets [1]. Always check from the origin of the coronary arteries as there can be anomalous origin from the wrong sinus or even the wrong vessel such as the pulmonary artery [2]. Both right and left coronary sinuses usually face the pulmonary trunk in front, before the pulmonary trunk goes to the left of the aorta. The sinus portion of the aorta is 1.5 times wider than the proximal tubular aorta, a fact that is appreciated more easily during life with angiography. Note also that any damage to the aortic root or aortic valve may involve the conduction tissue leading to heart block (Fig. 4.1).

4.1.1 Ascending Aorta M. N. Sheppard (*) CRY Department of Cardiovascular Pathology, Cardiovascular Sciences Research Centre, Level 1, Jenner Wing, St. George’s University of London, London, UK e-mail: [email protected]

The ascending aorta is the segment between the sinotubular junction and the largest aortic branch vessel, the innominate (brachiocephalic) artery (Fig.  4.2). This is the only portion of the aorta

© Springer-Verlag GmbH Austria, part of Springer Nature 2019 O. H. Stanger et al. (eds.), Surgical Management of Aortic Pathology, https://doi.org/10.1007/978-3-7091-4874-7_4

77

M. N. Sheppard

78

that does not give any branch vessels. The ascending aorta is the most anterior portion of the aorta. Therefore, the most common symptom from the ascending aorta and the aortic root is chest pain.

4.1.2 Aortic Arch

Fig. 4.1  Figure shows the left ventricular outflow tract with the aortic valve and ascending aorta. Note the three semilunar aortic leaflets above which are the sinotubular junction. Note the ostium of the left coronary artery in the left coronary sinus and the right coronary artery in the right coronary sinus. To the right is the non-coronary sinus. Note that the triangular area between the right coronary sinus and the non-coronary sinus contains the membranous septum through which the conduction tissue enters from the right atrium. Damage in this area can result in heart block. The pale area below shows branches of the left bundle

The aortic arch (transverse aorta) is a short segment where branch vessels to the head and arms originate. It has typically three branches: First, the innominate (brachiocephalic) artery supplies the right arm and right portion of the head and brain with oxygenated blood. Next, the left carotid artery carries blood to the left head and brain. The last branch vessel from the aortic arch is the left subclavian artery supplying the left arm with blood (Fig. 4.2). Just distal to the left subclavian artery, there is often a puckered or depressed area visible on the intima which is the site of the closed ductus arteriosus (Fig.  4.3). This is also the site of coarctation.

4.1.3 Descending Thoracic Aorta The descending thoracic aorta starts with the last branch vessel off the aortic arch and ends at the first branch in the abdominal aorta, the coeliac artery. The intercostal and bronchial arteries arise from this portion of the aorta (Fig.  4.3). The descending thoracic aorta is the most posterior portion of the aorta. Therefore, the most common symptom from the descending thoracic aorta is back pain. The descending thoracic aorta has several branches that supply a portion of the spinal cord. Any damage involving this portion of the aorta has the risk of spinal cord ischemic injury.

4.1.4 Abdominal Aorta

Fig. 4.2  Figure shows the aortic valve with the ascending aorta and arch. Note the brachiocephalic, left carotid and left subclavian arterial branches

The abdominal aorta branches to the intestine and the kidneys and divides into left and right common iliac arteries. The branch vessels of the abdominal aorta include the coeliac artery, the superior mesenteric artery, the left and right renal arteries and the inferior mesenteric artery.

4  The Normal Aorta and Changes with Age

Fig. 4.3  Figure shows the descending thoracic aorta. At the proximal end near the probe is the puckered area indicating the closed ductal tissue. Below this on both sides are the ostia of the intercostal and bronchial arterial branches

4.1.5 Thoracoabdominal Aorta The thoracoabdominal aorta is the segment starting past the last branch of the aortic arch and ends with the abdominal aortic bifurcation into the left and right common iliac arteries. The size of the aorta is directly proportionate to the patient’s height and weight. Its diameter may range from 1.2 to 3 cm. It is typically largest at the aortic root and smallest in the abdominal aorta. An increase in size is usually indicative of weakness of the wall and points towards the risk of rupture.

79

which impart the elastic properties and strength of the aorta. All arterial walls have the same basic triple-layer composition: intima, media and adventitia (Fig. 4.4). The normal aorta has a thin intima lined by endothelium, a prominent media containing parallel elastic lamellae separated by smooth muscle cells, collagen fibres and a mucoid ground substance rich in proteoglycans. Depending on their location in the vascular tree, these layers vary considerably in thickness, composition and biological properties. The media constitutes the largest component of the artery. The media is composed of ­concentrically arranged lamellar units of fenestrated elastic laminae that enclose smooth muscle cells, collagen fibres and large amounts of proteoglycans. The basic structural and functional unit in the aortic media is the lamellar unit, first defined by Wolinsky and Glagov [3]. Each lamellar unit is composed of a vascular smooth muscle cell (VSMC) sandwiched between two layers of elastin fibres, which contain microfibrils and proteoglycans (Figs.  4.5 and 4.6). The extracellular matrix stains positive with Alcian blue (Fig.  4.7). Lamellar units are

intima

adventitia

media

4.2

Histology of the Aorta

The aorta normally possesses a high degree of elasticity, which aids in the propulsion of blood downstream to the systemic vasculature, and a microstructure that supports this function. The aorta is thus an elastic artery of which the main structural components are elastin and collagen fibres, smooth muscle cells and a proteoglycan-­ rich ground substance. It is the connective fibres within this microstructure, elastin and collagen,

Fig. 4.4  Figure shows a histological section of the aorta including adventitia with connective tissue, vasa vasorum and nerve bundles. The media contains parallel elastic and collagen fibres with smooth muscle cells. The intima consists of a lining of endothelium with underlying thin layer containing collagen and elastic fibres. Haematoxylin and eosin stain

80

Fig. 4.5  Figure shows a histological section of the aortic media containing parallel elastic and collagen fibres with smooth muscle cells. The lamellar unit lies between parallel elastic fibres, containing smooth muscle cells and mucopolysaccharide ground substance. Haematoxylin and eosin stain

Fig. 4.6  Figure shows a histological section of the aortic media with parallel elastic fibres stained black. The space in between the parallel fibres is the lamellar unit. Elastin van Gieson stain

M. N. Sheppard

Fig. 4.7  Figure shows Alcian blue staining highlighting pools of mucopolysaccharide ground substance in the aortic media

intercalated by collagen fibres. Each elastic lamella and the adjacent zone containing the smooth muscle cells synthesize the connective tissue matrix within the lamellar unit. The lamellar unit has both tensile strength and elastic recoil properties, allowing the aorta to withstand high pressures and return to its initial diameter during diastole. The lamellar units are approximately 11  μm in thickness. The number and thickness of lamellae vary by age and topographic site in the aorta; at birth, there are about 35 lamellar units, increasing to 50–60  in adult life. In contrast to muscular arteries, the aorta contains no prominent internal elastic lamina, nor does it have a distinctive external elastic lamina. Thus, these innermost and outermost elastic lamellae do not differ substantially from other laminar units of the media. The adventitia is composed of loosely arranged connective tissues, vasa vasorum, including lymphatic vessels, nerve bundles and low numbers of perivascular lymphocytes (Fig. 4.4). In the aorta, these vasa vasorum normally extend into the outer third of the media and produce nonpathologic disruptions of the lamellar units. These small vessels

4  The Normal Aorta and Changes with Age

penetrate into the outer fifth of the media from the adventitia and form a capillary arcade. The inner four-­fifths of the aortic media is avascular. Aortic distensibility is the key to normal aortic function and relates to the lamellar unit in the media. The elastic lamellae are closely associated with thick collagen fibres containing type I, III and V collagen. Between these collagen fibres, numerous complex, circumferentially oriented streaks of elastin protrude from the lamellae. In contrast to what is usually reported in the aortas of experimental animals, the smooth muscle cells preferentially adhered to these ill-defined streaks rather than directly to the solid lamellae. Fibrillin-1- and type VI collagen-containing bundles of microfibrils are also involved in the smooth muscle cell-elastin contact. The smooth muscle cells are invested by basal lamina-like layers connecting them to each other as well as to oxytalan fibres. These layers are abundantly labelled by fibronectin, whereas type IV collagen, a specific basement membrane component, is mainly found in larger, flocculent deposits. The proteoglycans present are collagen-associated dermatan sulphate proteoglycan, cell-associated heparan sulphate proteoglycan and interstitial chondroitin sulphate proteoglycan. The extracellular matrix in the human aorta is extremely complex and therefore differs from most descriptions based on experimental animals [4]. The parallel elastic lamellae are more numerous in the ascending aorta, usually 55 units, while the descending aorta has less at 28  units. The media is not a static structure. The number of lamellae is only 35 at birth; by adult life, it is 55. After middle age, the number of lamellae is difficult to count because each unit has reduplicated finer elastic lamellae alongside the major one with increased cross-linking. The number of smooth muscle cells decreases, the mucoid ground substance becomes more prominent, and the number of collagen fibres increases. These medial changes alter the compliance of the aortic wall and result in dilatation with elongation, a process known as aortic ectasia. In the ascending aorta, the dilatation leads to ‘unfolding’ of the aorta as seen in a chest X-ray. The increase in rigidity of the aorta leads to the widening of the

81

pulse pressure in old age. With advancing age, the intima thickens and develops fine elastic laminae which make a clear distinction of the intima from the media far from easy histologically. The aortic sinus area contains predominantly fibrous tissue without elastic, explaining why the sinuses bulge in systole.

4.3

Ageing Changes in the Aorta

The aorta also possesses a non-uniform structure displaying distinct regions or segments that are more susceptible to certain types of disease than others. It stands to reason that each segment can undergo different types and degrees of remodelling during ageing and disease and structurally responds distinctly to the various loading conditions seen throughout the length of the aorta. For instance, it has been suggested that the aorta ages ‘from the bottom up’, i.e. biomechanical changes are manifested at an earlier age in the infrarenal abdominal aorta (IRAA) than in the descending thoracic aorta (DTA). It would be prudent to believe that this is a direct result of variable age-­ related architectural changes of the aortic wall, which occur earlier in the distal segments of the aorta than in the proximal ones. The histologic changes that occur in the media of the normal aorta at various ages were studied in 100 normal aortas. These changes encompassed (1) cystic medial degeneration defined as pooling of mucoid material; (2) elastin fragmentation, characterized by disruption of elastin lamellae; (3) fibrosis, defined as an increase in collagen at the expense of smooth muscle cells; and (4) loss of smooth muscle nuclei (Fig. 4.8). The changes showed a striking correlation with age and represent the normal ageing process for the aorta. The alterations showed a close relation in onset and location within the media, suggesting a phenomenon of injury and repair caused by haemodynamic events. These findings in the normal ageing aorta reveal that none of the histologic changes observed can be regarded as the specific structural alteration responsible for the development of dissecting aneurysm [5].

82

Fig. 4.8  Figure shows the aortic media with fragmentation of the elastic fibres and increase in background collagen highlighted in red. Elastin van Gieson stain

The thickness of the intima increases gradually with age; in newborns, the intimal layer is very thin, and the endothelial lining is closely apposed to the first elastic lamella of the media. Due to a process of low-grade injury and repair over many years, the intimal layer gradually expands and is composed of extracellular matrix proteins (mainly collagen and mucopolysaccharides), and sparse mesenchymal cells, which are best visualized with an antibody for the alpha-1 isoform of actin (vascular smooth muscle cell actin, SMA-1 antibody). One of the earliest effects of age shows the concentration of elastin within the aorta decreased with age, but the elastin content remained unchanged which also suggested that the decrease in elastin concentration was in part due to increases of other components, such as collagen, while maintaining total elastin content. Age can alter the amino acid structure of elastin, which might be responsible for the ageing-associated changes seen in the human aorta such as loss in elasticity. The remodelling of the human thoracic aorta that occurs with ageing is associated with

M. N. Sheppard

fragmentation of elastin fibres. This fragmentation has also been reported in old, highly calcified thoracic aortas accompanied by atherosclerotic lesions. It should be noted, however, that calcification of the aorta does not always require atherosclerosis and occurs in a process known as medial elastocalcinosis. While the exact mechanism for which these elastin fibres are fragmented with age is not known, it could be a consequence of mechanical fatigue failure caused by the pulsatile wall strain experienced by the aorta over the number of cardiac cycles experienced in a lifetime. It might also be due to chemical degradation caused by the upregulation of matrix metalloproteases (MMPs) with age owing to an imbalance with their inhibitors. Additionally, there is evidence that the increased expression of MMPs could accompany the aforementioned calcification of elastin. An increase in the content of collagen within the aortic wall with increased age has been well characterized. The content of hydroxyproline, which is proportional to the content of collagen, is increased with age. This increase in collagen content did not occur linearly with age for the thoracic aorta, as it is constant for ages younger than 45 and increased slowly thereafter. The structure of the collagen fibres themselves also change with advanced age, showing an increase in irregularly arranged fibres in the media of the human thoracic aorta. The cross-links between collagen fibres have also been found to increase with age because of two different mechanisms. The first is due to a marked increase in the content of cross-linking amino acids histidinoalanine and pentosidine in collagen itself. The second is due to a marked increase in the accumulation of advanced glycation end products with age, which are produced by the glycation and oxidizing reactions between sugars and the amino groups in protein molecules and form bridges between collagen fibres. With age, the aorta stiffens with these associated histopathologic changes and becomes tortuous with regional differences. A variety of risk factors contribute to aortic stiffness and propensity towards aneurysm and dissection with hypertension being the most critical factor. The greatest

4  The Normal Aorta and Changes with Age

difference in aortic stiffness occurs in the abdominal region, whereas the greatest difference in diameter occurs in the ascending aorta (annuloaortic ectasia). Fibrous intimal thickening (i.e. maximal in abdominal aorta), increased proteoglycans and CD68-positive macrophages are noted in ageing aortas. These findings are also related to atherosclerosis.

4.3.1 Elastic Component of Aorta Ageing is accompanied by a relative loss of elastin content in human arteries as other matrix materials, primarily collagen, increase. The remaining elastic fibres are more frequently fragmented and less cross-linked in the aged aorta as compared to younger individuals. Histologically, this appears as a thinner, more separated and more fragmented elastic meshwork in the media (Fig. 4.8). Fragmentation of intralamellar elastic fibres results in the separation of elastic laminas within the aortic media leaving gaps partially filled with proteoglycans. This histopathologic appearance of the media in older individuals is frequently comparable to the aortas resected from younger individuals with a genetic syndrome.

4.3.2 Extracellular Matrix As the vasculature ages, neointimal proliferation and frank atherosclerosis may occur. A large amount of extracellular matrix, particularly collagen types I and III, are produced, and they fill the subendothelial space. In the media, there is an increase in proteoaminoglycans and mucopolysaccharide ground substance between the elastic fibres increasing the space between fibres.

4.3.3 Smooth Muscle Cells Medial smooth muscle cells decrease in number in the ageing aorta. Medial smooth muscle cells also undergo age-related morphological changes towards a senescent phenotype. The loss of smooth muscle cells is often more apparent under

83

neointimal plaques. This might be secondary to reduced diffusion of reagents from the lumen to these cells as the neointimal plaque increases in size. Extension of the vasa vasorum deeper into the media may preserve the integrity of these cells. Medial collagen, which is absent to rare in young aortas, increases with age. This is noted by an expansion of collagen staining within the intralamellar space by a collagen stain such as Masson’s trichrome or similar. This collagen is described as disorganized and unstructured, forming cloudy spaces rather than crimped or wavy fibre bundles.

4.4

Conclusions

In the ageing aorta, the elastic component, medial smooth cells and extracellular matrix are progressively, negatively altered. This results in the regenerative potential declining significantly with increased age promoting aortic aneurysms and dissections. Features of medial degeneration in the ageing aorta bear close resemblance to medial changes in the genetic syndromes. However, it should be noted that, in the case of genetic syndromes, these degenerative changes are more extensive and they occur at an earlier age.

References 1. Muriago M, Sheppard MN, Ho SY, Anderson RH.  Location of the coronary arterial orifices in the normal heart. Clin Anat. 1997;10:297–302. 2. Krexi L, Sheppard MN.  Anomalous origin of the left coronary artery from the pulmonary artery (ALCAPA), a forgotten congenital cause of sudden death in the adult. Cardiovasc Pathol. 2013;22:294–7. 3. Wollinsky H, Glagov S.  A lamellar unit of aortic medial structure and function in mammals. Circ Res. 1967;20:99–111. 4. Dingemans KP, Teeling P, Lagendijk JH, Becker AE. Extracellular matrix of the human aortic media: an ultrastructural histochemical and immunohistochemical study of the adult aortic media. Anat Rec. 2000;258:1–14. 5. Schlatmann TJ, Becker AE. Histologic changes in the normal aging aorta: implications for dissecting aortic aneurysm. Am J Cardiol. 1977;39:13–20.

Part II Mechanisms

5

Molecular Mechanisms of Aortic Valve Pathology Ghada Mkannez, Deborah Argaud, Marie-­Chloé Boulanger, and Patrick Mathieu

5.1

Introduction

The knowledge about molecular processes involved into the development of valve disorders has expanded considerably over the last 15 years or so. Owing to the development of isolation technique suitable for the culture of valve interstitial cells (VICs), the biology of heart valves is now well established and has opened novel vistas to better understand the pathobiology of some common heart valve disorders. Heart valves are under intense mechanical constraint evolving in a dynamic environment where changes in blood flow during systole and diastole are transduced into biological responses allowing the normal

G. Mkannez · D. Argaud · M.-C. Boulanger Department of Surgery, Laboratory of Cardiovascular Pathobiology, Quebec Heart and Lung Institute/ Research Center, Laval University, Québec, QC, Canada e-mail: [email protected]; [email protected]; [email protected] P. Mathieu (*) Department of Surgery, Laboratory of Cardiovascular Pathobiology, Quebec Heart and Lung Institute/ Research Center, Laval University, Québec, QC, Canada Institut de Cardiologie et de Pneumologie de Québec, Quebec Heart and Lung Institute, Québec, QC, Canada e-mail: [email protected]

maintenance of extracellular matrix (ECM) components. The development of heart valve diseases is linked to several factors that have intricate relationships with the biology of VICs and valve endothelial cells (VECs). As such, biomechanical cues along with circulating factors may work together in modifying VIC and VEC biology leading to an abnormal homeostatic response, which ultimately may promote the development of valve disorders [1]. Moreover, studies have started to unravel the genetic architecture of some heart valve diseases and have thus provided a unique glimpse into the complex interplay between the genes and the environment in promoting the development of heart valve disorders [2]. In this chapter, we aimed to provide a comprehensive overview of aortic valve biology and to highlight the key underpinning processes involved in common aortic valve disorders, namely, calcific aortic valve disease (CAVD) and bicuspid aortic valve (BAV).

5.2

Structure and Biology of the Aortic Valve

5.2.1 Structure of the Aortic Valve By opening and closing for about three billion of cycles during a lifetime, the aortic valve exerts a crucial role to ensure the normal function of the

© Springer-Verlag GmbH Austria, part of Springer Nature 2019 O. H. Stanger et al. (eds.), Surgical Management of Aortic Pathology, https://doi.org/10.1007/978-3-7091-4874-7_5

87

88

heart. The normal aortic valve is composed of three leaflets or cusps attached to the aortic annulus. The highest point-of-attachments are the commissures, which reach the sinotubular junction. The crescent and semilunar structure of each leaflet is surrounded by the sinus of Valsalva. Competency of the aortic valve is ensured by a surface of coaptation where the leaflets meet during diastole. The central portion of the free edge of each leaflet corresponds to the nodule of Arantius. The complex architecture of the aortic root, which consists in the aortic leaflets, annulus, sinuses, commissures, and sinotubular junction, ensures the opening of the aortic valve during systole and prevents the regurgitation of blood flow during diastole. The resilience of the aortic valve is determined by its trilaminar structure. The fibrosa and ventricularis are the outermost and innermost layers that face the aorta and ventricle, respectively. The fibrosa is rich in circumferentially oriented collagen type I fibers, whereas the ventricularis is enriched in radially oriented elastin fibers. The middle layer or the spongiosa contains an abundance of glycosaminoglycans that provide a cushion in absorbing the mechanical load during the cardiac cycle [3].

5.2.2 Biology of the Aortic Valve The cellular population of the aortic valve is composed predominantly of VICs, which are of mesenchymal origin and have a high level of plasticity [4]. Smooth muscle cells (SMCs) are far less abundant compared to VICs and are located at the base of leaflets [5]. The aortic valve leaflets are covered by VECs, which have important crosstalks with VICs. VIC is a heterogeneous population which can differentiate in  vitro into osteoblast-like cells, chondrocytes, and adipocytes when grown with appropriate growth medium [6]. The VICs help maintain the integrity of the aortic valve by renewing the ECM. During embryogenesis, endothelial-to-­mesenchymal transition (EndoMT) is the leading process in the cardiac cushion that generates VICs. In rodents, the progressive transition of endothelial cells to

G. Mkannez et al.

VICs is accompanied by a loss of VE-cadherin and the expression of several genes such as Twist1, Msx2, and Sox9 that drives cell proliferation and differentiation process [7]. Of note, several genes involved in the EndoMT and embryogenesis of the aortic valve are reexpressed during the development of aortic valve disorders [8]. Also, in rodents, there is evidence that neural crest-derived VICs participate to valvulogenesis and populate, via the truncus arteriosus, the septal leaflet. During adulthood, studies carried out in mice have shown that VICs are renewed by EndoMT and by the recruitment of circulating hematopoietic stem cells (HSCs) (lin−, c-kit+, Sca-1+, CD34−), which participate to the maintenance of the ECM. Following engraftment of circulating HSCs in the aortic valve, VICs preferentially locate into the ventricularis and are CD45+/CD34− [9]. In mouse aortic valve, VIC population is characterized by cells that are CD45+/CD34− and CD45−/CD34+. The CD45+/ CD34− population is less abundant (~30%) and according to a report is prone to mineralize in vitro [10]. Circulating HSCs may also differentiate into dendritic cells (DCs) CD11c+ [11]. In mice, DCs are predominantly located at the base of leaflets in the subendothelium and have cellular extension toward the lumen. Hence, it is likely that DCs in the aortic valve may cross-present antigens to lymphocytes and thus activate an immune adaptive response. The VECs also provide a unique interface with the blood and the aortic valve. It is worth pointing out that the VECs from the aortic and ventricular sides are exposed to different blood flow conditions. During systole, the blood flow to which the ventricular VECs are exposed is mostly laminar. However, VECs facing the aorta are exposed to an oscillatory blood flow. The expression of genes with anti-mineralizing properties is reduced in porcine aortic VECs compared to ventricular VECs [12]. Moreover, it has been shown that VEC-derived nitric oxide (NO) decreased the tension and stiffness of the ECM [13]. Alternatively, decreased production of NO, which is promoted by an oscillatory flow, increases the tension on VICs and thereby activates these cells with a secretory phenotype [14].

5  Molecular Mechanisms of Aortic Valve Pathology

89

NO is well known to inhibit the transition of VICs into osteogenic cells and to prevent mineralization in vitro. By a process which is not yet fully elucidated, NO increased the activation of the NOTCH pathway in VICs and prevented the expression of key osteogenic genes such as RUNX2 and BMP2 [15]. Hence, oscillatory blood flow may impact on the gene expression pattern of aortic VECs and the release of NO and may explain, at least in part, why mineralization starts in the fibrosa during CAVD.

5.3

Calcific Aortic Valve Disease

5.3.1 Pathology CAVD is a chronic disorder, which is characterized by the fibrocalcific remodeling of the aortic valve [16, 17]. The striking features of surgically explanted aortic valves are the presence of mineralized nodules and the excess of ECM characterized by a fibrotic component (Fig. 5.1). As highlighted in the precedent section, the mineralization process starts in the fibrosa layer. With time, the accrual of mineral provokes a

a

b

d­ istortion of the normal anatomy of leaflets that modifies substantially the biomechanical characteristics of the aortic valve. The fibrocalcific response of the aortic valve is also accompanied by the formation of blood vessels. The aortic valve is normally avascular, and neoangiogenesis is one important feature of CAVD. Endothelial progenitor cells infiltrate the aortic valve and participate to the neovascularization process. Histopathologic analysis of surgically explanted stenotic aortic valves has revealed that the density of neovascularization is closely associated with the inflammatory infiltrate, which consists in macrophages and lymphocytes [18]. Though the role of neovascularization during CAVD is not clearly established, it is possible that it promotes the recruitment of inflammatory cells. Also, investigations have shown that neovascularization in surgically explanted stenotic aortic valves was often accompanied by intra-leaflet hemorrhage [19]. Though the significance of this observation is not yet elucidated, it is possible that it contributes to elevate the local load of iron, which is internalized by VICs. In turn, internalized iron in VICs increases the proliferation of cells [20]. Whether iron can increase the

c

*

d

e

*

Fig. 5.1  Sections of calcific aortic valve disease showing the presence of leukocytes (a and b), macrophages (c), and T cells (d). Inflammatory infiltrates co-localize with

f

mineralization (asterisk) and neoangiogenesis (e and f). Reproduced from Côté et al. [18] with kind permission of Springer

90

osteogenic transition of VICs is presently unknown. Another important feature of CAVD is the presence of lipid infiltrates in explanted stenotic valves. Immunohistological analyses have shown that apolipoprotein E, apolipoprotein B, apolipoprotein(a), and apolipoprotein A1 are present near calcified areas [21]. These observations raised the hypothesis that lipids may play a significant role into the development of CAVD.

5.3.2 Role of Lipids in CAVD Recently, studies with a Mendelian randomization design have shown a causal link between LDL and Lp(a) with CAVD. The single nucleotide polymorphism rs10455872 located in the LPA gene, which encodes for apolipoprotein(a), was associated with CAVD at the genome-wide level [22]. Also, the plasma level of Lp(a) is a predictor for a faster progression of aortic stenosis. Lp(a) is a LDL-like particle in which the apolipoprotein(a) is linked to the apolipoprotein B. Quantitative assessment of lipids in explanted stenotic aortic valves showed that the level of oxidized low-­ density lipoprotein (OxLDL) is associated with the density of the inflammatory infiltrate and the expression of tumor necrosis alpha (TNF-α) [23]. Also, the valvular level of OxLDL is associated with the circulating proportion of small, dense LDL. Small, dense LDL particles have a high rate of oxidation and have a greater capability to infiltrate the aortic valve. Moreover, studies have underlined that reactive oxygen species (ROS) are generated during CAVD.  The increased production of ROS in CAVD is related to the uncoupling of the nitric oxide synthase (NOS) and also to the upregulation of the NADPH oxidase [24, 25]. Hence, it is likely that lipid infiltration and subsequent oxidation is promoted by the high level of ROS produced during CAVD.  In addition, there is evidence that the lipids are modified and retained in the aortic valve. In surgically explanted stenotic aortic valves, the expression of proteoglycans such as biglycan and decorin is increased [26]. Furthermore, experimental evidence sug-

G. Mkannez et al.

gests that transforming growth factor-β1 (TGFβ1), which is overexpressed in CAVD, promotes the elongation of glycosaminoglycan (GAG) chains on proteoglycans and increases the retention lipids. Increased transit time and retention of lipoproteins in aortic valves promote the modification of the lipoproteins by enzymes, which may result in the production of bioactive lipids that activate inflammation and mineralization. Studies of explanted stenotic aortic valves showed that lipoprotein lipase (LPL), which is secreted by macrophages, is overexpressed in mineralized valves and co-localized with decorin [27]. Experimental data suggest that interaction between LPL and decorin may increase the retention of lipids. We recently underlined that lipoprotein-associated phospholipase A2 (Lp-PLA2) and autotaxin (ATX), a lysophospholipase D, were overexpressed in stenotic, mineralized aortic valves [28, 29]. Lp-PLA2 transforms OxLDL into lysophosphatidylcholine (LPC). In turn, ATX converts LPC into lysophosphatidic acid (LPA), a bioactive lipid that incidentally accumulates in surgically explanted stenotic aortic valves. Noteworthy, we found that ATX enzyme activity is enriched in isolated Lp(a) fraction and was transported into the aortic valves. Moreover, in response to TNF-α, VICs secrete ATX, which mediates the mineralization of cell culture induced by LPC.  In LDLR−/− apoB100/100 IGF2 transgenic mice, the administration of LPA accelerated the mineralization of the aortic valve and the development of CAVD [29]. In patients with CAVD, we found that the circulating level of Lp-PLA2 was associated with the progression rate of aortic stenosis and the calcification of aortic valve bioprostheses [30–32]. Recently, we highlighted that a high level of ATX activity combined with an elevated Lp(a) concentration in the blood plasma increased the risk of CAVD by 3.5fold in patients with coronary artery disease [33]. Taken together, these data suggest that during CAVD, lipid-modifying enzymes, such as Lp-PLA2 and ATX, produce bioactive lipids with pro-osteogenic activity and contribute to the development/progression of aortic valve mineralization (Fig. 5.2).

5  Molecular Mechanisms of Aortic Valve Pathology Lipid infiltration

91

Inflammation

Fibro-calcific response

Radiation Mechanical stress Lipid-derived species Cytokines Lipids LDL

Lp(a)

Calcium hydroxyapatite

NOS uncoupling ROS Ox-LDL Ox-PL ATX

Chymase ACE Lp-PLA2

MMPs VEGF

TNF IL-1b

lysoPC lysoPA

sPLA2

BMP2

Osteoprogenitor cell Inflammation

RUNX2 MSX2

ENPP1 NT5E

A2AR

AA

Leukotrienes

LDL

RANKL TNF

VIC

5-LO

VEGF

Angiotensin II

IL-6 WNT3a TGFb

ATX LPAR

Blood vessel

Angiotensin I

Collagen Fibrosis

Mineralization COX2 Prostaglandins

ATP

AMP +PPi ALP Pi

Adenosine +Pi

Apoptosis Osteogenic transition

Monocyte

Macrophage

Mastocyte

Calcifying microvesicles

T cell Time

Nature Reviews Disease Primers

Fig. 5.2  Scheme showing the pathobiology of calcific aortic valve disease. Reproduced from Lindman et al. [17] with kind permission from Nature Publishing Group

5.3.3 Inflammation Chronic low-grade inflammation is suspected to play a significant role in the development and progression of CAVD [34]. The transcriptome of CAVD is typified by a dysregulation of genes involved in inflammation [35]. In 285 surgically explanted stenotic aortic valves, Cote et al. found that the presence of dense inflammatory infiltrates was linked to the remodeling process and osteochondrogenic metaplasia [18]. The valvular density of leukocytes was related to the expression of TNF-α and preoperative progression rate of aortic stenosis. A prominent feature of inflammation in CAVD is related to the expression of several proteases, which may remodel the ECM. In this regard, the expression of metalloproteinase 2 (MMP2) and MMP9 is increased, and their levels are associated with the ­angiogenic

activity. Secreted protein acidic and rich in cysteine (SPARC), which is overexpressed in CAVD, is cleaved by MMPs into small peptide fragments with angiogenic activity [36]. Mice deficient for chondromodulin-I, which protects against angiogenesis, expressed higher valvular level of vascular endothelial growth factor A (VEGFA) and developed neovascularization of the aortic valve accompanied by a thickening of aortic valve leaflets [37]. Therefore, it is possible that inflammation, neoangiogenesis, and remodeling of the ECM are intertwined in promoting the development of CAVD. Cytokines produced by inflammatory cells and VICs also play an important role to the pathophysiology of CAVD.  Mice deficient for the interleukin-1 receptor antagonist (IL1-Ra), which acts as a decoy receptor and prevents the activation of the IL1 receptor, develop CAVD [38].

92

Of interest, the double knockout mice deficient in both IL1-Ra and TNF-α are protected against CAVD [38]. It is thus possible that TNF-α is one key regulator of ectopic aortic valve mineralization. In isolated SMCs, TNF-α promotes the expression of Msx2, a homeobox transcription factor that entrains the osteogenic transition of cells [39]. Also, the expression of TNF-related apoptosis-inducing ligand (TRAIL) is increased in CAVD and promotes apoptosis-­mediated mineralization of isolated VIC cultures [40]. The receptor activator of nuclear factor kappa-B ligand (RANKL), a member of the TNF family, is expressed by CD4+ and CD8+ cells and could play a role in CAVD.  RANK is activated by RANKL and promotes in isolated VIC cultures the mineralization process. A study based on immunohistology has revealed that RANKL is present in stenotic aortic valves, whereas the expression of osteoprotegerin (OPG), a circulating and decoy receptor for RANKL, was decreased in the same tissues [41]. Also, in LDLR−/− mice, the administration of OPG decreased the mineralization of the aortic valve and the expression of osteogenic genes [42]. IL6 is a multifaceted and pleiotropic cytokine that is expressed by a wide variety of cells. A study using a transcriptomic approach has shown that IL6 is highly expressed during CAVD [43]. Upon a treatment with an osteogenic medium, VICs produced large amount of IL6, which promoted the expression of BMP2, a potent morphogen that drives the osteogenic transition of cells [43]. Of interest, IL6 is also a potent factor involved in the EndoMT process in the aortic valve. As such, it is possible that IL6 participates to the remodeling process of the aortic valve by promoting the recruitment and formation of mesenchymal cells with osteogenic properties. Receptors that are sensitive to the danger-­ associated molecular pattern molecules (DAMPs) may also enhance the innate immune response and exacerbate inflammation during CAVD.  DAMPs are endogenous factors that can stimulate the Tolllike receptors (TLRs), which, in turn, activate VICs. Biglycan is an agonist of TLR-2, which is expressed by VICs. Stimulation of TLR-2 with biglycan promotes the activation of the NF-κB

G. Mkannez et al.

pathway and expression of phospholipid transfer protein (PLTP) [26]. In turn, PLTP in the aortic valve may promote the retention of lipids by altering apolipoprotein A1 and the reverse cholesterol transport. Moreover, studies have shown that biglycan and OxLDL may promote an osteogenic activity in VICs through TLR-2 and TLR-4 [44]. Increasing evidence suggest that the adaptive immune system may also play a role in the pathogenesis of CAVD [45]. The proportion of circulating CD3+ cells that expresses HLA-DR is increased in patients with CAVD.  Moreover, in mineralized aortic valves, there is a clonal expansion of the TCR repertoire. Analysis of explanted stenotic aortic valves also revealed that CD8+ and CD28 null cells were present in the vicinity of mineralized areas. Though functional studies are clearly needed to address the role of adaptive immune response in CAVD, these observational studies performed in humans indicate that lymphocytes are activated during the mineralization of the aortic valve.

5.3.4 T  he Renin Angiotensin System and CAVD Accumulation of visceral ectopic fat, a feature of the metabolic syndrome, is associated with hypertension, a prevalent risk factor in patients with CAVD [46]. In prehypertensive patients with CAVD, the circulating level of angiotensin II is elevated and correlated with the expression of IL6  in surgically explanted stenotic aortic valves [47]. The administration of angiotensin II in mice resulted in a thickening of the aortic valve [48]. Studies have shown that angiotensin-­ converting enzyme (ACE) is expressed in CAVD and co-localized with angiotensin II [49]. Moreover, mastocyte-derived chymase, an angiotensin II-producing enzyme, is elevated in surgically explanted stenotic aortic valves [50]. Observational studies have shown that administration of angiotensin receptor blockers (ARBs) was associated with a lower fibrocalcific remodeling score of the aortic valve and with a slower progression rate of aortic stenosis [51].

5  Molecular Mechanisms of Aortic Valve Pathology

In ­hypercholesterolemic rabbit, the thickening of the aortic valve is diminished by the blockade of angiotensin receptors [52]. These observational data suggest that activation of the renin angiotensin system may participate to the development of CAVD, but further mechanistic studies are needed to uncover key processes behind these findings.

5.3.5 Purinergic System A seminal paper by Bertazzo and colleagues has shown that mineralization of the aortic valve in humans is composed of highly crystalline apatite of calcium organized as microparticles that tend to coalesce [53]. In isolated VICs, we observed that upon a treatment with a mineralizing medium, VICs secrete ectonucleotidase-rich microparticles that progressively accumulate hydroxyapatite of calcium [54]. These findings highlight that a regulated process under the control of VICs is involved during ectopic mineralization of the aortic valve. The ectonucleotidase is a family of enzymes located at the cell membrane that produces nucleotide and nucleoside derivatives. By driving the metabolism of extracellular nucleotides, the ectonucleotidases exert a powerful control over purinergic signaling and the level of phosphate-­ pyrophosphate [55]. Single nucleotide polymorphisms (SNPs) for ectonucleotide pyrophosphatase/phosphodiesterase-1 (ENPP1) have been associated with CAVD.  One SNP located in the intron 9 was associated with higher level of ENPP1 in surgically explanted aortic valves [56]. ENPP1 transforms adenosine triphosphate (ATP), which is secreted in minute amount by VICs, into adenosine monophosphate (AMP) and pyrophosphate, an inhibitor of mineralization. However, overexpression of ENPP1 in VICs induced the mineralization of cell cultures [56]. In VICs, during mineralization of cell cultures, ENPP1 is overexpressed along with alkaline phosphatase (ALP) and 5’nucleotidase (NT5E)/CD73 [57]. ALP transforms pyrophosphate into phosphate with strong osteogenic and mineralizing properties, whereas NT5E/CD73 uses AMP as a substrate

93

and produces adenosine [57, 58]. In turn, adenosine promotes in VICs the mineralization of cell culture through the adenosine receptor A2a (A2aR), which increases, in a feed forward mechanism, the expression of ectonucleotidases [57]. In addition, consequential to the high ectonucleotidase activity during the mineralization of the aortic valve, purinergic signaling through P2Y2 receptor (P2Y2R) is decreased. In VICs, P2Y2R triggers survival signal and prevents apoptosis-mediated mineralization of the aortic valve [56]. Also, Bouchareb et  al. provided the first evidence that VICs can resorb mineral through the activation of carbonic anhydrase XII (CAXII) through a pathway involving P2Y2R [59]. Following stimulation of P2Y2R, CAXII is recruited to the cell membrane of VICs where it acidifies the ECM.  By using nanobiosensors, it was found that this process lowers the pericellular pH down to 5.5, which is sufficient to dissolve mineral [59, 60]. In mice with CAVD, the administration of a P2Y2R agonist lowered the amount of mineral in the aortic valve, which was measured by microCT, and decreased the transaortic velocities [59]. Taken together, these data suggest that a dysregulation of the purinergic system exerts an important control on the osteogenic fate of VICs and the resorption of mineral in the aortic valve.

5.3.6 NOTCH and Wnt Pathways On histologic examination, ~15% of surgically explanted stenotic, mineralized aortic valves show the presence of osteochondrogenic metaplasia. These observations suggest that the expression of genes that promote an osteogenic transdifferentiation may play a role during CAVD. In 2005, Garg and colleagues have found mutations in the NOTCH1 gene in families of patients with BAV, which are prone to develop CAVD [61]. Also, rare mutations in the NOTCH1 gene were later documented in patients with CAVD and a tricuspid anatomy [62]. The NOTCH1 receptor binds to the delta-like proteins and Jagged1–2 and following this event activates the γ-secretase complex that cleaves the receptor

G. Mkannez et al.

94

into its active notch intracellular domain (NICD). In the nucleus, the NICD combines with recombining binding protein suppressor of hairless (RBPjk) and activates its downstream targets, which include the hairy family of repressors (Hrt). In turn, Hrt represses the expression of RUNX2 and BMP2, two important drivers of osteogenic transition in VICs [61]. Of note, the NICD also interferes with the Wnt pathway through a process that remains to be fully elucidated. Wnt3a is increased in CAVD and may be involved in the activation of an osteogenic program through the activation of a co-receptor formed by Lrp5/6 and frizzled [63]. After activation of Lrp5/6 and frizzled, GSK-3β is recruited by an inhibitory complex and cannot phosphorylate β-catenin. As a result, β-catenin is stabilized and translocates to the nucleus where it promotes the expression of BMP2. Hence, a dysregulation of both NOTCH1 and Wnt pathways promotes the formation of osteoblast-­like cells and plays a significant role in the osteogenic differentiation of VICs.

5.4

Bicuspid Aortic Valve

5.4.1 Altered Biomechanics BAV, a common developmental anomaly (prevalence 1–2%) with a male predominance (3:1), is associated with aortic valve and aorta disorders. Late complications include CAVD, aortic regurgitation, aorta dilation, and aortic dissection [64]. BAV represents 30–50% of surgically explanted aortic valves for CAVD [65]. On microscopic examination, the aortic valve of newborn infants with BAV shows alteration of the trilaminar architecture of valve leaflets [66]. The compartmentalization of VICs is perturbed, and BAVs are characterized by an abundance of proteoglycans. Moreover, BAV is characterized by an abnormal blood flow pattern, which may participate to the development of CAVD and aortopathy/aortic regurgitation. It is likely that both altered structural components and abnormal blood flow participate to increase the mechanical load on the leaflets. Studies have highlighted that mechanical

stress is maximal at the raphe, an area where mineralization is exacerbated during CAVD.  We showed that VICs located in the raphe expressed a higher level of ectonucleotidases and were subjected to a higher mechanical stress with activation of the RhoA pathway [54]. In turn, activation of RhoA by mechanical constraint activates a series of events in VICs that promote the production of mineralized microparticles in the extracellular matrix.

5.4.2 Genetic Control As previously highlighted, frameshift mutations in NOTCH1 are associated with BAV [61]. Also, rare (4% of cases) non-synonymous variants in GATA5, which encodes for a transcription factor, have been documented in patients with BAV [67]. GATA5−/− mice develop BAV in about 25% of littermates [68]. Of interest, the NOS3 gene has conserved binding sequence for GATA5  in its promoter, and a significant proportion (~20– 30%) of NOS3−/− mice also develop BAV. Taken together, these data suggest that GATA5 and NOS3 may have reciprocal crosstalk and are important factors for the development of the aortic valve during the embryonic period.

5.4.3 Aortopathy BAV is often accompanied by structural abnormalities of the aortic wall, which may predispose to the development of aortic dilation-­dissection and aortic regurgitation. Additionally, there is an association between BAV and aortic coarctation, pointing to developmental anomaly of the aortic wall in BAV.  In mice, a disruption of NOTCH and its downstream signaling in the second heart field, which plays a role in outflow tract septation and patterning of cells derived from the neural crest during embryogenesis, led to aortic arch abnormalities, disorganized aortic wall histology, and the development of aortic regurgitation with a bicuspid-like valve [69]. Of note, neural crest cells participate to the development of aorta and are involved in the late phase of

5  Molecular Mechanisms of Aortic Valve Pathology

v­ alvulogenesis. In patients with BAV, histologic examination of human dilated aortas has shown the presence of fragmented elastic fibers and higher levels of MMP2. The expression of lysyl hydroxylase, a key factor involved in collagen cross-linking, is decreased in dilated aortas. Also, apoptosis is increased, whereas the levels of collagen types I and III are decreased at the convexity compared to the concavity of dilated aortas in BAV [70]. This asymmetrical pattern in BAV-associated aortopathy suggests that hemodynamic factors related to an abnormal blood flow pattern in BAV contribute to the dilation of aorta. Hence, it is likely that several factors, related to both inborn genetic defects and abnormal blood flow, participate to aortic wall abnormality in patients with BAV.

5.5

Conclusion and Perspectives

The aortic valve, including the root and sinuses, is a dynamic structure exposed to a high mechanical load that necessitates adaptive mechanisms to repair and maintain homeostasis of the ECM.  The development of pathologies such as CAVD and aortic regurgitation is intimately linked to maladaptive response of VECs and VICs to environmental cues. Complex molecular cascades promote the mineralization of the aortic valve. In the last decade, there have been a growing number of reports that have underlined the pathways involved in the fibrocalcific remodeling process of the aortic valve. These works have underscored the role of key pathways involved in aortic valve mineralization and may open new therapeutic opportunities to treat pharmacologically CAVD [71]. However, further work is needed to refine and identify druggable targets that would impact clinical outcomes such as the progression of aortic valve stenosis. Acknowledgments  The work of the authors is supported by Canadian Institutes of Health Research grants to P.M. (MOP114893, MOP114893, MOP114893, MOP365029), the Heart and Stroke Foundation of Canada, and the Quebec Heart and Lung Institute Fund. P.M. holds a FRQS Research Chair on the Pathobiology of Calcific Aortic Valve Disease.

95

References 1. Mathieu P, Boulanger MC.  Basic mechanisms of calcific aortic valve disease. Can J Cardiol. 2014;30:982–93. 2. Guauque-Olarte S, Messika-Zeitoun D, Droit A, Lamontagne M, Tremblay-Marchand J, Lavoie-­ Charland E, et  al. Calcium signaling pathway genes RUNX2 and CACNA1C are associated with calcific aortic valve disease. Circ Cardiovasc Genet. 2015;8:812–22. 3. Hinton RB, Yutzey KE.  Heart valve structure and function in development and disease. Annu Rev Physiol. 2011;73:29–46. 4. Wirrig EE, Yutzey KE.  Conserved transcriptional regulatory mechanisms in aortic valve development and disease. Arterioscler Thromb Vasc Biol. 2014;34:737–41. 5. Latif N, Sarathchandra P, Chester AH, Yacoub MH. Expression of smooth muscle cell markers and co-activators in calcified aortic valves. Eur Heart J. 2015;36:1335–45. 6. Chen JH, Yip CY, Sone ED, Simmons CA.  Identification and characterization of aortic valve mesenchymal progenitor cells with robust osteogenic calcification potential. Am J Pathol. 2009;174:1109–19. 7. Mahler GJ, Farrar EJ, Butcher JT. Inflammatory cytokines promote mesenchymal transformation in embryonic and adult valve endothelial cells. Arterioscler Thromb Vasc Biol. 2013;33:121–30. 8. Lincoln J, Yutzey KE. Molecular and developmental mechanisms of congenital heart valve disease. Birth Defects Res A Clin Mol Teratol. 2011;91:526–34. 9. Hajdu Z, Romeo SJ, Fleming PA, Markwald RR, Visconti RP, Drake CJ. Recruitment of bone marrow-­ derived valve interstitial cells is a normal homeostatic process. J Mol Cell Cardiol. 2011;51:955–65. 10. Nomura A, Seya K, Yu Z, Daitoku K, Motomura S, Murakami M, et  al. CD34-negative mesenchymal stem-like cells may act as the cellular origin of human aortic valve calcification. Biochem Biophys Res Commun. 2013;440:780–5. 11. Choi JH, Do Y, Cheong C, Koh H, Boscardin SB, Oh YS, et  al. Identification of antigen-presenting dendritic cells in mouse aorta and cardiac valves. J Exp Med. 2009;206:497–505. 12. Simmons CA, Grant GR, Manduchi E, Davies PF.  Spatial heterogeneity of endothelial phenotypes correlates with side-specific vulnerability to calcification in normal porcine aortic valves. Circ Res. 2005;96:792–9. 13. El-Hamamsy I, Balachandran K, Yacoub MH, Stevens LM, Sarathchandra P, Taylor PM, et  al. Endothelium-dependent regulation of the mechanical properties of aortic valve cusps. J Am Coll Cardiol. 2009;53:1448–55. 14. Richards J, El-Hamamsy I, Chen S, Sarang Z, Sarathchandra P, Yacoub MH, et  al. Side-specific

96 endothelial-dependent regulation of aortic valve calcification: interplay of hemodynamics and nitric oxide signaling. Am J Pathol. 2013;182:1922–31. 15. Bosse K, Hans CP, Zhao N, Koenig SN, Huang N, Guggilam A, et al. Endothelial nitric oxide signaling regulates Notch1  in aortic valve disease. J Mol Cell Cardiol. 2013;60:27–35. 16. Rajamannan NM, Evans FJ, Aikawa E, Grande Allen KJ, Demer LL, Heistad DD, et  al. Calcific aortic valve disease: not simply a degenerative process: a review and agenda for research from the National Heart and Lung and Blood Institute Aortic Stenosis Working Group. Executive summary: calcific aortic valve disease-2011 update. Circulation. 2011;124:1783–91. 17. Lindman BR, Clavel MA, Mathieu P, Iung B, Lancellotti P, Otto CM, et al. Calcific aortic stenosis. Nat Rev Dis Primers. 2016;2:16006. 18. Cote N, Mahmut A, Bosse Y, Couture C, Page S, Trahan S, et  al. Inflammation is associated with the remodeling of calcific aortic valve disease. Inflammation. 2013;36:573–81. 19. Akahori H, Tsujino T, Naito Y, Yoshida C, Lee-­ Kawabata M, Ohyanagi M, et  al. Intraleaflet haemorrhage as a mechanism of rapid progression of stenosis in bicuspid aortic valve. Int J Cardiol. 2013;167:514–8. 20. Laguna-Fernandez A, Carracedo M, Jeanson G, Nagy E, Eriksson P, Caligiuri G, et al. Iron alters valvular interstitial cell function and is associated with calcification in aortic stenosis. Eur Heart J. 2016;37:3532–5. 21. O'Brien KD, Reichenbach DD, Marcovina SM, Kuusisto J, Alpers CE, Otto CM. Apolipoproteins B, (a), and E accumulate in the morphologically early lesion of ‘degenerative’ valvular aortic stenosis. Arterioscler Thromb Vasc Biol. 1996;16:523–32. 22. Thanassoulis G, Campbell CY, Owens DS, Smith JG, Smith AV, Peloso GM, et al. Genetic associations with valvular calcification and aortic stenosis. N Engl J Med. 2013;368:503–12. 23. Mohty D, Pibarot P, Despres JP, Cote C, Arsenault B, Cartier A, et al. Association between plasma LDL particle size, valvular accumulation of oxidized LDL, and inflammation in patients with aortic stenosis. Arterioscler Thromb Vasc Biol. 2008;28:187–93. 24. Miller JD, Chu Y, Brooks RM, Richenbacher WE, Pena-Silva R, Heistad DD. Dysregulation of antioxidant mechanisms contributes to increased oxidative stress in calcific aortic valvular stenosis in humans. J Am Coll Cardiol. 2008;52:843–50. 25. Lassegue B, Griendling KK. NADPH oxidases: functions and pathologies in the vasculature. Arterioscler Thromb Vasc Biol. 2010;3:653–61. 26. Derbali H, Bosse Y, Cote N, Pibarot P, Audet A, Pepin A, et  al. Increased biglycan in aortic valve stenosis leads to the overexpression of phospholipid transfer protein via Toll-like receptor 2. Am J Pathol. 2010;176:2638–45. 2 7. Mahmut A, Boulanger MC, Fournier D, Couture C, Trahan S, Page S, et  al. Lipoprotein lipase

G. Mkannez et al. in aortic valve stenosis is associated with lipid retention and remodelling. Eur J Clin Investig. 2013;43:570–8. 28. Mahmut A, Boulanger MC, El HD, Fournier D, Bouchareb R, Despres JP, et  al. Elevated expression of lipoprotein-associated phospholipase A2 in calcific aortic valve disease: implications for valve mineralization. J Am Coll Cardiol. 2014;63:460–9. 29. Bouchareb R, Mahmut A, Nsaibia MJ, Boulanger MC, Dahou A, Lepine JL, et  al. Autotaxin derived from lipoprotein(a) and valve interstitial cells promotes inflammation and mineralization of the aortic valve. Circulation. 2015;132:677–90. 30. Capoulade R, Mahmut A, Tastet L, Arsenault M, Bedard E, Dumesnil JG, et  al. Impact of plasma Lp-PLA2 activity on the progression of aortic stenosis: the PROGRESSA study. JACC Cardiovasc Imaging. 2015;8:26–33. 31. Mahmut A, Mahjoub H, Boulanger MC, Fournier D, Despres JP, Pibarot P, et  al. Lp-PLA2 is associated with structural valve degeneration of bioprostheses. Eur J Clin Investig. 2014;44:136–45. 32. Mahmut A, Mahjoub H, Boulanger MC, Dahou A, Bouchareb R, Capoulade R, et  al. Circulating Lp-PLA2 is associated with high valvuloarterial impedance and low arterial compliance in patients with aortic valve bioprostheses. Clin Chim Acta. 2016;455:20–5. 33. Nsaibia MJ, Mahmut A, Boulanger MC, Arsenault BJ, Bouchareb R, Simard S, et al. Autotaxin interacts with lipoprotein(a) and oxidized phospholipids in predicting the risk of calcific aortic valve stenosis in patients with coronary artery disease. J Intern Med. 2016;280:509–17. 34. Mohty D, Pibarot P, Despres JP, Cartier A, Arsenault B, Picard F, et al. Age-related differences in the pathogenesis of calcific aortic stenosis: the potential role of resistin. Int J Cardiol. 2010;142:126–32. 35. Guauque-Olarte S, Droit A, Tremblay-Marchand J, Gaudreault N, Kalavrouziotis D, Dagenais F, et  al. RNA expression profile of calcified bicuspid, tricuspid and normal human aortic valves by RNA sequencing. Physiol Genomics. 2016;48:749–61. 36. Charest A, Pepin A, Shetty R, Cote C, Voisine P, Dagenais F, et al. Distribution of SPARC during neovascularisation of degenerative aortic stenosis. Heart. 2006;92:1844–9. 37. Yoshioka M, Yuasa S, Matsumura K, Kimura K, Shiomi T, Kimura N, et al. Chondromodulin-I maintains cardiac valvular function by preventing angiogenesis. Nat Med. 2006;12:1151–9. 38. Isoda K, Matsuki T, Kondo H, Iwakura Y, Ohsuzu F.  Deficiency of interleukin-1 receptor antagonist induces aortic valve disease in BALB/c mice. Arterioscler Thromb Vasc Biol. 2010;30:708–15. 39. Lee HL, Woo KM, Ryoo HM, Baek JH.  Tumor necrosis factor-alpha increases alkaline phosphatase expression in vascular smooth muscle cells via MSX2 induction. Biochem Biophys Res Commun. 2010;391:1087–92.

5  Molecular Mechanisms of Aortic Valve Pathology 40. Galeone A, Brunetti G, Oranger A, Greco G, Di BA, Mori G, et al. Aortic valvular interstitial cells apoptosis and calcification are mediated by TNF-­related apoptosis-inducing ligand. Int J Cardiol. 2013;169:296–304. 41. Kaden JJ, Bickelhaupt S, Grobholz R, Haase KK, Sarikoc A, Kilic R, et  al. Receptor activator of nuclear factor kappaB ligand and osteoprotegerin regulate aortic valve calcification. J Mol Cell Cardiol. 2004;36:57–66. 42. Weiss RM, Lund DD, Chu Y, Brooks RM, Zimmerman KA, El AR, et al. Osteoprotegerin inhibits aortic valve calcification and preserves valve function in hypercholesterolemic mice. PLoS One. 2013;8:e65201. 43. El Husseini D, Boulanger MC, Mahmut A, Bouchareb R, Laflamme MH, Fournier D, et  al. P2Y2 receptor represses IL-6 expression by valve interstitial cells through Akt: implication for calcific aortic valve disease. J Mol Cell Cardiol. 2014;72:146–56. 44. Zeng Q, Song R, Ao L, Xu D, Venardos N, Fullerton DA, et al. Augmented osteogenic responses in human aortic valve cells exposed to oxLDL and TLR4 agonist: a mechanistic role of Notch1 and NF-kappaB interaction. PLoS One. 2014;9:e95400. 45. Mathieu P, Bouchareb R, Boulanger MC. Innate and adaptive immunity in calcific aortic valve disease. J Immunol Res. 2015;2015:851945. 46. Capoulade R, Clavel MA, Dumesnil JG, Chan KL, Teo KK, Tam JW, et al. Impact of metabolic syndrome on progression of aortic stenosis: influence of age and statin therapy. J Am Coll Cardiol. 2012;60:216–23. 47. Cote N, Pibarot P, Pepin A, Fournier D, Audet A, Arsenault B, et al. Oxidized low-density lipoprotein, angiotensin II and increased waist circumference are associated with valve inflammation in prehypertensive patients with aortic stenosis. Int J Cardiol. 2010;145:444–9. 48. Fujisaka T, Hoshiga M, Hotchi J, Takeda Y, Jin D, Takai S, et  al. Angiotensin II promotes aortic valve thickening independent of elevated blood pressure in apolipoprotein-E deficient mice. Atherosclerosis. 2013;226:82–7. 49. O'Brien KD, Shavelle DM, Caulfield MT, McDonald TO, Olin-Lewis K, Otto CM, et  al. Association of angiotensin-converting enzyme with low-density lipoprotein in aortic valvular lesions and in human plasma. Circulation. 2002;106:2224–30. 50. Helske S, Lindstedt KA, Laine M, Mayranpaa M, Werkkala K, Lommi J, et al. Induction of local angiotensin II-producing systems in stenotic aortic valves. J Am Coll Cardiol. 2004;44:1859–66. 51. Cote N, Mahmut A, Fournier D, Boulanger MC, Couture C, Despres JP, et  al. Angiotensin receptor blockers are associated with reduced fibrosis and interleukin-6 expression in calcific aortic valve disease. Pathobiology. 2014;81:15–24. 52. Arishiro K, Hoshiga M, Negoro N, Jin D, Takai S, Miyazaki M, et  al. Angiotensin receptor-1 blocker inhibits atherosclerotic changes and endothelial disruption of the aortic valve in hypercholesterolemic rabbits. J Am Coll Cardiol. 2007;49:1482–9.

97 53. Bertazzo S, Gentleman E, Cloyd KL, Chester AH, Yacoub MH, Stevens MM.  Nano-analytical electron microscopy reveals fundamental insights into human cardiovascular tissue calcification. Nat Mater. 2013;12:576–83. 54. Bouchareb R, Boulanger MC, Fournier D, Pibarot P, Messaddeq Y, Mathieu P.  Mechanical strain induces the production of spheroid mineralized microparticles in the aortic valve through a RhoA/ROCK-dependent mechanism. J Mol Cell Cardiol. 2014;67:49–59. 55. Mathieu P.  Pharmacology of ectonucleotidases: relevance for the treatment of cardiovascular disorders. Eur J Pharmacol. 2012;696:1–4. 56. Cote N, El Husseini D, Pepin A, Guauque-Olarte S, Ducharme V, Bouchard-Cannon P, et  al. ATP acts as a survival signal and prevents the mineralization of aortic valve. J Mol Cell Cardiol. 2012;52:1191–202. 57. Mahmut A, Boulanger MC, Bouchareb R, Hadji F, Mathieu P. Adenosine derived from ecto-­nucleotidases in calcific aortic valve disease promotes mineralization through A2a adenosine receptor. Cardiovasc Res. 2015;106:109–20. 58. Mathieu P, Voisine P, Pepin A, Shetty R, Savard N, Dagenais F.  Calcification of human valve interstitial cells is dependent on alkaline phosphatase activity. J Heart Valve Dis. 2005;14:353–7. 59. Bouchareb R, Cote N, Marie CB, Le Quang K, El Huseini D, Asselin J, et  al. Carbonic anhydrase XII in valve interstitial cells promotes the regression of calcific aortic valve stenosis. J Mol Cell Cardiol. 2015;82:104–15. 60. Asselin J, Roy C, Boudreau D, Messaddeq Y, Bouchareb R, Mathieu P.  Supported core-shell nanobiosensors for quantitative fluorescence imaging of extracellular pH.  Chem Commun (Camb). 2014;50:13746–9. 61. Garg V, Muth AN, Ransom JF, Schluterman MK, Barnes R, King IN, et  al. Mutations in NOTCH1 cause aortic valve disease. Nature. 2005;437:270–4. 62. Ducharme V, Guauque-Olarte S, Gaudreault N, Pibarot P, Mathieu P, Bosse Y. NOTCH1 genetic variants in patients with tricuspid calcific aortic valve stenosis. J Heart Valve Dis. 2013;22:142–9. 63. Caira FC, Stock SR, Gleason TG, McGee EC, Huang J, Bonow RO, et al. Human degenerative valve disease is associated with up-regulation of low-density lipoprotein receptor-related protein 5 receptor-mediated bone formation. J Am Coll Cardiol. 2006;47:1707–12. 64. Michelena HI, Prakash SK, Della CA, Bissell MM, Anavekar N, Mathieu P, et  al. Bicuspid aortic valve: identifying knowledge gaps and rising to the challenge from the International Bicuspid Aortic Valve Consortium (BAVCon). Circulation. 2014;129:2691–704. 65. Unger P, Clavel MA, Lindman BR, Mathieu P, Pibarot P. Pathophysiology and management of multivalvular disease. Nat Rev Cardiol. 2016;13:429–40. 66. Hinton RB Jr, Lincoln J, Deutsch GH, Osinska H, Manning PB, Benson DW, et al. Extracellular matrix

98 remodeling and organization in developing and diseased aortic valves. Circ Res. 2006;98:1431–8. 67. Padang R, Bagnall RD, Richmond DR, Bannon PG, Semsarian C.  Rare non-synonymous variations in the transcriptional activation domains of GATA5  in bicuspid aortic valve disease. J Mol Cell Cardiol. 2012;53:277–81. 68. Laforest B, Andelfinger G, Nemer M.  Loss of Gata5  in mice leads to bicuspid aortic valve. J Clin Invest. 2011;121:2876–87. 69. Jain R, Engleka KA, Rentschler SL, Manderfield LJ, Li L, Yuan L, et al. Cardiac neural crest orchestrates

G. Mkannez et al. remodeling and functional maturation of mouse semilunar valves. J Clin Invest. 2011;121:422–30. 70. Cotrufo M, Della CA, De Santo LS, Quarto C, De FM, Romano G, et al. Different patterns of extracellular matrix protein expression in the convexity and the concavity of the dilated aorta with bicuspid aortic valve: preliminary results. J Thorac Cardiovasc Surg. 2005;130:504–11. 71. Mathieu P, Boulanger MC, Bouchareb R. Molecular biology of calcific aortic valve disease: towards new pharmacological therapies. Expert Rev Cardiovasc Ther. 2014;12:851–62.

6

Functional and Morphological Interplay of the Aortic Valve, the Aortic Root, and the Left Ventricle Marie-Annick Clavel and Philippe Pibarot

6.1

Introduction

A harmonious anatomical and functional interrelation between the left ventricle, the aortic valve, and the aorta is key to ensure adequate blood flow and blood pressure as much as necessary to assure perfusion to all body organs. Any abnormality, congenital or acquired, of the left ventricular (LV) myocardium, LV outflow tract, aortic valve, aortic root, or arteries may compromise this anatomical and functional coupling between the left ventricle and the arterial system and ultimately lead to heart failure. The purpose of this chapter is to describe the morphological and functional interplay between the left ventricle, the aortic valve, and the aorta and its implications with regard to the diagnosis and treatment of aortic valve disease.

M.-A. Clavel · P. Pibarot (*) Département de Médecine, Université Laval, Québec, QC, Canada Canada Research Chair in Valvular Heart Diseases, Institut Universitaire de Cardiologie et de Pneumologie de Québec, Québec Heart and Lung Institute, Québec, QC, Canada e-mail: [email protected]; [email protected]

6.2

Anatomical Interaction Between the Left Ventricle, Aortic Valve, and Aortic Root

6.2.1 Anatomical Interrelation Between the LV Outflow Tract, Aortic Valve, and Aortic Root The aortic valve is composed of three semilunar leaflets that are attached to the aortic root. Thus, the aortic root can be divided by the semilunar attachment of the leaflets into supra-valvular and sub-valvular components (Fig.  6.1) [1]. The supra-valvular component is represented by the three sinuses of Valsalva, and two of them host a coronary ostium. The flow vortices created in the sinuses lead to stress reduction on the aortic leaflets and support coronary blood flow. Hence, effacement of the sinuses may impair coronary flow and increase the mechanical stress and ensue structural degeneration of aortic valve leaflets. The sub-valvular component consists in the LV outflow tract that is surrounded by the myocardium and by fibrous tissue. Under each commissure of the leaflets lies one of the three inter-leaflet triangles, which are, in fact, extensions of the LV outflow tract and reach the level of the sino-tubular junction in the area of the commissures. The triangle between the left and non-coronary sinuses forms a part of the mitroaortic curtain that leads to the anterior mitral leaflet. The triangle between the right and

© Springer-Verlag GmbH Austria, part of Springer Nature 2019 O. H. Stanger et al. (eds.), Surgical Management of Aortic Pathology, https://doi.org/10.1007/978-3-7091-4874-7_6

99

M.-A. Clavel and P. Pibarot

100 Fig. 6.1  Anatomy of the LV outflow tract, aortic valve, and aortic root. (a) Three-­ dimensional arrangement of the aortic root, which contains three circular “rings,” but with the leaflets suspended within the root in crown-like fashion. (b) The leaflets have been removed from this specimen of the aortic root, showing the location of the three rings relative to the crown-like hinges of the leaflets. VA ventriculo-­ arterial, A-M aortic-­ mitral. Adapted with permission from [1]

a

Sinutubular junction

Crown-like ring Anatomic ventriculoarterial junction Virtual ring formed by joining basal attachments of aortic valvar leaflets

b Anatomic VA junction

Sinutubular junction Crown-like ring

A-M curtain

Membranous septum

non-coronary sinuses forms part of the membranous septum [2]. This area is of particular importance during aortic valve procedures, as injury at this level can lead to conduction abnormalities, which may require the implantation of a permanent pacemaker. Due to their semilunar configuration, the leaflets of the aortic valve are attached in crown-­like fashion with a basal ring at the bottom of leaflet insertion, i.e., the aortic “annulus,” and an upper ring at the level of the sino-tubular junction (Fig.  6.1). The aortic annulus that is measured clinically by echocardiography, computed tomography (CT), or cardiac magnetic resonance (CMR) for diagnostic purposes of for therapeutic purposes (e.g., selection of valve size for transcatheter aortic valve implantation; TAVI) corresponds to the

Virtual ring formed by joining basal attachments of aortic valvar leaflets

circle at the base of the crown formed by the leaflets. However, the aortic a­nnulus is not a real and distinct anatomic entity as it does not correspond to the basal contour of the aortic valve which is not circular but rather has a tripoint crown shape. Nevertheless, the aortic annulus represents the area of the smallest diameter in the blood path between the left ventricle and the aorta and thus determines the size of the prosthetic valve to be implanted.

6.2.2 Bicuspid Aortic Valve and Aortopathy The normal architecture of the aortic valve is a 3-cusp configuration. However, some subjects may have a bicuspid, unicuspid, or quadricuspid

6  Functional and Morphological Interplay of the Aortic Valve, the Aortic Root, and the Left Ventricle

configuration. Bicuspid aortic valve (BAV) is the most common congenital heart defect, affecting 1–2% of the population and three men for one woman [3, 4]. All combination of cusps fusion with and without raphe are theoretically possible; however, the most common types are fusion of the right and left coronary leaflets (60%) and fusion of the right and the non-coronary leaflets (35%) (Fig.  6.2) [5]. Despite the evidence of an autosomal dominant pattern of BAV inheritance with variable expression and incomplete penetrance in families, the genetic determinants of BAV are far from being elucidated [6, 7]. Patients with BAV have a high risk to develop aortic valve dysfunction, i.e., stenosis and/or regurgitation. In particular, the lifetime risk of aortic valve replacement for aortic stenosis in subjects with BAV is about 50%.

a

BAV has been associated with other congenital defects and also with aorta dilation, aneurism, and dissection [4]. The link between BAV and aorta dilation/aneurism is probably multifactorial with a genetic basis and a possible impact of flow pattern induced by BAV type (Fig. 6.3a, b) [8]. The eccentric aortic blood flow created by the BAV increases the radial pressure and thus the stress on the aortic wall, thus causing a dilation of the aortic root [7, 9]. This phenomenon is more pronounced in patients with BAV and aortic stenosis (Fig. 6.3a, b) [7]. On the other hand, the genetic basis for aorta dilation is supported by the fact that aortopathy is often present in relatives of patients with BAV.  The structural support and elasticity of the aorta are achieved by alternating layers of elastic lamellae and smooth muscle cells (Fig. 6.3c). In subjects with

b

RCA Single cusp

Fused cusps

101

c Raphe

RC LC

LCA

NC

d

Fused cusps

e

Fused cusps

Fig. 6.2  Comparison of tricuspid and bicuspid aortic valve structures. Schematic representation of a normal-­ tricuspid-­aortic valve with the three cusps (a), a bicuspid valve with right non-coronary cusp fusion and one raphe (the line of union between the fused cups) (b), a bicuspid valve with fusion of the right and left coronary cusps and

no raphe (c), a bicuspid valve with right-left coronary cusp fusion and one raphe (d), and a bicuspid valve with fusion of the left and non-coronary cups and one raphe (e). LC left coronary, LCA left coronary artery, NC non-­ coronary, RC right coronary, RCA right coronary artery. Reproduced with permission from [5]

M.-A. Clavel and P. Pibarot

102

a1

a2

Right-left fusion pattern

Raphe

b1

b3

b2

Right-noncoronary fusion pattern

6  Functional and Morphological Interplay of the Aortic Valve, the Aortic Root, and the Left Ventricle

c1

103

Elastin and collagen

Fibrillin 1 microfibril Smooth-muscle cells

c2 Disrupted elastin and collagen

Loss of fibrillin 1 microfibrils

Smooth-muscle cell loss

MMP release

Fig. 6.3  Morphological and pathophysiological features of aortopathy in subjects with a bicuspid aortic valve. Panel a shows the fusion configuration of the aortic valve cusps lays the foundation for changes in aortic wall shear stress and the resultant flow pattern. In the right-left fusion pattern (A-1), the jet is directed toward the right anterior wall of the ascending aorta, where it travels in a right-­ handed helical direction to promote dilatation predominantly of the ascending aorta. In the pattern with fusion of the right and non-coronary cusps (A-2), the jet is directed toward the posterior wall of the aorta, whereby the pattern of wall shear stress it causes may promote aortic dilatation within the proximal arch. Reproduced with permission from [7]. Panel b shows the flow patterns in bicuspid aortic valve disease. (B-1) Normal flow pattern, (B-2) righthanded helical flow, and (B-3) left-­handed helical flow. The systolic flow angle (θ) is demonstrated in B-2: the

angle between the aortic midline (dashed) and the instantaneous mean flow vector at peak systole (arrow). Reproduced with permission from [9]. Panel c shows the structural support and elasticity are afforded to the aorta by means of alternating layers of elastic lamellae and smooth muscle cells. At the histologic level, the smooth muscle cells in the aorta in persons with tricuspid valves are secured to the adjacent elastin and collagen matrix by fibrillin 1 microfibrils (C-1). The aorta in persons with bicuspid valves may be deficient in fibrillin 1. This deficiency culminates in a disrupted architecture whereby smooth muscle cells detach, accompanied by a surge in local levels of matrix metalloproteinases (MMPs), leading to loss of integrity in the extracellular matrix and the accumulation of apoptotic cells. These events may lead to an aorta with weakened structural integrity and reduced elasticity (C-2). Reproduced with permission from [7]

M.-A. Clavel and P. Pibarot

104

tricuspid valves, the smooth muscle cells in the aorta are secured to the adjacent elastin and collagen matrix by fibrillin 1 microfibrils. The aorta of subjects with BAV may be deficient in fibrillin 1, which causes a disruption of the histological architecture of the aortic wall whereby smooth muscle cells detach, leading to weakened structural integrity and reduced elasticity of the aorta [7]. Disruption of the histologic architecture of the aortic wall has also been observed in subjects with bicuspid valves in the absence of any genetic or mechanical explanation.

6.3

Functional Interaction Between the Left Ventricle, Aortic Valve, and Aorta

6.3.1 Ventriculo-Valvulo-Arterial Coupling The dynamic interaction between the heart and the systemic circulation allows the cardiovascular system to be efficient in providing adequate cardiac output and arterial pressures to ensure adequate organ perfusion in different physiological (rest and exercise) conditions. The ­cardiovascular system indeed works better when the heart and the arterial system are coupled. The ventriculo-arterial coupling is achieved by the continuous modulation of the arterial system compliance and resistance with respect to LV systolic performance, and this physiological process is key to maintain adequate LV stroke volume and cardiac output. Because LV stroke volume depends on myocardial contractility and loading conditions, i.e., preload and afterload, myocardial, valvular, and arterial dysfunction can lead to ventriculo-arterial decoupling with resulting decrease in stroke volume, cardiac output, and organ perfusion. Ventriculo-arterial coupling can be defined as the ratio of the effective arterial elastance (Ea) to the ventricular elastance (Ees) measured on the LV pressure-volume loop (Fig.  6.4) [10]. Ea is calculated as the ratio of the end-systolic pressure to the LV stroke volume and is a measure of the arterial hemodynamic load imposed on the

left ventricle. Ees describes the maximal pressure that can be developed by the ventricle at any given LV volume. Ees is an index of myocardial contractility that is relatively insensitive to changes in preload, afterload, and heart rate. The Ea/Ees ratio is useful to evaluate the mechanical efficiency of the cardiovascular system and the interaction between cardiac performance and systemic vascular function. An unbalance between Ea and Ees as reflected by an increase or decrease in the Ea/Ees ratio may lead to heart failure.

6.3.2 I mpact of Aortic Stenosis on Ventriculo-Arterial Coupling Aortic stenosis (AS) is the most common disease of the aortic valve. In AS, the LV afterload is increased, and as a consequence, the LV stroke volume is reduced, and the LV peak systolic pressure and end-diastolic pressures are increased (Fig.  6.4). The stroke work, which is the area enclosed within the LV pressure-volume loop, is increased. With advanced severity of AS, Ea increases, and Ees decreases resulting in a marked augmentation of the Ea/Ees ratio and ventriculo-arterial “decoupling.” A ratio > 1.0 is a marker for development of heart failure. Ea is calculated from the LV end-systolic pressure and therefore primarily reflects the arterial load but does not include the valvular load per se. This valvular load is negligible in patients with a normal aortic valve but is markedly increased in patients with severe AS. To estimate the true total, i.e., arterial + valvular hemodynamic load that the left ventricle is facing in patients with AS, one can calculate the valvuloarterial impedance, which is the ratio of the LV peak systolic pressure to the stroke volume index (Fig. 6.5b) [11, 12]. The LV systolic pressure can be estimated noninvasively by Doppler echocardiography by adding the mean transvalvular gradient to the systolic blood pressure. The rationale for using the stroke volume index rather than the unindexed stroke volume as in the Ea is that a small-sized subject may have a much smaller stroke volume than a large-sized subject but

6  Functional and Morphological Interplay of the Aortic Valve, the Aortic Root, and the Left Ventricle

Isovolumetric relaxation

Ea Isovolumetric contraction

40 20

EDPVR Mitral valve opens

0

50

70

End-systolic volume

90 Stroke volume

110

130

Ees

200 160

Ea

120 80 40

EDPVR

0 0

80 Ea

40 0

0

40

120 160 200 40 80 Left ventricular volume, (ml)

80

120

160

EDPVR

200

Left ventricular volume, (ml)

End-diastolic volume

Left ventricular volume, (ml)

c Left interventricular pressure, (mm Hg)

Mitral valve closes

Filling period

0

120

d Paradoxical Low Flow Low Gradient AS

200

Ees

60

160

160

Classical Low Flow Low Gradient AS

s

Aortic valve opens

80

200

Ee

Aortic valve closes

Left interventricular pressure, (mm Hg)

100

Ee s

Ejection period

120

Left interventricular pressure, (mm Hg)

b Ees

Left interventricular pressure, (mm Hg)

a

105

120 Ea

Ea

80 40

EDPVR

EDPVR

0 0

40 80 120 160 200 Left ventricular volume, (ml)

Fig. 6.4  Ventriculo-valvulo-arterial coupling in normal condition, aortic regurgitation, and aortic stenosis. This figure shows the LV pressure-volume loops for different situations including normal condition (a), severe aortic regurgitation (b), severe aortic stenosis with high gradient (c), paradoxical (preserved LV ejection fraction) and classical (reduced LV ejection fraction) low-flow, low-­

gradient aortic stenosis (d). The dashed pressure-volume loop on panels b, c, and d represents the loop in normal condition. AS aortic stenosis, Ea arterial elastance, EDPVR end-diastolic pressure-volume relationship, Ees ventricular elastance (end-systolic pressure-volume relationship)

nonetheless similar arterial and LV pressures. The valvulo-arterial impedance represents the valvular and arterial factors that oppose ventricular ejection by absorption of the mechanical energy developed by the left ventricle. Of the difference of Ea, this parameter also includes the valvular load (i.e., the mean transvalvular ­gradient), and it accounts for the interindividual variability in body size. Values of impedance >4.5  mmHg.mL−1.m2 have been shown to

p­ rovide incremental value to predict symptoms, heart  failure, and mortality in patients with AS [12, 13]. The stroke volume index is a good surrogate marker of the ventriculo-arterial coupling and of the overall performance of the cardiovascular system in patients with AS.  Several studies and meta-analyses reported that a low-flow state defined as a stroke volume index 65 years) population. The International receptor blockers (ARBs), diuretics, and statins Registry of Acute Aortic Dissection (IRAD) was were prescribed more frequently at hospital disestablished in 1996 with the idea to create a new charge following acute aortic dissection [1]; and understanding of this old disease including overall hospital mortality had improved for type demographics, presenting history, physical A aortic dissection but less so for type B aortic examination, imaging studies, and management. dissection (Table 20.1). The conception and implementation of the IRAD was inaugurated with the mission to IRAD database under the direction of Drs. Eric improve the care of dissection patients worldIsselbacher, Christoph Nienaber, and Kim Eagle wide, in the expectation that information from a represented a practical resource has contributed large registry will influence diagnostic and theraimmensely to the advancement in the manage- peutic management in the years to come. ment of aortic diseases with more than 80 peer-­ However, there are drawbacks as IRAD data reviewed publications. Since its inception, are collected retrospectively, there is no core laboIRAD has expanded to 43 active sites in 12 ratory for image analysis, and the tertiary referral countries in the USA, Europe, and Asia. A nature of the IRAD centres impairs its ability to recent trend analysis has shown that clinical be “representative of all patients with acute aortic presentation of acute aortic dissection had not dissection”. IRAD information comes from a changed. However, the use of computed tomog- referral hospital basis rather than a community raphy (CT) as a diagnostic modality increased; population basis, with subsequently some inherent potential for misleading statistics [2]. It would be helpful to have population-based information X. Yuan · C. A. Nienaber (*) Faculty of Medicine, Cardiology and Aortic Centre, in addition to knowledge of non-­consecutive disRoyal Brompton & Harefield NHS Trust and sected patients only. Nevertheless, IRAD has shed Cardiovascular Department, National Heart and Lung tremendous light on the “silent killer” and “great Institute, Imperial College London, London, UK masquerader” or aortic dissection. e-mail: [email protected]; [email protected]

© Springer-Verlag GmbH Austria, part of Springer Nature 2019 O. H. Stanger et al. (eds.), Surgical Management of Aortic Pathology, https://doi.org/10.1007/978-3-7091-4874-7_20

277

X. Yuan and C. A. Nienaber

278 Table 20.1  Twelve important publications from the IRAD consortium First author, year of publication (Ref.#) Hagan et al., 2000 [3] Mehta et al., 2002 [69]

Journal Journal of the American Medical Association Circulation

Moore et al., 2002 [27]

American Journal of Cardiology

Mehta et al., 2002 [16]

Circulation

Januzzi et al., 2004 [10]

Journal of the American College of Cardiology

Nienaber et al., 2004 [4]

Circulation

Tsai et al., 2007 [59]

New England Journal of Medicine

Pape et al., 2007 [11]

Circulation

Suzuki et al., 2009 [32]

Circulation

Suzuki et al., 2012 [60]

American Journal of Cardiology

Dean et al., 2014 [7]

American Journal of Cardiology

Pape et al., 2015 [1]

Journal of the American College of Cardiology

Major findings Describes manifestations of acute aortic dissection in the twenty-first century Develops a simple risk prediction tool for patients with type A dissection Demonstrates high sensitivity for all four imaging techniques: CT, MRI, TEE, and angiography Describes circadian and seasonal variations in the incidence of aortic dissection Identifies the importance of Marfan’s syndrome, bicuspid aortic valve, and aortic enlargement in young patients Demonstrates gender-­related differences in aortic dissection with regard to clinical characteristics, management, and outcomes of patients with aortic dissection Suggests partial thrombosis as a predictor of poor post-discharge prognosis after dissection Shows that many dissections occur before current dimensional criteria are achieved diameters Shows the high sensitivity of D-dimer elevation in detecting aortic dissection Demonstrates the benefit of beta blockers and lack of benefit of ACE inhibitors in the follow-up of patients after aortic dissection Focuses on cocaine as an important etiological factor for acute aortic dissection in the present era Demonstrates that there was no change in the presenting symptoms and physical findings of aortic dissection over two decades

ACE angiotensin-converting enzyme, CT computed tomography, IRAD International Registry of Acute Aortic Dissection, MRI magnetic resonance imaging, TEE transesophageal echocardiography

20.2 D  emographics at the Start of the Twenty-First Century In the IRAD, 67% of patients presented with type A dissection and 33% with type B; twothirds of patients were men, with an average age of 63  years. Many risk factors were related to aortic dissection, with hypertension (76.6%) being the most common [1, 3]. Twice as many hypertensive women than men older than 70  years experienced an acute aortic syndrome [4]. Iatrogenic dissection occurred in 3.3% of patients [5], cocaine use was recorded in 1.8% [6, 7], and 17% of patients with type A dissection had previous cardiac surgery [8]. Black patients who were younger (mean age 55 years)

suffered more frequently (52.4%) from type B dissection and more likely to have a history of cocaine abuse (12%), hypertension (89.7%), and diabetes (13.2%) [9]. Young patients 60

2.13 ± 0.35 2.33 ± 0.53 2.64 ± 0.48 3.27 ± 0.60 3.64 ± 0.40 4.82 ± 1.56

1.65 ± 0.24 1.62 ± 0.19 2.17 ± 0.33 1.96 ± 0.37 2.26 ± 0.35 2.91 ± 0.90

1.09 ± 0.21 1.23 ± 0.21 1.57 ± 0.20 1.89 ± 0.22 1.88 ± 0.19 2.82 ± 0.90

Ascending aorta, AA; aotic arch, AR; descending thoracic aorta, DA. Measurements were normalized for body surface area (cm2/m2).

Fig. 23.8  Age-related changes in the ascending (AA), arch (AR) and descending thoracic aorta. Mohiaddin RH, et al. J Comput Assist Tomog 1990;14(5):748–52

the elasticity of the vessel via its sheets of elastin, collagen and smooth muscle cells; and the outermost layer is the collagenous tunica adventitia containing the vasa vasorum and lymphatics. Over time, subjected to haemodynamic stresses, the ageing aorta dilates at a rate of approximately 0.9  mm per decade in men and 0.7 mm per decade in women [28]. However, this is an average measurement, and more detailed analysis via echo has shown that different segments of the proximal aorta dilate at different rates, with the ascending aorta and sinuses of Valsalva more significantly affected than the annulus (1.1 and 1 mm per decade versus 0.2 mm per decade, respectively) [29]. It is thought that the repeated pulsatile stress to which the aorta is exposed in its proximal ­portions leads to fragmentation of elastic tissues with resulting fibrotic tissue replacement [30]. This leads to increased arterial stiffness, decreased distensibility and transverse dilatation of the aorta. The entire thoracic aortic anatomy changes with age with a progressive uncoiling of the aortic arch (Fig. 23.8) [31, 32].

Whilst these age-related changes can be appreciated on imaging, their exact importance and clinical integration have not been fully elucidated, although multiple authors argue that aortic diameters should be measured against age-­ specific norms [33]. A distinction should be made between this process of arteriosclerosis and the atherosclerosis discussed below.

23.4 Acquired Aortic Diseases 23.4.1 Atherosclerotic Aneurysm An aortic aneurysm is defined as an increase in vessel diameter of >50% of the normal diameter for a given age and body surface area [3]. Aneurysm formation may occur in any part of the aorta. Many medical practitioners consider aneurysm formation to be a disease of the abdominal aorta, and, when it comes to those of an atherosclerotic aetiology, there is some truth in this. Fewer aneurysms occur in the thoracic aorta, and those that

23  Imaging of the Thoracic Aorta

do are marked by less lymphocytic infiltration, which often occur at a younger age and are more likely to have a genetic component. Thoracic aortic aneurysms occur most commonly in the ascending and descending aorta: 10% can be classed as thoracoabdominal; and 10% involve the aortic arch [34]. Most aneurysms are asymptomatic, and initial presentation may be as a medical emergency. Imaging is therefore key to diagnosis, assessment of aetiology, monitoring of progression and guide timing and nature of intervention. Thereafter, it provides non-invasive follow-up to identify complications or need for further intervention at the same site or remotely in the vascular tree. From a diagnostic perspective, TTE will readily identify aneurysmal root dilatation and may reveal arch abnormalities, but these are less likely to be atherosclerotic in origin. However, with the use of harmonic imaging, now the default on most echo machines, protruding arch plaques as small as 4 mm may be reliably visualised when an arch view of adequate quality can be obtained [35]. Descending thoracic aneurysms which may be atherosclerotic may be viewed in cross-section in a four-chamber view with echogenic matter clearly visible within the lumen, but this is far more likely to be an incidental finding in an asymptomatic patient but may be encountered in the assessment of a patient who has suffered an embolic event [36]. TTE is not the diagnostic modality of choice. TOE offers a far more extensive view of the aorta and can readily identify atherosclerotic lesions. These lesions are dynamic and often have superimposed mobile components [37]. These can be assessed in real time using ultrasound. Combine this with TOE’s ability for measuring the size of the aortic arch and descending aorta, and it becomes the modality of choice for visualising atherosclerotic lesions. However, it can be difficult to establish at which aortic level a lesion is without readily visible adjacent landmarks on an image and when image quality is poor, particularly likely when the aorta is tortuous. Therefore, other than investigating source of cardiac/vascular embolism in a known atherosclerotic aneurysm, it is better used as a peri- and

343

intraoperative imaging technique when intervention is required in a previously diagnosed atherosclerotic aneurysm. (For the role of echocardiography in non-atherosclerotic aneurysms, see section on heritable diseases.) Once an aortic aneurysm is suspected, CT or MRI is required to adequately visualise the entire aorta and identify the affected parts, remembering that whatever the aetiology, lesions may occur in tandem (Fig. 23.9). In an individual with a newly diagnosed atherosclerotic ascending aortic aneurysm, without confounding features, e.g. connective tissue disease, or suspected dissection, surgical intervention is advised in those who have a maximal aortic diameter of ≥55 mm [2]. Imaging not only provides this measurement but also is used to assess for coronary artery and aortic valve pathology. When the aneurysm has a smaller diameter, the role of imaging is in assessing rate of change. Cross-sectional imaging should be performed and the aneurysm morphology and any potential complications defined. When this is performed by CT, an ECG-gated protocol is the method of choice to minimise the effects of motion, particularly in the aortic root. Unenhanced imaging followed by contrast-enhanced images should be obtained to assess for acute aortic syndromes, which may complicate aneurysm formation, as described in the next section. CT images can then be reconstructed and manipulated to permit accurate measurements of even the most tortuous and complex of aneurysms and to plan intervention. The maximum diameter of the aneurysm perpendicular to the centreline of the vessel should be measured, using curved multiplanar reconstructions whenever possible [38]. Where this isn’t possible, thin, ≤3 mm, axial slices should be acquired. External diameters of an aneurysm should be measured as described earlier. However, where there is a marked burden of atheroma or thrombus within an aneurysm which alters the luminal diameter, the authors would recommend that a second ‘internal measurement’ is made and both given in the report. This approach needs to be agreed at a departmental level for consistency.

344

H. Dormand and R. H. Mohiaddin

Fig. 23.9  CMR volume spin-echo image (left) and volume-rendered contrast-enhanced MRA (right) of a diffusely atherosclerotic thoracic aorta with a large mainly

thrombosed aneurysm involving the aortic arch and proximal descending aorta

It should be recognised that CT is the imaging modality of choice for identifying calcification (Fig.  23.10). This often complicates atherosclerosis. When extensive this may be described as a ‘porcelain aorta’, necessitating modification of surgical technique. MRI may be used to identify and monitor thoracic aneurysms with a similar accuracy to CT and without radiation. It is useful for serial monitoring and may be performed without contrast and using 3D-SSFP sequences for aortography. Highly mobile intra-aneurysmal components may not be visualised, but it is more reliable in this respect than CT, but inferior to echo [39].

FDG-PET combined with CT can demonstrate atherosclerotic plaques and metabolically active plaques and is likely to have a role in the future, but its clinical applications have not yet been fully elucidated [40, 41]. Post-intervention, cross-sectional imaging is the mainstay of monitoring (Fig. 23.11). Where intervention is in the form of endovascular stent-­ graft implantation, imaging must be performed to assess for complications and must continue for life. Complications may take the form of endoleak, thrombosis, kinking or distortion of a graft (may occur slowly over time), aortic dissection or extravascular necrosis or perforation [42].

23  Imaging of the Thoracic Aorta

345

Fig. 23.10  Volume-rendered contrast-enhanced CTA of the thoracic and abdominal aorta showing large ascending aortic aneurysm and diffusely calcified aortic wall

Fig. 23.11  CTA of a malalignment of multiple thoracoabdominal covered aortic stents

The initial imaging, ideally by CT, should be performed before discharge. The most common complication at this stage is a leakage into the aneurysm, ‘endoleak’. This should be described by the imager in relation to its site of origin, e.g. proximal, middle or distal graft. Images should be obtained looking for limited apposition of the stent-graft to aortic walls, for stent-graft defects and for collateral vessel flow bypassing normal channels. Both axial slices and multiplanar reconstructions are useful in this setting. Endoleaks can be classified by their mechanism, but in the setting of the thoracic aorta, an attachment site leak (type I) is the most common mechanism. Type II, collateral vessel leak;

III, graft failure; IV, graft-wall porosity; and V, endotension are usually encountered in the abdominal aorta [43]. Delayed contrast imaging is important to detect endoleaks, which have variable flow rates. An unenhanced scan to identify calcification in the aneurysm sac is usually performed first. This is followed by a contrast-enhanced arteriogram and then further imaging in the post-contrast delayed phase. The exact parameters should be determined by the scanner used since technology in this area is rapidly changing and faster scans with lower radiation doses are possible in scanners with greater numbers of detectors and better dose modulation.

346

Contrast-enhanced CT may also identify graft thrombosis. The intraluminal thrombus will not enhance and may take a variety of forms. Mural thrombus within the graft may resolve or may progress leading to complete occlusion, and hence close follow-up is necessary with targeted CT to minimise radiation dose wherever possible. If an aneurysm has reached a large size, then when it decreases in diameter after stent-graft implantation, it will decrease in length as well, and this may result in graft kinking. This may result in graft migration and is a later phenomenon occurring months to years after implantation. Axial slices may miss this phenomenon, and MIP or multiplanar reconstructions are preferred. The rare complication of graft infection is difficult to appreciate on imaging since peri-­ prosthetic thickening may represent post-­implantation inflammation and clinical correlation is required. Infection at distal sites or septic emboli may provide assistance in the diagnosis. When suspected this is a medical emergency, and it is essential that the referrer and imager(s) discuss their concerns and review images jointly. MRI may be used for long-term surveillance but may be limited by artefacts, depending on stent composition. Nitinol stents are most MR-compatible and gadolinium-enhanced angiography can be used to demonstrate endoleaks. Other stent materials generally cause too much artefact. Follow-up is advised after 1  month, 6  months, 12  months and then yearly. If no endoleak after 2 years, this may be extended to every 2 years, in line with European guidance. Where surgical repair has been undertaken rather than endovascular repair, the same imaging principles apply, but MR and CT can be used more interchangeably since artefact is not a significant issue.

23.4.2 Acute Aortic Syndromes The term ‘acute aortic syndromes’ refers to several conditions, which are often medical emergencies and present with similar clinical symptoms. These may be defined by their specific aortic appearance, but essentially occur

H. Dormand and R. H. Mohiaddin

when a tear or ulcer permits blood to enter and disrupt the aortic media. They are further classified by their location. When they affect the ascending aorta, they are referred to as type A and, for the descending aorta, type B [2]. Although divided into their underlying pathologies in the sections below, these are diagnostic divisions made initially on the basis of imaging and then confirmed, or otherwise, at intervention or post mortem. These conditions carry a high mortality, and it is incumbent on the imager to provide a rapid, reliable test. Although CT, TOE and MRI all have something to offer, CT has emerged as the front runner because of its noninvasive nature allowing rapid vascular and extravascular coverage in all but the most unstable of patients with a sensitivity and specificity approaching 100%. Although individual studies have demonstrated benefits of different scanning protocols, the accepted approach is an unenhanced scan followed by a contrast-enhanced scan [44]. Where intraoperative imaging is available, or preoperative imaging, if the patient has not been unstable, TOE can be utilised. For long-term follow-up in which ionising radiation needs to be minimised, MRI has a clear advantage.

23.4.2.1 Intramural Haematoma (IMH) This is a diagnosis made by imaging. IMH is a haematoma within the aortic media without the presence of a false lumen. As a result it takes up a crescentic or circular-shaped thickening of the aortic wall, defined as >5 mm, in the absence of blood flow. The majority of cases occur in the descending thoracic aorta [45]. Cross-sectional imaging by either CT or MRI is recommended, and CT is often used first line. In this setting, an unenhanced scan is performed first looking for a high-attenuation area of the aorta with linear extension, taking up a circular or crescentic shape. This is followed by a contrast enhanced to assess for an intimal flap, which would alter the diagnosis. In IMH the aortic lumen is not normally compromised. CT offers sensitivities as high as 100% for IMH detection, but this is reduced in highly atherosclerotic aortas or those with previous dissection [44].

23  Imaging of the Thoracic Aorta

347

Fig. 23.12  T1W spin-echo image in a sagittal plane acquired in a patient with intramural haematoma at the initial presentation of severe back pain (left) and at 9  months follow-up (right). Note that the intramural

haematoma has intense signal intensity at the initial phase due to the release of methaemoglobin. There is near-­ complete resolution of the haematoma in the follow-up study

MRI may also be used to assess for IMH, and, when discovered incidentally or for serial monitoring, the age of the IMH episode can be estimated by the signal characteristics of the haematoma (Fig. 23.12). (Methaemoglobin, a byproduct of the early disintegration of haemoglobin in haematoma, has a short T1 relaxation time value and appears bright on T1-weighted spin echo.) Site of IMH, (A worse than B), maximum aortic diameter, maximum and changing wall thickness over time, recurrent pleural effusion (Fig. 23.13), organ ischaemia and an ‘ulcer-like projection’ secondary to the depth of local involvement are all prognostic markers. Imaging reports should highlight these features where relevant. More than one imaging test may be needed to make this challenging diagnosis [46, 47].

more common in the descending rather than ascending thoracic aorta (Fig. 23.14). With its ability to image the most unstable of patients and to delineate calcification, CT is the imaging of choice in the acute setting. On unenhanced CT, penetrating aortic ulceration will resemble IMH. However, after the administration of contrast, there will be transit of contrast into the aortic wall in an area of ulceration, often within a calcified atherosclerotic section. The calcification may be displaced beyond its normal intimal alignment. The role of FDG-PET/CT is monitoring of ulcers not requiring immediate intervention is being evaluated [48].

23.4.2.2 Penetrating Aortic Ulcer This is ulceration of an atherosclerotic plaque penetrating into the media. It may progress to other forms of acute aortic syndrome or aneurysm formation. Lesions may be multiple and are

23.4.2.3 Aortic Dissection This occurs when intramural bleeding via an intimal tear or flap leads to disruption of the aortic media. The aortic layers are disrupted, and blood may track back into the lumen via a second intimal flap, or there may be aortic rupture. The passage of blood results in the formation of a true and false lumen (TL and FL).

348

Fig. 23.13  Diastolic frame from complete cine acquisitions acquired in the oblique sagittal plane in a patient with a large intramural haematoma of the ascending aorta and dissection of the descending thoracic aorta (left). The

Fig. 23.14  Volume-rendered contrast-enhanced MRA of a diffusely atherosclerotic thoracic aorta with large arch aneurysms and multiple aortic ulcerations

H. Dormand and R. H. Mohiaddin

corresponding image acquired after a 9 months follow-up (right). Note the near complete resolution of ascending aortic haematoma

Complications depend up on the site and extent of dissection. Imaging can be used to assess the entire aorta, side branches, organ compromise, effusion, tamponade and valvular regurgitation or coronary compromise. Contrast-­enhanced CT is most commonly used to identify a dissection flap, measure aortic diameters and assess extent of dissection. Branch involvement can be assessed. The move towards ECG-gated studies has reduced motion artefact making it less likely that this is misdiagnosed as dissection and is of particular importance when assessing the aortic root. However, pulsation artefact, if a triple rule out protocol is used, may mimic root dissection [2]. Using CT, the vascular tree from neck to pelvis can be imaged rapidly, and it is possible to obtain an aortogram and coronary angiogram in one ECG-gated CT acquisition (Fig. 23.15). The FL usually has slower flow and hence may contain thrombi. Its diameter is often larger than the TL, and the convex surface of the intimal flap is usually towards the FL. The ‘cobweb’ sign refers to incompletely dissected media, which occur within the FL and are seen as linear areas of low attenuation on the scan. Efforts must be made to identify the TL and FL to advise the ‘interventionalist’ as to which organs are dependent on FL flow [49, 50]. MRI is as sensitive and specific for aortic dissection and has the advantage of being able to assess cardiac and valvular performance

23  Imaging of the Thoracic Aorta

349

Fig. 23.15  Volume-rendered contrast-enhanced CTA of an extensive aortic dissection involving the thoracic and abdominal aorta and extending cephalically into

innominate and right carotid arteries and caudally into the left common iliac artery

(Fig.  23.16). A combination of T1-weighted black blood images followed by SSFP-based cine images defines the anatomy. Phase-contrast imaging can then be used to assess flow in TL and FL as well as quantification of valvular and large vessel performance. T2-weighted black blood sequences can be used for follow-up and tissue characterisation. However, access to this technology is still limited, and it is not the ideal environment for a patient presenting acutely. Its role is therefore as a second-line test and for longer-term serial monitoring. In the most unstable of patients, there may only be an opportunity for a transthoracic echo probe to be placed on the chest before the clinical situation deteriorates. It is recommended that a

TTE be performed in all patients with a suspected acute aortic syndrome [2]. This has a sensitivity and specificity of up to 80% and 96% for ascending aortic involvement. It will also identify aortic valvular involvement and provide an assessment of myocardial performance. Addition of colour Doppler to images may demonstrate communication between the intima and media across a flap(s). However, as discussed earlier in this chapter, TTE is restricted by many patient features. Therefore additional imaging, e.g. by TOE, is required in all patients. In experienced hands, it has a sensitivity and specificity comparable to CT and MRI [20]. TOE offers flap visualisation and identification of flow and thrombus in TL and FL.  The TL usually expands in systole and

350

H. Dormand and R. H. Mohiaddin

Fig. 23.16  Chronic type A aortic dissection. Trans-axial T1W spin-echo images from aortic arch to aortic root (left panel). The false lumen (F) is partially thrombosed and

larger than the true lumen. The corresponding CE-MRA (right panel) showing the relation of the arch vessels to the aneurysmal segment

c­ollapses in diastole with forward flow during systole. A FL shows the reverse expansion pattern, and there may be reversed, delayed or absent flow. However, it must be remembered that the ‘blind spot’ of TOE is in the area of the brachiocephalic artery. As with all imaging in this context, when clinical suspicion is high, if one imaging modality is negative, then a second should be performed [3]. Long-term follow-up follows the same principles as for an aneurysm or post-aneurysm repair, unless an existing comorbidity or predisposition to further aortic complications is present, e.g. Marfan syndrome.

by an enhanced scan. In this setting, coverage of iliac and femoral vessels is needed to assist in planning intervention.

23.4.3 Contained Thoracic Aortic Rupture This occurs when an aneurysm ruptures and the ensuing haematoma is contained by peri-aortic structures including pleura and pericardium within the chest. Again, CT is the imaging modality of choice, using an unenhanced scan followed

23.4.4 Traumatic Aortic Injury Blunt force trauma to the thoracic aorta is usually secondary to a road traffic accident or fall from significant height (Fig. 23.17). This may result in sudden deceleration that translates into destructive forces maximised at relatively immobile portions of the aorta. Within the thorax these include the aortic root, the region of the left subclavian artery and the level of the diaphragm. It is the second most common cause of death after brain injury in blunt trauma patients. Formal classification systems exist with injury ranging from type I (intimal tear) to type IV (rupture). However, what is most important is a high index of clinical and ‘imaging’ suspicion based on mechanism of injury. Additional features pointing to the diagnosis include haemothorax, mediastinal haematoma/tamponade, aortic valve regurgitation or signs of myocardial contusion [51].

23  Imaging of the Thoracic Aorta

351

Fig. 23.17  CE-MRA in the LAO and RAO views in a patient with traumatic aneurysm (arrow) seen several years after a car accident

Where traumatic aortic injury is suspected, CT is the imaging modality of choice for speed and reproducibility [2]. It allows vascular and extravascular assessment in a non-invasive manner without patient movement being required. Beyond its diagnostic abilities, CT may help in the planning of emergency intervention. This may be supplemented with or substituted by TOE in the unstable patient in the intraoperative or perioperative setting. However, other organ damage cannot be assessed in this way.

23.4.5 Post-stenotic Aneurysms In isolated stenosis of a tricuspid aortic valve, the ascending aorta may dilate secondary to ­abnormal flow patterns in up to 60% patients. It may not be necessary to intervene on the aorta in many of these patients in the absence of connective tissue disorders since, following valve surgery, the rate of aortic expansion is minimal. This is in contrast to an atherosclerotic or connective tissue-related ascending aortic aneurysm [52]. The role of preoperative imaging is therefore to

352

define the aortic valve morphology and note whether there is aortic dilatation. Post-operatively, surveillance may be by whichever modality visualises the aorta and assesses aortic valve replacement function whilst complying with the international guideline recommendations of patient acceptance and minimal radiation dose.

23.4.6 Aortitis Aortitis refers to inflammation of the aortic wall. The causes may be a non-infectious inflammatory vasculitis or, uncommonly, infectioninduced inflammation. Takayasu arteritis is a rare, large vessel vasculitis, which most commonly affects the thoracic aorta and its branches. It is more common in young women. Presentation ranges from the incidental finding of a thoracic aortic aneurysm without active symptoms to fever and severe aortic regurgitation. Symptoms vary depending on the site and severity of aortic involvement. For example, upper extremity claudication may indicate supra-aortic vessel compromise, whereas peripheral embolisation may result from thrombus formation within the aortic lumen. Aneurysm formation and aortic stenosis are later consequences. Imaging of the entire aorta is needed. Echocardiography is rapid and non-invasive and may identify wall thickening. This needs to be complemented by use of cross-sectional imaging for better aortic coverage and branch vessel assessment. MRI as a nonionising radiation technique has the advantage of demonstrating vessel wall thickening and oedema using T2-weighted scans when the disease is active and hence is useful in demonstrating response to corticosteroid therapy and timing of intervention (Fig. 23.18). This can be supplemented by contrast-enhanced angiography (Fig.  23.19). At a later stage, lesions may become calcified, and this is better seen with CT. A more common vasculitis with thoracic aortic involvement in up to 18% of cases is giant cell arteritis (GCA). In contrast to Takayasu arteritis, this is a disease of the elderly. When the temporal

H. Dormand and R. H. Mohiaddin

arteries are involved, it is commonly referred to as temporal arteritis. However, if the aorta is involved, there may be aortic root and ascending aortic dilatation, dissection or rupture. Echocardiography, MR and CT can all diagnose extra-cardiac GCA and demonstrate aortic wall thickening and aneurysm formation but may not be good at tracking active disease [53]. PET scanning is less widely available but, when combined with contrast-enhanced CT, has shown an ability to identify early aortic and pulmonary lesions in Takayasu arteritis and track their response to therapy [54]. IN GCA PET scanning has shown subclinical aortic inflammation in patients. Intervention is best performed on patients in the quiescent phase of their disease. Other noninfectious causes which should be considered in cases of aortic thickening include ankylosing spondylitis, Reiter’s syndrome, Behcet’s disease and SLE [55]. Infectious causes include Staphylococcus, Mycobacteria, Salmonella and Treponema pallidum. HIV may be associated with aortic root dilatation, perhaps due to increasing vascular stiffness [56].

23.4.7 Aortic Tumours Primary tumours are very rare and may present with embolic phenomena. MRI with gadolinium is the modality of choice.

23.5 Congenital Aortic Diseases 23.5.1 Coarctation This describes a narrowed segment of the aorta, most frequently in the area of the ductus a­ rteriosus insertion, distal to the left subclavian artery. Although manifested by anything from a discrete stenosis with collateral formation to a long hypoplastic segment, it is regarded as a generalised arteriopathy [57]. It may occur in isolation or be associated with other abnormalities, most commonly bicuspid aortic valve disease, but also supra, sub- or valvular aortic stenosis, mitral ste-

23  Imaging of the Thoracic Aorta

353

Fig. 23.18 Takayasu’s aortitis: grossly thickened wall of the ascending aorta depicted on a trans-axial T1W image (upper panel). Oblique sagittal images of the thoracic aorta (lower panel) of the same patient using T1W (left) and STIR T2W images (right). Note the thickened aortic wall and the corresponding high signal on the STIR images consistent with wall oedema/ inflammation

nosis or complex congenital heart disease. An imager must investigate with an open mind! If a bicuspid aortic valve is seen on any imaging modality, a coarctation should be looked for, and vice versa. Coarctation may be readily seen on a transthoracic arch view and velocity in the proximal descending aorta measured, even when anatomy cannot be seen clearly. A significantly elevated velocity with poor images should prompt a second imaging modality. However, Doppler gradients are not useful for quantification of coarctation; a diastolic ‘tail’ may reflect significant coarctation or recoarctation. A standard TTE will though provide a comprehensive assessment of left ventricular structure and function, with left

ventricular hypertrophy in the absence of significant aortic stenosis or left ventricular outflow tract obstruction, demonstrating haemodynamic effects from a coarctation. Cross-sectional imaging by either MRI or CT is preferred for aortic assessment in adults (Fig.  23.20). Both demonstrate anatomy and ­collaterals, regardless of chest wall structure or body habitus. These are used for initial diagnosis and assessment and then for follow-up post-­ intervention, surgical or stenting. Intervention may be undertaken on the basis of a non-invasive pressure difference between the upper and lower limbs with upper limb hypertension, abnormal blood pressure response during exer-

354

Fig. 23.19  Takayasu’s aortitis: T1W spin-echo image in oblique sagittal plane (left) and CE-MRA (right). There is complete occlusion of the proximal left subclavian artery

Fig. 23.20  CE-MRA of severe aortic coarctation (arrow) with extensive arterial collaterals

H. Dormand and R. H. Mohiaddin

and moderate stenosis at the origin of the left common carotid artery

cise or significant LVH. However, independent of the pressure gradient, patients with >50% aortic narrowing relative to aortic diameter at the diaphragm should be considered for intervention. Hence, imaging is crucial [2]. Aortography is still regarded as the gold standard though. For MRI, a standard scanning protocol will incorporate scouts in three planes, long-axis and short-axis cine imaging of the heart and valvular assessment. To assess the aorta, cine imaging of the aorta will demonstrate anatomical and crude flow abnormalities and, supported by either 3D contrast-enhanced or unenhanced aortography, will demonstrate anatomy and collaterals. Velocity-encoded sequences should be used to demonstrate flow acceleration [58]. Analysis of a jet just beyond a coarctation, which demonstrates a diastolic tail, represents a significant narrowing. Additionally, collateral flow can be estimated by comparing through-plane flow immediately prior to a coarctation with that at the level of the diaphragm. The expected decrease in the latter is ≥10%. Any increase represents significant collateral flow [59]. CT will have better spatial resolution for small-calibre collaterals but is unable to provide this haemodynamic information.

23  Imaging of the Thoracic Aorta

355

Fig. 23.21  CE-MRA of aneurysm (arrow) at the site of patch repair of aortic coarctation (left) and postsurgical repair with excision of aneurysmal segment and interposition aortic graft placement (right)

Follow-up of these patients after intervention aims to identify complications such as recoarctation and aneurysm formation (Fig.  23.21). If stenting has taken place, then CMR may be affected by artefact, but haemodynamic assessment via velocity-encoded sequences will still be accurate. CT will be unhindered by artefact but unable to provide haemodynamic data; additionally, this is at the cost of exposure to ionising radiation, often in a young or relatively young population. Imaging intervals for re-evaluation depend on clinical status and post-intervention appearance, but full clinical evaluation and echo should be on a 2 yearly basis unless a clinical change is detected [57]. Often a composite approach utilising multimodality imaging is needed throughout a patient’s life, with close collaboration between imager, specialist physician in adult congenital heart disease and interventionalist or surgeon. Note: For supravalvular aortic stenosis, similar principles apply.

23.5.2 Bicuspid Aortic Valve (BAV) Our understanding of bicuspid aortic valve, once considered an isolated anomaly, has progressed significantly over recent years. It remains the most common congenital cardiac defect, usually resulting from fusion of the left and right coronary cusps. When secondary to right and left cusp fusion BAV is associated with both aortic coarctation and dilatation, other forms are being associated with dilatation, making a multimodality imaging approach essential in this population [60]. There is a heterogeneity of patterns and rate of aortic root and ascending aortic dilatation in this population. The arch is rarely involved, but maximum aortic diameter is often within the tubular section of the aorta (Fig.  23.22) rather than the aortic sinuses [61]. Dissection rates are also increased, irrespective of valve function, with growth rates of >5  mm/year and larger absolute diameters being markers of highest risk.

356

H. Dormand and R. H. Mohiaddin

Fig. 23.22  Bicuspid aortic valve: diastolic and systolic frames from the complete cine acquisition in the short-­ axis view of the aortic root (left panel). There is fusion between the left and right coronary cuspid and significant

stenosis of the aortic valve. The ascending aorta and the proximal aortic arch are dilated (right panel, top). Valve leaflets are restricted and there is in narrow stenotic jet in systole

When a BAV is identified by whatever means, initial imaging assessment should be by TTE to assess valve function, diameters of aortic root and ascending aorta. If these aortic regions cannot be adequately visualised by TTE, then MRI or CT is required (Class I ESC indication) [2]. Thereafter, serial follow-up is required and frequency determined by initial aortic and valvular appearance and family history. The authors would advocate that some form of c­ ross-­sectional imaging is used in all these patients to identify those with arch involvement and for greater reproducibility of serial measurement (MRI is preferable to CT whenever possible to minimise lifetime radiation exposure in this population). The American

guidelines also advocate evaluation of both root and ascending aorta [3]. The latest ESC guidelines recommend that if the aortic root or ascending aorta is >45 mm or is increasing at >3 mm/year measured by echocardiography, then annual measurement is an appropriate time interval. If, however, the absolute diameter is >50  mm or the rate of increase is >3 mm/year measured by echocardiography, then this requires a second imaging modality to be undertaken. Surgery is indicated when the maximum aortic diameter at any level is >55 mm or at the lower level of 50  mm in the presence of coarctation, hypertension, family history of dissection or rate of growth >3  mm/year. An even lower level of

23  Imaging of the Thoracic Aorta

357

>45  mm is suggested if a concomitant valve replacement is planned. The advent of 4D-MRI makes the future of this area of aortic imaging particularly exciting (Fig. 23.23). This time-resolved three-­dimensional phase contrast MRI has begun to unravel some of the haemodynamic mysteries of the aorta, and the future should be one of personalised aortic imaga

b

healthy volunteer

ing with time intervals determined by our knowledge of aetiology of disease and vascular dynamics. Elevated and asymmetric wall shear stress in the ascending aorta has been demonstrated in a subgroup of patients with BAV using this technique [62, 63], whilst others are demonstrating specific patterns of abnormality [64]. Recent work raises the possibility of 4D-MRI screening in apparently c

aorta size control

velocity [m/s] 1.0 0.5

d

RL-BAV

RN-BAV

flow jet

left-anterior view

0

velocity [m/s] 1.0 0.5

anterior view

0

Fig. 23.23  3D streamline visualisation of peak systolic blood flow in patients with BAV (c, d) in comparison with an aorta size-matched control subject (b) and a healthy volunteer (a). Note the presence of distinctly different 3D outflow flow jet patterns (black dashed arrows) in the ascending aorta (AAo) for patients (b) and (c). Bottom, 3D flow patterns in the left ventricular outflow tract (LVOT) and AAo distal to the aortic valve. Note the different systolic aortic valve outflow flow jet patterns (red indicating high veloci-

ties >1  m/s) and wall impingement zones that correspond to variable exertion of high wall shear forces between different valve groups (c, d) and aorta sizematched controls (b) and healthy volunteers (a). BAV indicates bicuspid aortic valve; RL, right and left coronary leaflet; and RN, right and noncoronary leaflet. Mahadevia R et  al. Bicuspid aortic cusp fusion morphology alters aortic 3D outflow patterns, wall shear stress and expression of aortopathy [101]

358

unaffected relatives and tailoring of therapy in those affected [65, 66]. Screening note: The European and American guidelines both give a nod to screening but do not stipulate this as a requirement for family members in most cases. The exception to this is in first-­degree relatives of patients with a BAV and premature thoracic aortic disease or familial thoracic aneurysm and dissection. These should be evaluated not just for a BAV but also for thoracic aortic disease in the absence of BAV, which can also occur [3, 67].

H. Dormand and R. H. Mohiaddin

Vascular rings are congenital vessel anomalies of the aortic arch. They are referred to as complete when they encircle both the trachea and oesophagus and incomplete (sling) when one or other is involved. Due to their position, they may result in compression of these structures and either respiratory or gastrointestinal symptoms. Complete forms include a double aortic arch (Fig. 23.24) or

a right-sided arch with an aberrant left subclavian artery and persistent ductus arteriosus. Incomplete forms or ‘slings’ are usually from a pulmonary artery, compression of the brachiocephalic artery or an aberrant right subclavian artery, as discussed in the section below [68]. In a double aortic arch, the ascending aorta divides into two arches anteriorly to the trachea and oesophagus. They rejoin to form a descending aorta posteriorly, thus creating a ring. Imaging with CT or MRI will readily demonstrate this arrangement. Careful attention should be paid to identifying the origins of head and neck vessels and any underdeveloped aortic arch segments. In a right-sided arch (see below) with an aberrant left subclavian artery and ductus arteriosus, the trachea and oesophagus end up being encircled by the right-sided arch, the base of the subclavian artery and the ductus arteriosus. Again this will be identifiable on cross-sectional imaging. The administration of contrast and careful 3D reconstruction are particularly helpful. Crosssectional imaging has overtaken previous strategies of echo, CXR and aortography [69].

Fig. 23.24  Chest X-ray (left). No left aortic knuckle visible and there is a prominent knuckle instead. CE-MRA reconstructed in coronal and trans-axial plane (right).

There is complete aortic ring. Dominant right aortic arch and atretic left arch, each gives the corresponding common carotid and subclavian arteries

23.5.3 Rings

23  Imaging of the Thoracic Aorta

In a pulmonary artery sling, the left pulmonary artery arises from the right and passes between the trachea and oesophagus as it heads to the left lung. In brachiocephalic compression, the artery arises more distally along the arch and compresses the trachea as it passes in front of it in its proximal course.

23.5.4 Arch Vessel Anomalies and Right-Sided Arch There are several aortic arch variants and vessel anomalies distinct from those above. The most common aortic arch pattern has separate origins for the brachiocephalic/innominate artery, left common carotid and left subclavian arteries, as described at the beginning of the chapter. The second most common is the misnamed ‘bovine arch’ in which the brachiocephalic and left common carotid have a common origin. In a smaller group of individuals, the left common carotid artery arises from the brachiocephalic artery itself just after its origin from the aortic arch, again some erroneously refer to this as a ‘bovine arch’. Between them, these arrangements account for about 93% of the population [70]. Other variants include the left vertebral artery arising from the arch or brachiocephalic trunk, aberrant right subclavian artery, which is rarely combined with a bicarotid trunk and origin of the thyroid arterial supply directly from the arch [71]. For the imager, these are best identified using cross-sectional imaging. Most are asymptomatic and picked up incidentally but may have significant influence of the planning of intervention. An aberrant right subclavian artery is the most significant. Where it occurs, it takes a retro-­oesophageal course in 80% patients and may cause dysphagia. In a percentage of patients, the origin of this artery from the descending thoracic aorta becomes aneurysmal and may be referred to as a Kommerell diverticulum. The adjacent descending thoracic aorta may become aneurysmal as well [72]. Whilst

359

the discussion around timing of intervention is ongoing, it is based on the factors of symptoms and aneurysm dimensions. Serial imaging is therefore required, and whilst CT has been used in most studies, MRI would be expected to provide comparable results. These arch abnormalities have been described in the setting of a left-sided aortic arch, which occurs in 99.5% of the population. In the remainder the arch is right-sided. This results from persistence of the embryonic right dorsal arch with involution of a section of the left arch. The resulting arch runs to the right of the trachea and crosses over the right main bronchus before reaching the descending aorta on the right of the vertebra and then may descend on either side of the vertebral column. This can be identified on TTE in centres with experience of scanning in a population with congenital heart disease. However, in most cases it is identified by, and imaged with, CT or MRI. There are two main types defined according to their vessels. In a right-sided arch with ‘mirrored vessels’, there is a left brachiocephalic trunk, right common carotid artery and a right subclavian artery. This form of right-sided arch is associated with complex congenital heart disease, in particular tetralogy of Fallot, in whom up to 25% may have a right-sided arch. In the other form, the left subclavian artery is aberrant and may form a Kommerell diverticulum as described above. A cervical arch occurs when the arch extends into the lower portion of the neck or upper mediastinum. Its significance relates to intervention or line positioning, in which inadvertent puncture can be deleterious [73].

23.6 Heritable Disease The principles of imaging in heritable disease are the same as in other areas of aortic imaging. However, in several of these conditions, it needs to be recognised that there is a systemic pathology, which may affect many organs, and imaging should be tailored accordingly.

360

23.6.1 Marfan Syndrome This is an autosomal dominant disorder of connective tissue structure and function. It is the most common and quintessential connective tissue disorder with vascular involvement, but the phenotypic overlap with the conditions below is considerable. It is diagnosed using the revised Ghent nosology [74]. The systems classically thought of as being involved, are cardiovascular, ocular and musculoskeletal. Cardiovascular manifestations affect the aorta, myocardium, valves or pulmonary arteries. Aortic dissection is the most common cause of death. Cardiovascular imaging is therefore fundamental to the diagnosis and lifelong monitoring of these patients [75]. In the absence of dissection, the criteria for aortic surgical intervention are based on aortic diameters on imaging. Both the American and European guidelines advocate surgery if the external diameter is >5.0 cm. The former advocate consideration of surgery at /=15 years of age. Am J Cardiol. 2012;110:1189–94. 7. Burman ED, Keegan J, Kilner PJ.  Aortic root measurement by cardiovascular magnetic resonance: specification of planes and lines of measurement and corresponding normal values. Circ Cardiovasc Imaging. 2008;1:104–13. 8. Tamborini G, Galli CA, Maltagliati A, Andreini D, Pontone G, Quaglia C, et  al. Comparison of feasibility and accuracy of transthoracic echocardiography versus computed tomography in patients with known ascending aortic aneurysm. Am J Cardiol. 2006;98:966–9. 9. La Canna G, Maisano F, De Michele L, Grimaldi A, Grassi F, Capritti E, et al. Determinants of the degree of functional aortic regurgitation in patients with anatomically normal aortic valve and ascending thoracic aorta aneurysm. Transoesophageal Doppler echocardiography study. Heart. 2009;95:130–6. 10. Quint LE, Liu PS, Booher AM, Watcharotone K, Myles JD.  Proximal thoracic aortic diameter measurements at CT: repeatability and reproducibility according to measurement method. Int J Cardiovasc Imaging. 2013;29:479–88. 11. Hayter RG, Rhea JT, Small A, Tafazoli FS, Novelline RA. Suspected aortic dissection and other aortic disReferences orders: multi-detector row CT in 373 cases in the emergency setting. Radiology. 2006;238:841–52. 1. Evangelista A, Flachskampf F, Lancellotti P, 12. Francois CJ, Tuite D, Deshpande V, Jerecic R, Weale P, Carr JC.  Unenhanced MR angiography of the Badano L, Aguilar R, Monaghan M, et al. European thoracic aorta: initial clinical evaluation. AJR Am J Association of Echocardiography for standardizaRoentgenol. 2008;190:902–6. tion of performance, digital storage and reporting of echocardiographic studies. Eur J Echocardiogr. 13. Teixido-Tura G, Redheuil A, Rodriguez-Palomares J, Gutierrez L, Sanchez V, Forteza A, Lima JA, Garcia-­ 2008;9:438–48. Dorado D, Evangelista A.  Aortic biomechanics by 2. Goldstein SA, Evangelista A, Abbara S, Arai A, magnetic resonance: early markers of aortic disease Asch FM, Badano LP, et  al. Multimodality Imaging in Marfan syndrome regardless of aortic dilatation? of Diseases of the Thoracic Aorta in Adults: From Int J Cardiol. 2014;171:56–61. the American Society of Echocardiography and the 14. Evangelista A.  Imaging aortic aneurismal disease. European Association of Cardiovascular Imaging: Heart. 2014;100:909–15. Endorsed by the Society of Cardiovascular Computed 15. Hagan PG, Nienaber CA, Isselbacher EM, Tomography and Society for Cardiovascular Magnetic Bruckman D, Karavite DJ, Russman PL, et  al. The Resonance. J Am Soc Echocardiogr. 2015;28:119–82. International Registry of Acute Aortic Dissection 3. Li Y, Fan Z, Xu L, Yang L, Xin H, Zhang N, et  al. (IRAD): new insights into an old disease. JAMA. Prospective ECG-gated 320-row CT angiography of 2000;283:897–903. the whole aorta and coronary arteries. Eur Radiol. 16. Nienaber CA, von Kodolitsch Y, Nicolas V, Siglow V, 2012;22:2432–40. Piepho A, Brockhoff C, et  al. The diagnosis of tho 4. Lin FY, Devereux RB, Roman MJ, Meng J, Jow VM, racic aortic dissection by noninvasive imaging proceJacobs A, et  al. Assessment of the thoracic aorta by dures. N Engl J Med. 1993;328:1–9. multidetector computed tomography: age- and sex-­ 17. Nienaber CA, Kische S, Skriabina V, Ince specific reference values in adults without evident H.  Noninvasive imaging approaches to evaluate the cardiovascular disease. J Cardiovasc Comput Tomogr. patient with known or suspected aortic disease. Circ 2008;2:298–308. Cardiovasc Imaging. 2009;2:499–506. 5. Evangelista A, Garot J. In: Zamorano J, editor. Role of magnetic resonance imaging in aortic diseases. 18. Evangelista A, Carro A, Moral S, Teixidó-Turà G, Rodríguez-Palomares JF, Cuéllar H, García-Dorado ESC textbook of non-invasive of imaging techniques. D. Imaging modalities for the early diagnosis of acute London: Springer; 2010. aortic síndrome. Nat Rev Cardiol. 2013;10:477–86. 6. Devereux RB, de Simone G, Arnett DK, Best LG, 19. Shiga T, Wajima Z, Apfel CC, Inoue T, Ohe Boerwinkle E, Howard BV, et al. Normal limits in rela- Y.  Diagnostic accuracy of transesophageal echocartion to age, body size and gender of two-dimensional

24  Cardiovascular Imaging in Aortic Diseases: Multimodality Approach in Clinical Practice diography, helical computed tomography, and magnetic resonance imaging for suspected thoracic aortic dissection: systematic review and meta-­analysis. Arch Intern Med. 2006;166:1350–6. 20. Cecconi M, Chirillo F, Costantini C, Iacobone G, Lopez E, Zanoli R, et  al. The role of transthoracic echocardiography in the diagnosis and management of acute type A aortic syndrome. Am Heart J. 2012;163:112–8. 21. Evangelista A, Avegliano G, Aguilar R, Cuellar H, Igual A, Gonzalez-Alujas T, et  al. Impact of contrast-­enhanced echocardiography on the diagnostic algorithm of acute aortic dissection. Eur Heart J. 2010;31:472–9. 22. Evangelista A, Garcia-del-Castillo H, Gonzalez-­ Alujas T, Dominguez-Oronoz R, Salas A, Permanyer-­ Miralda G, et  al. Diagnosis of ascending aortic dissection by transesophageal echocardiography: utility of M-mode in recognizing artifacts. J Am Coll Cardiol. 1996;27:102–7. 23. Sommer T, Fehske W, Holzknecht N, Smekal AV, Keller E, Lutterbey G, et al. Aortic dissection: a comparative study of diagnosis with spiral CT, multiplanar transesophageal echocardiography, and MR imaging. Radiology. 1996;199:347–52. 24. Evangelista A, Aguilar R, Cuellar H, Thomas M, Laynez A, Rodriguez-Palomares J, et  al. Usefulness of real-time three-dimensional transoesophageal echocardiography in the assessment of chronic aortic dissection. Eur J Echocardiogr. 2011;12:272–7. 25. Rocchi G, Lofiego C, Biagini E, Piva T, Bracchetti G, Lovato L, et  al. Transesophageal echocardiography-­ guided algorithm for stent-graft implantation in aortic dissection. J Vasc Surg. 2004;40:880–5. 26. Chung JH, Ghoshhajra BB, Rojas CA, Dave BR, Abbara S.  CT angiography of the thoracic aorta. Radiol Clin N Am. 2010;48:249–64. 27. Zeman RK, Berman PM, Silverman PM, Davros WJ, Cooper C, Kladakis AO, et al. Diagnosis of aortic dissection: value of helical CT with multiplanar reformation and three-dimensional rendering. AJR Am J Roentgenol. 1995;164:1375–80. 28. Sebastia C, Pallisa E, Quiroga S, Alvarez-­Castells A, Dominguez R, Evangelista A. Aortic dissection: diagnosis and follow-up with helical CT. Radiographics. 1999;19:45–60. 29. Chang JM, Friese K, Caputo GR, Kondo C, Higgins CB. MR measurement of blood flow in the true and false channel in chronic aortic dissection. J Comput Assist Tomogr. 1991;15:418–23.

383

30. Nienaber CA, Spielmann RP, von Kodolitsch Y, Siglow V, Piepho A, Jaup T, et al. Diagnosis of thoracic aortic dissection. Magnetic resonance imaging versus transesophageal echocardiography. Circulation. 1992;85:434–47. 31. Evangelista A, Salas A, Ribera A, Ferreira-Gonzalez I, Cuellar H, Pineda V, et al. Long-term outcome of aortic dissection with patent false lumen: predictive role of entry tear size and location. Circulation. 2012;125:3133–41. 32. Movsowitz HD, Levine RA, Hilgenberg AD, Isselbacher EM.  Transesophageal echocardiographic description of the mechanisms of aortic regurgitation in acute type A aortic dissection: implications for aortic valve repair. J Am Coll Cardiol. 2000;36:884–90. 33. Evangelista A, Dominguez R, Sebastia C, Salas A, Permanyer-Miralda G, Avegliano G, et al. Long-term follow-up of aortic intramural hematoma: predictors of outcome. Circulation. 2003;108:583–9. 34. Evangelista A, Dominguez R, Sebastia C, Salas A, Permanyer-Miralda G, Avegliano G, et  al. Prognostic value of clinical and morphologic findings in short-­ term evolution of aortic intramural haematoma. Therapeutic implications. Eur Heart J. 2004;25:81–7. 35. Coady MA, Rizzo JA, Hammond GL, Pierce JG, Kopf GS, Elefteriades JA. Penetrating ulcer of the thoracic aorta: what is it? How do we recognize it? How do we manage it? J Vasc Surg. 1998;27:1006–5. 36. Quint LE, Williams DM, Francis IR, Monaghan HM, Sonnad SS, Patel S, et  al. Ulcerlike lesions of the aorta: imaging features and natural history. Radiology. 2001;218:719–23. 37. Tsai TT, Evangelista A, Nienaber CA, Myrmel T, Meinhardt G, Cooper JV, et al. Partial thrombosis of the false lumen in patients with acute type B aortic dissection. N Engl J Med. 2007;357:349–59. 38. Pitcher A, Cassar TE, Leeson P, Francis JM, Blair E, Wordsworth PB, et al. Aortic dissection: visualisation of aortic blood flow and quantification of wall shear stress using time-resolved, 3D phase-contrast MRI. J Cardiovasc Magn Reson. 2011;13:1–2. 39. Weber TF, Ganten MK, Bockler D, Geisbusch P, Kopp-Schneider A, Kauczor HU, et  al. Assessment of thoracic aortic conformational changes by four-­ dimensional computed tomography angiography in patients with chronic aortic dissection type b. Eur Radiol. 2009;19:245–53.

Multidetector Computed Tomography of the Aorta

25

Alistair C. Lindsay, Arjun Nair, and Michael B. Rubens

25.1 Technical Considerations The CT angiography (CTA) technique chosen to image the thoracic aorta should always be tailored to the diagnostic information being sought and is dependent on a number of clinical, technical and patient-based factors (Table  25.1). The main technical considerations in CT protocol selection are discussed below, and situation-specific protocols are described in Table  25.2 and discussed in later sections on different aortic pathologies.

25.1.1 Scanner Technology and ECG Synchronisation The introduction of multidetector row CT (MDCT) systems in the late 1990s afforded the opportunity for considerably greater acquisition speeds, longitudinal (z) axis coverage and spatial and temporal resolution, all of which improved further as the number of detector rows with each successive generation of CT scanner increased [1–4]. These improvements over helical single-­slice CT are parA. C. Lindsay Department of Cardiology, Royal Brompton Hospital, London, UK A. Nair · M. B. Rubens (*) Department of Radiology, Royal Brompton Hospital, London, UK e-mail: [email protected]

Table 25.1  Factors that influence the choice of CT scanning technique Patient factors Body mass index Heart rate Heart rhythm (beat-to-beat variability) Excessive respiratory motion Scan factors Number of detector rows Number of X-ray tubes (single- vs. dual-source) CT gantry speed CT reconstruction technique (e.g. statistical iterative reconstruction) Intravenous iodinated contrast: volume, phase, timing of injection, scan initiation (bolus tracking or test bolus) Clinical factors (see also Table 25.2) Suspected diagnosis (e.g. aortic dissection versus acute coronary syndrome) Acute versus stable presentations Pre-operative workup (e.g. aortic root replacement versus trascatheter aortic valve implantation)

ticularly advantageous to aortic imaging; the entire thoracoabdominal aorta can now be scanned in as little as 10 s with a 128-detector row scanner [5]. Narrow detector collimation provides the option of thinner sections (for high-resolution detail) or thicker sections (for quick review and with less noise) from a single acquisition while simultaneously generating an isotropic volumetric dataset [1]. In this way, accurate depiction of multiple ­facets of aortic pathology (e.g. intimal tears, fenestrations and side-branch involvement in aortic dissection) can be complemented by smooth ­

© Springer-Verlag GmbH Austria, part of Springer Nature 2019 O. H. Stanger et al. (eds.), Surgical Management of Aortic Pathology, https://doi.org/10.1007/978-3-7091-4874-7_25

385

A. C. Lindsay et al.

386 Table 25.2  Suggested protocols for common aortic CT indications Clinical indication Acute aortic syndromes

Acquisition Non-contrast, top of arch to groin Gated thoracic aorta post-contrast (“step-and-­ shoot” prospective gating if available) If extension of dissection above arch/ below heart: angiogram from base of skull to lesser trochantera

Pre-operative evaluation: aortic root replacement

Non-contrast gated thoracic aorta (for calcium scoring)b

Transcatheter aortic valve implantation planning

Gated thoracic aorta post-contrast As for aortic valve replacement but with the addition of a whole body angiogram

IV contrast injection protocol Reconstructions Gated thoracic aorta: Non-­ contrast 3 mm thickness/3 mm increment, soft-tissue algorithm

Gated scan: Test bolus, dual-phase injection (e.g. lopromide, 370 60 mL, then saline flush 40 mL, both at 6 mL/s) Whole body scan: Test bolus, dual-phase injection (e.g. lopromide 370 60 mL, then saline flush 40 mL, both at 4 mL/s)

Post-contrast: Thinnest available (0.5–0.7 mm, overlapping increment desirable) 60–70% R-R interval reconstructionsc Non-contrast and post-contrast whole body scans: 5 mm thickness/5 mm increment, soft-tissue algorithm Post-contrast whole-body scans only: Thinnest available (0.5–0.7 mm, overlapping increment desirable), soft-tissue algorithm

These protocols can be adjusted according to available technology and local radiation protection rules We advocate a separate scan (rather than combining this scan with the gated thoracic aorta scan) for a suspected extensive aortic dissection as image quality is improved in our experience b The method with the least dose is recommended (e.g. high-pitch spiral acquisition) c Further cardiac cycle phases can be provided if available; these can provide additional information when opportunistically evaluating the coronary arteries a

two- (2D), three- (3D) and even four-dimensional (4D) post-processing. The shorter acquisition time of MDCT systems has also increased temporal resolution. At the same time, the advent of ECG-gated synchronisation has meant that cardiac motion can be “frozen” within a particular phase of the cardiac cycle, to minimise or even eliminate motion artefact. In this way, unhampered assessment of the aortic root and the coronary arteries can be performed. Retrospective gating, as the name implies, involves acquiring images across the whole cardiac cycle and then retrospectively assigning phases of the cardiac cycle to post-processed datasets. Although this technique has demonstrated improved thoracic aorta visualisation [6], it is not routinely recommended due to its high radiation dose. Prospective ECG gating, where acquisition is only performed during ­diastole (typically at 70% of the R-R interval),

has been shown to substantially reduce radiation dose in both coronary [7] and aortic [8–10] CTA. In addition to dose reduction, a sequential prospective ECG-­ gating mode (the so-called “step-and-shoot” mode) has demonstrated higher aortic attenuation values, with no detriment to image quality [8]. Recent CT innovations, such as high-pitch dual-source CT [11] and wide-detector 320-row MDCT [12, 13], have harnessed the advantages of improved temporal resolution and prospective ECG synchronisation to deliver helical prospectively gated scanning in one heartbeat, allowing temporal homogeneity of the acquired dataset with an even greater dose reduction. The implication for aortic imaging is that prospective ECG-synchronised imaging of the thoracic aorta to minimise motion artefacts is widely available and may also allow simultaneous diagnostic visualisation of coronary arteries [14].

25  Multidetector Computed Tomography of the Aorta

387

Furthermore, for patients presenting to the emergency department with acute chest pain, the desire to exclude flow-limiting coronary artery disease, pulmonary embolus and acute aortic syndromes with a single investigation has prompted the evolution of “triple rule-out” CT protocols [15]. Such protocols may have a positive impact on patient management and cost-­ saving potential [16–18] when used in the appropriate patient population, but high-quality imaging of the thoracic aorta when there is a high index of suspicion of aortic pathology may still require a dedicated rather than “catch-all” imaging protocol.

­plateau [20] with a peak enhancement of greater than 200 Hounsfield units. Numerous patient factors (e.g. patient size, cardiac output, venous access), scanning factors (scan duration, direction, tube potential and tube current, timing of contrast) and contrast medium factors (volume, concentration, rate of injection, number of injection phases) contribute to the enhancement profile; the reader is directed to the article by Bae for an in-depth discussion on this topic [21]. Although some of these factors can be modulated on a patient-specific basis, most institutions create standardised contrast injection protocols for aortic imaging, which are adapted from manufacturer-specified scanning protocols. It is worth noting that reducing tube potential 25.1.2 Radiation Dose Reduction (Box 25.1) as a dose-reduction technique may simultaneously also provide a secondary benefit In addition to prospective ECG gating, a variety of reducing the contrast volume required, a of other measures to reduce dose are increasingly desirable benefit in patients with acute aortic employed (Box 25.1). Radiation dose should pathology who are commonly at risk of renal always be kept as low as is reasonably achievable dysfunction, circulatory overload and contrast-­ (ALARA) but not at the expense of compromis- induced nephropathy [22]. For instance, ing the diagnostic quality of an examination. Nakayama et  al. demonstrated that a low Furthermore, acquisition protocols should always kilovoltage setting (90  kVp, as opposed to the take advantage of capabilities introduced on the standard 120  kVp) with a standardised iodine latest scanners to help achieve this balance. For delivery rate of 300  mg  I/mL could be used to instance, thoracic aortic CT using reduced tube lower the volume of contrast administered potential at 70 kVp, and a high-pitch acquisition (40  mL as opposed to 100  mL) for patients mode that obviated the need for ECG gating, has weighing less than 70  kg, with acceptable recently been explored. With the use of a second-­ interobserver agreement for image quality [23]. generation method of iterative reconstruction, The penalty for such a reduction in tube potential mean radiation dose could be reduced by 85% is usually an increase in image noise, but recent without compromising diagnostic quality [19]. experience with dual-energy CT scanning using dual-source systems suggests that simultaneous acquisition of low- and high-kVp aortic datasets 25.1.3 Contrast Enhancement using smaller contrast volumes (thereby providing a supplementary less noisy dataset as a The desired optimal contrast enhancement profile backup) is feasible in nonobese patients [22]. for CT angiography is a long and consistent Box 25.1  Dose-reduction strategies in aortic CT

25.1.4 Post-processing

ECG-controlled tube current modulation (“padding”) Prospective ECG-gating (“step-and-shoot”) High-pitch spiral acquisition Weight-based reduction of tube potential Anatomical-based tube current modulation Statistical iterative reconstruction

Review of the transverse (axial) image dataset remains the initial assessment of choice; besides allowing evaluation of aortic enhancement and the presence of any pulsation artefacts, it also allows assessment of coexistent thoracoabdominal

A. C. Lindsay et al.

388

pathology. However, 2D and 3D post-processing techniques that complement the axial dataset permit (a) quicker visualisation of structures in a manner more familiar to the physician and (b) more accurate pre-surgical and follow-up measurement and branch vessel assessment. Two-dimensional techniques include maximum intensity projection (MIP) and multiplanar reconstructions (MPR) [24]. MIP images increase the conspicuity of structures of highest attenuation and as such are especially useful for depicting the aorta and branch vessels [25]. MPR images permit orthogonal and non-­ orthogonal visualisation and can be used to generate curved planar reformats to extract the contour of the aorta or branch vessels of interest. 3D aortic visualisation now largely relies on volume rendering (VR), a method in which an attenuation-based histogram classification is applied to the entire CT dataset, to map CT attenuation values to opacity, brightness and colour. In this way, structures of different densities can be selectively concealed or revealed [26]. Bone removal techniques can be applied to VR images to reveal cardiac and aortic structures obscured by the thoracic cage and shoulder girdle. Volumetric cine reconstructions (4D imaging) which visualise the cardiac and aortic structures in motion can also be performed when retrospective ECG gating has been used for acquisition, but the penalty of the high radiation dose does not justify this technique as routine. Fig. 25.1 (a) Sagittal MIP reconstruction of a normal undilated aorta. (b) Coronal view of the normal aortic root, showing valve annulus, sinus of Valsalva and sinotubular junction

a

b

25.2 Acquired Abnormalities of the Aorta 25.2.1 Normal Anatomy The entire aorta—from the aortic valve annulus to the iliac bifurcation and beyond—can be clearly imaged and its dimensions accurately measured using MDCT (Fig.  25.1). Although exact measurements have varied between studies [27], an ascending aortic diameter of greater than or equal to 40 mm is generally accepted as being abnormal (Table  25.3). A variety of conditions can lead to abnormal dilatation of the aorta—and in particular the aortic root—the most important of which are discussed in more detail below. A comprehensive list of conditions can be found in Table  25.4. Frequently, the finding of a dilated aortic root will be incidental on a CT scan performed for another indication and will often be due to common pathologies such as long-­ standing hypertension or post-stenotic dilatation due to severe aortic valve disease. However, for those cases with no clear cause, a thorough knowledge of the many potential causes of root dilatation is essential.

25.2.2 Acute Aortic Syndromes Acute aortic syndrome (AAS) is an umbrella term encompassing the entities of aortic dissection, penetrating atherosclerotic ulcer

25  Multidetector Computed Tomography of the Aorta

(PAU) and intramural haematoma (IMH) (28). Except for factors such as genetic predisposition (e.g. Marfan syndrome), these entities may be indistinguishable clinically. Contemporary classification of non-traumatic aortic dissection recognises five variants, namely, classic aortic dissection, IMH, limited dissection without haematoma, PAU and iatrogenic dissection [29–31]. Furthermore, it has been argued that IMH is better thought of as an imaging finding indicating an Table 25.3  Normal adult thoracic diameters (from [31])

Thoracic aorta Root (female) Root (male) Ascending Mid-­ descending (female) Mid-­ descending (male) Diaphragmatic (female) Diaphragmatic (male)

Range of reported mean (cm) 3.50–3.72 3.63–3.91 2.86 2.45–2.64

Reported SD (cm) 0.38 0.38 NA 0.31

Assessment method CT CT CXR CT

2.39–2.98

0.31

CT

2.40–2.44

0.32

CT

2.43–2.69

0.27– 0.40

CT, arteriography

389

acute process, while PAU and aortic dissection are manifestations of a disease process [5]; for simplicity, we have described the three terms under the umbrella of AAS here.

25.2.3 Aortic Dissection Aortic dissection occurs when diseased aortic media are disrupted, resulting in bleeding within and along the wall of the aorta and then separation of its layers [28]. CT evaluation of aortic dissection should address identification and characterisation of the type and extent of the lesion, the primary intimal tear (when present in a classic dissection), evidence of rupture or leakage and complications. A classic aortic dissection on contrast-­enhanced CT manifests as a true and false lumen separated by an intimal tear (Fig. 25.2). The true lumen can usually be identified by its continuity with an intact portion of the thoracic aorta. When the distinction between false and true lumen is difficult, the presence of certain ancillary signs such as the “cobweb” sign [32], “beak” sign and the larger cross-sectional area of the false lumen may be of value [33]. An atypical appearance due to circumferential

Table 25.4  Conditions that cause dilation of the aortic root Acquired Hypertension Atherosclerosis Annuloaortic ectasia Trauma Post-stenotic dilatation Aortic dissection Iatrogenic

Inflammatory/infective Takayasu’s arteritis Giant cell arteritis Ankylosing spondylitis Syphilis Tuberculosis HIV Reiter’s syndrome Acute anterior uveitis Psoriatic arthritis Juvenile rheumatoid arthritis Kawasaki disease Salmonella Aspergillosis Wiskott-Aldrich syndrome Behcet’s syndrome Relapsing polychondritis Idiopathic aortitis Sarcoidosis Bacterial aortitis

Congenital Marfan syndrome Ehlers-Danlos syndrome IV Coarctation of the aorta Bicuspid aortic valve Turner syndrome Familial aortic aneurysm Tuberous sclerosis Fabry’s disease Adult polycystic kidney disease Osteogenesis imperfecta Homocystinuria Noonan syndrome

A. C. Lindsay et al.

390

a

c

b

Fig. 25.2  Type A aortic dissection extending from the aortic root (a) to the descending aorta. Lack of contrast enhancement of the false lumen within the descending thoracic aorta may imply that the false lumen is undergoing thrombosis (a, b). A coronal MIP reconstruction

(c) elegantly demonstrates the termination of the dissection at the level of the renal arteries, with both renal arteries arising from the true lumen; such a finding is first confirmed on axial images

d­ issection along the intimal flap, with subsequent intimointimal intussusception, may sometimes be seen, giving rise to a “windsock” appearance in the central aorta [34, 35]. Limited dissection, or limited intimal tear, is an under-recognised form of dissection, which is often only suspected on high-quality, thin-slice CT acquisition by the presence of an abnormally bulging aortic contour (Fig.  25.3). This abnormality occurs mostly in the ascending aorta and

can be associated with aortic dilatation, especially in the context of Marfan syndrome [5, 29]. The extent of dissection is usually described using the Stanford classification, based on the presence (type A) or absence (type B) of a dissection flap in the ascending aorta [36]. Although there is no consensus on the best classification system [31], the Stanford classification has generally superseded the earlier DeBakey classification, chiefly because it has implications for

25  Multidetector Computed Tomography of the Aorta

a

b

391

c

Fig. 25.3 Acute aortic syndrome. The patient had a dilated thoracic aorta, optimally visualised on the sagittal oblique 8 mm MIP (a) and 3D volume-rendered (b) images. However, evaluation of the axial images is still

crucial: there is a bizarrely shaped focus of low density in the descending aorta thought to represent an intramural haematoma or the false lumen of an aortic dissection (c)

surgical versus non-surgical management, for type A and B dissections, respectively. The distal extent of the dissection should also be evaluated, and recanalisation and fenestration points should be described, because they may guide the choice and location of surgical repair. The assessment of branch vessel involvement in dissection is also crucial in this regard. As described earlier, the ability of modern CT scanning techniques to simultaneously assess the coronary arteries and depict the aortic root provides information as to whether reimplantation of the coronary arteries or root replacement is also required. The dissection may also cause either static or dynamic obstruction of a branch thoracoabdominal aortic vessel, as described by Williams et  al. [37]. Distinguishing these types of obstruction is also important as their management differs; static obstruction requires endovascular stenting, while dynamic obstruction requires fenestration of the prolapsing intimal flap that is causing the obstruction. Local complications of aortic dissection that should be sought on CT include evidence of rupture, especially with type A dissections, such as the presence of haemomediastinum or haemopericardium. Evidence of organ perfusion deficits, especially bowel or renal ischaemia, may not be immediately apparent on the initial aortic phase imaging and so necessitate delayed phase imaging.

Yoshida et  al. [38] demonstrated a 100% diagnostic accuracy rate of MDCT for type A aortic dissection, using corroboration with findings at surgery. Furthermore, the same investigators proved that not only did MDCT demonstrate the highest sensitivity for depicting arch branch vessel involvement (95%), followed by primary intimal tear (83%) and pericardial effusion (82%), but importantly that it had 100% specificity for all of these features. Given this high accuracy and comprehensive assessment of MDCT when performed to a high standard, it should be emphasised that the imaging of acute aortic syndromes is best performed with the supervising radiologist in the scan control room, reviewing the need for additional phases or coverage as necessary, while of course always being mindful of radiation dose. Several pitfalls that can lead to false-positive interpretations of aortic dissection on CT should also be recognised. These include (1) circumferential and perpendicular motion artefact, especially in non-gated aortic CT, usually at the left anterior and right posterior aspect of the aortic circumference [39] (Fig.  25.4), (2) prominent fluid in the anterior superior pericardial recess, and (3) mural thrombus in the atherosclerotic aorta. The latter can usually be distinguished by an irregular internal border, in contrast to the smooth internal border of a dissection.

A. C. Lindsay et al.

392

a

b

Fig. 25.4  Motion artefact mimicking an ascending aortic dissection on a non-ECG-gated contrast-enhanced scan (a). When the scan was repeated with gating (b), a dissection could be confidently excluded

25.2.4 Penetrating Atherosclerotic Ulcer (PAU) PAUs are ulcerations of atheromatous plaque that have eroded into the media from the internal elastic lamina [40] and are most often found in elderly patients with atherosclerosis. At CT, these lesions normally appear as outpouchings of contrast material that are in continuity with the aortic lumen, with or without focal intramural haematoma (see below), and often in association with a thickened aortic wall that may enhance [41, 42] (Fig.  25.5). Fissuring or ulceration confined to the intimal layer of atheromatous plaques is a common finding in the asymptomatic patient that can mimic PAU; as such, the clinical status of a patient with an incidental finding of such lesions should be ascertained to avoid misinterpretation. The risk of rupture from PAU is significant and is in the region of 20–50%. However, it should also be considered that surgical repair for ruptured PAU is also more complex as a consequence of the more extensively diseased aortic wall in these elderly patients [43].

Fig. 25.5  Penetrating atherosclerotic ulcer in a markedly atherosclerotic aorta. Sagittal CT image demonstrates an outpouching of contrast with no intimal flap visible

25  Multidetector Computed Tomography of the Aorta

25.2.5 Intramural Haematoma (IMH) Intramural haematoma is the development of haematoma within the media, usually without an intimal tear [44]. The exact relationship between IMH and aortic dissection has not been elucidated. Unenhanced CT is useful for depicting IMH [45], which usually manifests as a crescentic focus of high attenuation paralleling the aortic wall (in contrast to the false lumen of an aortic dissection), with or without displacement of intimal calcification (Fig. 25.6). An IMH should not enhance on contrast-enhanced CT and may in fact be obscured by a dense contrast bolus. For this reason, an initial non-gated, unenhanced evaluation of the thoracic aorta in patients with a high suspicion of AAS can be extremely helpful (Table 25.2). It should also be remembered that IMH can occur even when the initial site of disruption to the media is not within the aorta itself, for example, in coronary artery dissection with retrograde distension (Fig.  25.7). Techniques such as virtual non-contrast imaging in dual-­energy CT of the aorta show promise in allowing a simultaneous enhanced and pseudounenhanced evaluation from a single evaluation. However, thus far the use of such virtual unenhanced imaging has been confined to graft evaluation following endovascular aortic repair [46, 47] Fig. 25.6 Intramural haematoma (IMH). A crescentic rim of slightly altered density on a non-contrast image (a) shows no enhancement (b); the morphology is suggestive of IMH. Note the patient’s coronary vein graft arising from the anterior ascending aorta

a

393

but has not yet been validated in the setting of acute aortic syndromes.

25.2.6 Aortic Aneurysm An aneurysm is a permanent localised dilatation of an artery having at least a 50% increase in diameter compared with the expected normal diameter of the artery in question and involving all three vessel layers [31]. MDCT angiography has replaced invasive angiography as the method of choice for the diagnosis of thoracoabdominal aortic aneurysms. Concomitant thoracic and abdominal aortic aneurysms are not unusual, occurring in 15–28% of individuals [48, 49]. Therefore, comprehensive evaluation of thoracic aortic aneurysms should include the entire thoracoabdominal aorta. A few important guiding principles should be kept in mind when deciding if a thoracic aorta is aneurysmal on CT (Fig.  25.8). The normal thoracic diameter increases with age, at a rate of between 0.12 and 0.29 mm per year in one study [50]. Men have an upper limit of ascending aortic diameter that is between 2.0 and 2.6 mm greater than women, and this difference increases with increasing age [51]. Variation of ascending aortic b

A. C. Lindsay et al.

394

a

b

c

Fig. 25.7  Intramural haematoma in a patient with suspected right coronary artery dissection following catheter angiography 24  h prior. (a) Initial CT performed postcontrast demonstrates crescentic, predominantly low density around the ascending aorta, with some high attenuation posteriorly. It was uncertain whether the high density had originated from contrast at catheter

a

b

angiography, contrast during the CT due to acute extravasation or reflected haematoma maturing at different ages. ECG-gated post-contrast CTA performed 5 days later demonstrates a clear continued extravasation of contrast from the RCA on axial (b) and 3D volume-­ rendered (c) images

c

Fig. 25.8  Aneurysmal aortic root and ascending aorta. Note the effacement of the sinotubular junction. (a) Coronal view showing measurement of the valve annulus

and upper ascending aorta. (b) 3D reconstruction of the root. (c) Measurement of the ascending aorta at the level of the pulmonary artery bifurcation

measurement during the cardiac cycle [51, 52] reinforces the importance of noting the phase of the cardiac cycle that measurements have been obtained in on ECG-gated studies, although in practice this variation is only in the order of a few millimetres (3  mm in a study by Lu et  al., for instance [52]). Normal ranges for aortic diameters that are tailored to age, gender and body surface area have been proposed on echocardiography [53] and unenhanced CT [27]. Atherosclerosis remains the commonest cause of aortic aneurysm, but other causes such as infec-

tion, trauma, aortic dissection and genetic predisposition can also be suggested by their various imaging appearances on CT. For instance, aneurysms that are the consequence of hypertension and atherosclerosis are characteristically fusiform, while mycotic aneurysms usually have a saccular morphology and contain eccentric thrombus [54] with a predilection for the ascending aorta [55]. An important feature that can provide a clue to the underlying pathology of a thoracic aortic aneurysm (particularly in a young patient) is the

25  Multidetector Computed Tomography of the Aorta

presence of annuloaortic ectasia. In this condition, CT demonstrates dilated sinuses of Valsalva with concomitant effacement of the normal sinotubular junction, producing a dilated aorta that tapers to a normal aortic arch. This condition is seen in 60–80% of patients with Marfan syndrome, and other causes include homocystinuria and Ehlers-­Danlos syndrome [56, 57].

25.2.7 Inflammatory and Infectious Aortitis CT imaging of the aorta in the setting of inflammation and infection is mainly performed for the assessment of aortic wall thickness and aortic diameter and the presence of mural calcification and for the assessment of the degree of wall irregularity. As described above, an initial nonenhanced scan can be useful to show intramural calcification and exclude intramural haematoma, followed by a contrast-enhanced scan of the aorta. In CT surveillance for chronic inflammation, the unenhanced scan can be eliminated, as periaortic thickening and calcification, together with aneurysm formation or stenosis, can all be evaluated on the post-­contrast acquisition (Fig. 25.9).

395

25.3 Inherited Abnormalities of the Aortic Root and Ascending Aorta 25.3.1 Transposition of the Great Arteries (TGA) Transposition of the great arteries (TGA) is characterised by discordant ventriculoarterial connections [58, 59] (Fig.  25.10a). TGA is a cyanotic condition only compatible with life in the presence of an atrial or ventricular septal defect, or a patent ductus, allowing mixing of the systemic and pulmonary circulations. On CT imaging, the aorta is often seen anteriorly (as the main vascular structure closest to the chest wall), and the pulmonary artery is noted to arise from the morphological left ventricle (Fig.  25.10b) [60]. For many years the Mustard procedure, which used artificial baffles to redirect blood appropriately, was used [61]; however in contemporary practice the condition is corrected by means of the Jatene (arterial switch) procedure, ideally performed within the first 2 weeks of life [62, 63] (Fig. 25.11). Congenitally corrected transposition of the great arteries (ccTGA) is characterised by ventriculoarterial and atrioventricular discordance (Fig. 25.12). As the pulmonary and systemic connections are preserved, this is an acyanotic condition, although it is often associated with other abnormalities (Fig. 25.13a, b). CT imaging shows the aorta to be located anterior and to the left of the pulmonary artery (Fig. 25.13a) [64]. Both these conditions differ from the so-called criss-cross heart, whereby the inlets of the two ventricles appear elongated and cross each other, thereby causing the apical portions of the ventricles to be situated opposite their expected locations [65]. In such cases, the ventricular outflow tracts and arterial connections are preserved.

25.3.2 Tetralogy of Fallot Fig. 25.9 Prospective ECG-gated aortic CT demonstrates inflammatory periaortic thickening in a 39-yearold patient with known Takayasu’s arteritis

Tetralogy of Fallot (ToF) accounts for 3.5% of infants born with congenital heart disease and is

A. C. Lindsay et al.

396

a

b

Ao

RV

Ao

PA

RV LV

c

PA

LV

d RV LV

RA

PV

PV

SVC

PA

LV

Fig. 25.10  A 30-year-old male with a history of transposition of the great arteries (TGA) and Mustard procedure. (a) 3D volume-rendered CT image and (b) sagittal CT image demonstrate ventriculoarterial discordance with the aorta (Ao) originating from the right ventricle (RV) and the pulmonary artery (PA) originating from the left ven-

tricle (LV). (c) Axial CT image demonstrates RV hypertrophy (white arrowheads) and drainage of the pulmonary veins (PV) into the right atrium (RA). (d) Coronal CT image demonstrates a metallic stent (black arrow) in the channel connecting the superior vena cava (SVC) to the left ventricle (LV)

characterised by pulmonary infundibular stenosis, right ventricular hypertrophy, a ventricular septal defect and an overriding aorta [66, 67]. The latter in particular can be well visualised by CT and involves a biventricular connection of the aorta, which is situated above the VSD and connected to both the left and right ventricles (Fig.  25.14), the degree of override being less than 50% (in contrast to double-outlet right ventricle, defined by override of greater than 50%). Total surgical correction is the preferred method of treatment, and preoperative CT can be used to delineate aortic and coronary anatomy and exclude other abnormalities [68, 69].

25.3.3 Truncus Arteriosus and Aortopulmonary Window The normal embryonic arterial trunk (truncus arteriosus) normally divides into the aorta and pulmonary trunk. If it fails to do so, then a persistent truncus arteriosus with a single truncal valve is connected to both ventricles above a VSD and can present at birth with cyanosis, and the clinical syndrome rapidly deteriorates to heart failure [70] (Fig. 25.15). Two main classification systems are described: Collett-Edwards [71] and Van Praagh [72]; each system contains four types based on the origin of the pulmonary

25  Multidetector Computed Tomography of the Aorta

397

a

b

Ao PA

PA Ao

RV LV

Fig. 25.11  Jatene procedure (arterial switch repair). A 4-month-old male with a history of transposition of the great arteries (TGA) and status post-Jatene procedure. (a) 3D volume-rendered and (b) axial maximum intensity projection (MIP) images demonstrate switched positions of the aorta (Ao) and pulmonary artery (PA). The neopul-

a

monary artery (nPA) is located anterior to and its right, and left branches straddle the ascending neo-aorta (nAo) (Lecompte manoeuvre), resulting in a characteristic appearance. Patent ductus arteriosus (PDA) clip is noted (black arrow)

b

PA

Ao

c

mLV

RAA

RA

PA

Ao

mRV

Post

LA

LAA

LA Ant

Fig. 25.12  Congenitally corrected transposition of the great arteries (ccTGA) in a 42-year-old female. (a) 3D volume-rendered image demonstrates the aorta (Ao) arising from the anterior ventricle (Ant) and the pulmonary artery (PA) arising from the posterior ventricle (Post). (b) Axial CT image shows that the anterior ventricle corresponds to a morphological left ventricle (mLV) and

the posterior ventricle corresponds to a morphological right ventricle (mRV). (c) There is atrial situs solitus, as demonstrated by the broad-based right atrial appendage (RAA) and finger-like trabeculated left atrial appendage (LAA). In ccTGA, the switched positions of the great vessels are “corrected” by atrioventricular discordance

A. C. Lindsay et al.

398

a

b

RV

Ao PA

LV

RA

* LA

LV

Fig. 25.13  Congenitally corrected transposition of the great arteries and double inlet left ventricle in a 26-year-­ old female. (a) 3D volume-rendered image demonstrates the aorta (Ao) lying anterior to the pulmonary artery (PA). There is a dominant left ventricle (LV) communicating with a smaller hypertrophied anterior right ventricle (RV) via a large ventricular septal defect (VSD) (black arrowhead). There is ventriculoarterial discordance with the dominant LV supplying the pulmonary circulation and

a

the small RV supplying the systemic circulation. (b) Axial CT image demonstrates atrial situs solitus with both atria draining into the dominant LV. An atrial septal defect is also noted (asterisk). In contrast, with “criss-cross heart”, the AV and VA connections are preserved, although the long axis of the ventricles is twisted with respect to the atrial axis. Hence, a four-chamber view is often hard to obtain

b

Ao

Ao

*

Fig. 25.14  Overriding aorta in a 2-year-old male with tetralogy of Fallot (ToF). (a) Coronal CT image and (b) half-volume 3D image demonstrate the aorta (Ao) overriding the interventricular septum by >50%. There is a large ventricular septal defect (VSD) (asterisk) and

mixing of contrast in the aortic root (black arrow) which represents mixing of blood from the left and right ventricles. Patent ductus arteriosus (PDA) (white arrowhead) is also noted

25  Multidetector Computed Tomography of the Aorta

399

b

a

PA

c

CT

PA

CT CT PA

PA RV

LV

Fig. 25.15  Truncus arteriosus. (a) Part volume-rendered images showing a type A1 truncus arteriosus with a single main pulmonary artery (PA) arising from the common trunk (CT) in a 36-year-old female. The main pulmonary artery then branches to a normal calibre left PA and a markedly stenotic right PA (arrowhead). Axial (b) and 3D

half-volume-rendered (c) images showing a type A2 truncus arteriosus in a 42-year-old female. The right and left pulmonary arteries (PA) originate from the posterior aspect of the common trunk (CT). In (c), the common trunk (CT) straddles the right ventricular (RV) and left ventricular (LV) outflow tracts

arteries. CT can be used to classify patients according to either system. Early treatment with surgical repair in the neonatal stage is advised, with evidence that surgical outcomes are improving [73]. Aortopulmonary (AP) window is another type of persistent communication between the ascending aorta and pulmonary trunk [74] but above separate semilunar valves. Although the condition can be diagnosed by echocardiography, MDCT can be useful in determining the location and size of the window. In both tetralogy of Fallot (Fig.  25.16a) and truncus arteriosus (Fig. 25.16c), major aortopulmonary collateral arteries (MAPCAs) develop to supply blood to the lungs, due to the underdeveloped native pulmonary circulation. MAPCAs are also found in disorders of pulmonary arterial deficiency, such as pulmonary atresia or hypoplastic pulmonary arteries (Fig.  25.16b). They are generally small (often with a diameter of less than 1 mm) and are therefore best visualised by CT [75].

25.3.4 Marfan Syndrome Marfan syndrome [76] is an autosomal dominant condition characterised by defects in the gene for fibrillin-1, a connective tissue protein [77, 78]. It is characterised by a number of skeletal abnormalities (most characteristically tall height); however the majority of morbidity and mortality is caused by abnormalities of the cardiovascular system. The most severe of these is progressive dilatation of the aortic root, which often commences at an early age (Fig. 25.17). Once the orthogonal diameter of the root exceeds 40 mm, regular surveillance scanning is recommended by current guidelines.

25.3.5 O  ther Inherited Conditions Associated with Aortic Disease Other conditions that cause an aortopathy include the vascular form of Ehlers-Danlos syndrome (type IV), Loeys-Dietz syndrome, Turner syndrome and bicuspid aortic valve-associated ­

A. C. Lindsay et al.

400

a

b

Fig. 25.16 (a) 3D volume-rendered image of a 1-year-­ old female with tetralogy of Fallot demonstrates multiple major aortopulmonary collaterals (MAPCAs) (arrowheads) arising from the descending aorta (dAo). (b) Infant with hypoplastic left pulmonary artery with two Fig. 25.17 Marfan syndrome in a 48-year-old male patient. (a) Sagittal CT image and (b) 3D volume-rendered images demonstrate dilatation of the aortic root with effacement of the sinotubular junction (black arrow)

a

aortopathy. The imaging schedule for both pre-­ intervention monitoring and postoperative surveillance is tailored to the individual rather than the underlying cause of the aortopathy. In general, however, both echocardiography (for the assessment of the aortic root and proximal ascending aorta) and CT (for the assessment of the arch and

c

MAPCAs arising from the descending aorta (arrowheads). (c) Multiple MAPCAs arising from the descending aorta (dAo) to supply the right lung in a 1-year-old female with truncus arteriosus associated with marked hypoplasia of the right pulmonary artery

b

descending thoracic aorta) are used at least annually for surveillance, with increased (usually every 6 months) or decreased (biennially or even every 3–5 years) frequency dictated by clinical concern or the rate of progressive d­ilatation [58]. CT ­protocols for surveillance in such cases must be mindful of the young age of these patients and

25  Multidetector Computed Tomography of the Aorta

therefore should take advantage of all possible modern dose-reduction strategies.

25.4 C  T Imaging for Congenital Abnormalities of the Aorta The multiple visualisation techniques offered by CT lend themselves very well to preoperative depiction and postoperative assessment of a variety of congenital aortic abnormalities, including hypoplastic aorta, aortic atresia, arch abnormalities (such as interrupted and persistent fifth arch)

a

401

and coarctation (Figs. 25.18, 25.19, 25.20, 25.21, 25.22 and 25.23). It should be noted that CT imaging can complement MRI for such conditions, for example, in distinguishing coarctation from pseudocoarctation; the latter is a very rare congenital anomaly characterised by elongation and kinking of the aorta at the level of the ligamentum arteriosum without a pressure gradient across the lesion [79, 80]. In pseudocoarctation, the aortic arch may arise higher than the clavicle, but there is either absence or only a mild degree of stenosis of the aortic lumen. Correspondingly, the collateral circulation is also absent.

b Ao

PA

RV

LV

LV

c

RV

d

Ao

Ao

* LV

LV RV RV

Fig. 25.18  Hypoplastic ascending aorta. (a, b) Half-­ volume-­ rendered images of a 1-month-old male with aortic atresia. The left ventricle (LV) is a similar size to the right ventricle (RV). The ascending aorta (white arrowhead) is hypoplastic. The pulmonary artery (PA) is dilated and supplies the aorta via a patent ductus arteriosus (PDA) (arrow). (c) Coronal CT image and (d) half-­

volume-­ rendered image of a 1-year-old female with hypoplastic left heart syndrome and previous Norwood stage I procedure. Both the ascending aorta (white arrowhead) and the left ventricle (LV) are hypoplastic. The right ventricular outflow tract (RVOT) (asterisk) is connected to the transverse aorta (Ao)

A. C. Lindsay et al.

402

a

b

c

* *

Fig. 25.19  3D volume-rendered images demonstrating an interrupted arch (a, b). In (a), the brachiocephalic trunk (white arrowhead) arises immediately off the ascending aorta. However, the descending aorta (white arrow) is supplied by a common aortopulmonary trunk from which the pulmonary arteries are seen to arise laterally (yellow arrow). In this patient with Dandy-Walker syndrome (b), the brachiocephalic trunk (white arrowhead) bifurcates into the left and right common carotid arteries, while the descending aorta (white arrow) and pulmonary arteries

a

*

(black arrows) are once again supplied by a common trunk. However, the subclavian arteries (yellow arrowheads) also arise off the descending aorta and course cranially to their expected positions. 3D volume-rendered image (c) showing a persistent connection (black asterisk) between the pulmonary artery (black arrow) and the proximal aortic arch (white arrow). A ductus arteriosus would typically be found more distally along the aortic arch, and a diagnosis of persistent fifth arch was made

b

rAo

Fig. 25.20 Vascular rings. (a) 3D volume-rendered image of a complete vascular ring in a 4-month-old male. (b) A coronal CT image of the same patient demonstrates indentation of the trachea (arrow) by the larger right aortic arch (rAo) causing stridor. (c) Axial CT and (d) 3D

v­ olume-rendered image of an incomplete vascular ring in a 35-year-old male. There is a right-sided aortic arch and an aberrant left subclavian artery (arrowhead) with a dilated origin known as Kommerell’s diverticulum (white asterisk)

25  Multidetector Computed Tomography of the Aorta

c

403

d

Fig. 25.20 (continued)

a

b

*

LSA

* Fig. 25.21  Coarctation of the aorta. (a) 3D volume-­ rendered image of a 24-year-old male with a tight coarctation (arrowhead) distal to the left subclavian artery (LSA) resulting in the formation of multiple collateral arteries (arrows). (b) A chest radiograph of the same patient demonstrates rib notching (asterisks) due to collateral circulation. (c) 3D volume-rendered image of a 44-year-old

female patient with a history of coarctation repair at the age of 6 weeks. There is restenosis at the level of the LSA. (d) Pseudocoarctation in a 63-year-old male patient. There is elongation and kinking of the aorta at the level of the ligamentum arteriosum but no evidence of stenosis of the aortic lumen or collateral circulation

A. C. Lindsay et al.

404

c

d LSA

Fig. 25.21 (continued)

a

b

LSA

Ao Ao

dAo

dAo

Fig. 25.22 Coarctation repair. (a) 3D volume-rendered image of a 63-year-old male with a history of coarctation repair demonstrates a graft from the ascending aorta (Ao) to the descending aorta (dAo). (b) 3D volume-rendered image of a 44-year-old male with a history of coarctation repair demonstrates a graft from the left subclavian artery

(LSA) to the descending aorta. (c) Sagittal CT image of a 43-year-old male post-coarctation stenting. (d) 3D volume-rendered image of a 56-year-old male with a history of coarctation stenting demonstrates kinking of the stent (arrowhead) which can lead to restenosis

25  Multidetector Computed Tomography of the Aorta

c

405

d

Ao Ao dAo dAo

Fig. 25.22 (continued)

amount of information derived from, CT of the aorta for many years to come.

References

Fig. 25.23  Diffusely hypoplastic thoracic aorta (black asterisk) and hypoplastic pulmonary artery (white arrow) in a 10-month-old male with Williams syndrome. The ascending aorta (white arrowhead) has a normal calibre

25.5 Conclusions MDCT plays a crucial role in assessing congenital and acquired abnormalities of the aorta, facilitating diagnosis, permitting regular surveillance and optimising preoperative planning. Further improvements in scanning techniques and technology are likely to increase the use of, and

1. Flohr TG, Schaller S, Stierstorfer K, Bruder H, Ohnesorge BM, Schoepf UJ.  Multi-detector row CT systems and image-reconstruction techniques. Radiology. 2005;235:756–73. 2. Hsiao EM, Rybicki FJ, Steigner M. CT coronary angiography: 256-slice and 320-detector row scanners. Curr Cardiol Rep. 2010;12:68–75. 3. Hu H, He HD, Foley WD, Fox SH. Four multidetector-­ row helical CT: image quality and volume coverage speed. Radiology. 2000;215:55–62. 4. McCollough CH, Zink FE. Performance evaluation of a multi-slice CT system. Med Phys. 1999;26:2223–30. 5. Chin AS, Fleischmann D.  State-of-the-art computed tomography angiography of acute aortic syndrome. Semin Ultrasound CT MR. 2012;33:222–34. 6. Roos JE, Willmann JK, Weishaupt D, Lachat M, Marincek B, Hilfiker PR.  Thoracic aorta: motion artifact reduction with retrospective and prospective electrocardiography-assisted multi-detector row CT. Radiology. 2002;222:271–7. 7. Arnoldi E, Johnson TR, Rist C, et al. Adequate image quality with reduced radiation dose in prospectively triggered coronary CTA compared with retrospective techniques. Eur Radiol. 2009;19:2147–55. 8. Wu W, Budovec J, Foley WD.  Prospective and retrospective ECG gating for thoracic CT angiography: a comparative study. AJR Am J Roentgenol. 2009;193:955–63.

406 9. Blanke P, Bulla S, Baumann T, et  al. Thoracic aorta: prospective electrocardiographically triggered CT angiography with dual-source CT—feasibility, image quality, and dose reduction. Radiology. 2010;255:207–17. 10. Bolen MA, Popovic ZB, Tandon N, Flamm SD, Schoenhagen P, Halliburton SS.  Image quality, contrast enhancement, and radiation dose of ECG-­ triggered high-pitch CT versus non-ECG-triggered standard-pitch CT of the thoracoabdominal aorta. AJR Am J Roentgenol. 2012;198:931–8. 11. Achenbach S, Marwan M, Ropers D, et al. Coronary computed tomography angiography with a consistent dose below 1 mSv using prospectively electrocardiogram-triggered high-pitch spiral acquisition. Eur Heart J. 2010;31:340–6. 12. Rybicki FJ, Otero HJ, Steigner ML, et  al. Initial evaluation of coronary images from 320-detector row computed tomography. Int J Cardiovasc Imaging. 2008;24:535–46. 13. Kumamaru KK, Steigner ML, Soga S, et al. Coronary enhancement for prospective ECG-gated single R-R axial 320-MDCT angiography: comparison of 60- and 80-mL iopamidol 370 injection. AJR Am J Roentgenol. 2011;197:844–50. 14. Goetti R, Baumuller S, Feuchtner G, et  al. High-­ pitch dual-source CT angiography of the thoracic and abdominal aorta: is simultaneous coronary artery assessment possible? AJR Am J Roentgenol. 2010;194:938–44. 15. Halpern EJ. Triple-rule-out CT angiography for evaluation of acute chest pain and possible acute coronary syndrome. Radiology. 2009;252:332–45. 16. Gruettner J, Fink C, Walter T, et  al. Coronary computed tomography and triple rule out CT in patients with acute chest pain and an intermediate cardiac risk profile. Part 1: impact on patient management. Eur J Radiol. 2013;82:100–5. 17. Takakuwa KM, Halpern EJ.  Evaluation of a triple “rule-out” coronary CT angiography protocol: use of 64-Section CT in low-to-moderate risk emergency department patients suspected of having acute coronary syndrome. Radiology. 2008;248:438–46. 18. Henzler T, Gruettner J, Meyer M, et al. Coronary computed tomography and triple rule out CT in patients with acute chest pain and an intermediate cardiac risk for acute coronary syndrome: part 2: economic aspects. Eur J Radiol. 2013;82:106–11. 19. Zhang LJ, Zhao YE, Schoepf UJ, et al. Seventy-peak kilovoltage high-pitch thoracic aortic CT angiography without ECG gating: evaluation of image quality and radiation dose. Acad Radiol. 2015;22:890–7. 20. Rubin GD, Dake MD, Napel SA, McDonnell CH, Jeffrey RB Jr. Three-dimensional spiral CT angiography of the abdomen: initial clinical experience. Radiology. 1993;186:147–52. 21. Bae KT. Intravenous contrast medium administration and scan timing at CT: considerations and approaches. Radiology. 2010;256:32–61.

A. C. Lindsay et al. 22. Vlahos I, Godoy MC, Naidich DP. Dual-energy computed tomography imaging of the aorta. J Thorac Imaging. 2010;25:289–300. 23. Nakayama Y, Awai K, Funama Y, et  al. Lower tube voltage reduces contrast material and radiation doses on 16-MDCT aortography. AJR Am J Roentgenol. 2006;187:490–7. 24. Ravenel JG, McAdams HP.  Multiplanar and three-­ dimensional imaging of the thorax. Radiol Clin N Am. 2003;41:475–89. 25. Catalano C, Fraioli F, Laghi A, et al. Infrarenal aortic and lower-extremity arterial disease: diagnostic performance of multi-detector row CT angiography. Radiology. 2004;231:555–63. 26. Dalrymple NC, Prasad SR, Freckleton MW, Chintapalli KN. Informatics in radiology (infoRAD): introduction to the language of three-dimensional imaging with multidetector CT. Radiographics. 2005;25:1409–28. 27. Wolak A, Gransar H, Thomson LE, et al. Aortic size assessment by noncontrast cardiac computed tomography: normal limits by age, gender, and body surface area. JACC Cardiovasc Imaging. 2008;1:200–9. 28. Vilacosta I, Roman JA. Acute aortic syndrome. Heart. 2001;85:365–8. 29. Svensson LG, Labib SB, Eisenhauer AC, Butterly JR. Intimal tear without hematoma: an important variant of aortic dissection that can elude current imaging techniques. Circulation. 1999;99:1331–6. 30. Erbel R, Alfonso F, Boileau C, et  al. Diagnosis and management of aortic dissection. Eur Heart J. 2001;22:1642–81. 31. Hiratzka LF, Bakris GL, Beckman JA, et  al. ACCF/ AHA/AATS/ACR/ASA/SCA/SCAI/SIR/STS/SVM guidelines for the diagnosis and management of patients with Thoracic Aortic Disease: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, American Association for Thoracic Surgery, American College of Radiology, American Stroke Association, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society of Interventional Radiology, Society of Thoracic Surgeons, and Society for Vascular Medicine. Circulation. 2010;121:e266–369. 32. Williams DM, Joshi A, Dake MD, Deeb GM, Miller DC, Abrams GD.  Aortic cobwebs: an anatomic marker identifying the false lumen in aortic dissection—imaging and pathologic correlation. Radiology. 1994;190:167–74. 33. LePage MA, Quint LE, Sonnad SS, Deeb GM, Williams DM.  Aortic dissection: CT features that distinguish true lumen from false lumen. AJR Am J Roentgenol. 2001;177:207–11. 34. Nelsen KM, Spizarny DL, Kastan DJ. Intimointimal intussusception in aortic dissection: CT diagnosis. AJR Am J Roentgenol. 1994;162:813–4. 35. Karabulut N, Goodman LR, Olinger GN. CT diagnosis of an unusual aortic dissection with intimointimal intussusception: the wind sock sign. J Comput Assist Tomogr. 1998;22:692–3.

25  Multidetector Computed Tomography of the Aorta 36. Daily PO, Trueblood HW, Stinson EB, Wuerflein RD, Shumway NE.  Management of acute aortic dissections. Ann Thorac Surg. 1970;10:237–47. 37. Williams DM, Lee DY, Hamilton BH, et  al. The dissected aorta: Part III.  Anatomy and radiologic diagnosis of branch-vessel compromise. Radiology. 1997;203:37–44. 38. Yoshida S, Akiba H, Tamakawa M, et  al. Thoracic involvement of type A aortic dissection and intramural hematoma: diagnostic accuracy—comparison of emergency helical CT and surgical findings. Radiology. 2003;228:430–5. 39. Fleischmann D, Rubin GD, Paik DS, et al. Stair-step artifacts with single versus multiple detector-row helical CT. Radiology. 2000;216:185–96. 40. Stanson AW, Kazmier FJ, Hollier LH, et  al. Penetrating atherosclerotic ulcers of the thoracic aorta: natural history and clinicopathologic correlations. Ann Vasc Surg. 1986;1:15–23. 41. Hayashi H, Matsuoka Y, Sakamoto I, et al. Penetrating atherosclerotic ulcer of the aorta: imaging features and disease concept. Radiographics. 2000;20:995–1005. 42. Kazerooni EA, Bree RL, Williams DM.  Penetrating atherosclerotic ulcers of the descending thoracic aorta: evaluation with CT and distinction from aortic dissection. Radiology. 1992;183:759–65. 43. Castaner E, Andreu M, Gallardo X, Mata JM, Cabezuelo MA, Pallardo Y. CT in nontraumatic acute thoracic aortic disease: typical and atypical features and complications. Radiographics. 2003;23:S93–110. 44. Nienaber CA, von KY, Petersen B, et  al. Intramural hemorrhage of the thoracic aorta. Diagnostic and therapeutic implications. Circulation. 1995;92:1465–72. 45. Lemos AA, Pezzullo JC, Fasani P, et al. Can the unenhanced phase be eliminated from dual-phase CT angiography for chest pain? Implications for diagnostic accuracy in acute aortic intramural hematoma. AJR Am J Roentgenol. 2014;203:1171–80. 46. Sommer WH, Graser A, Becker CR, et al. Image quality of virtual noncontrast images derived from dual-­ energy CT angiography after endovascular aneurysm repair. J Vasc Interv Radiol. 2010;21:315–21. 47. Numburi UD, Schoenhagen P, Flamm SD, et  al. Feasibility of dual-energy CT in the arterial phase: imaging after endovascular aortic repair. AJR Am J Roentgenol. 2010;195:486–93. 48. Bickerstaff LK, Pairolero PC, Hollier LH, et  al. Thoracic aortic aneurysms: a population-based study. Surgery. 1982;92:1103–8. 49. Yu T, Zhu X, Tang L, Wang D, Saad N.  Review of CT angiography of aorta. Radiol Clin N Am. 2007;45:461–83. 50. Hannuksela M, Lundqvist S, Carlberg B.  Thoracic aorta—dilated or not? Scand Cardiovasc J. 2006;40:175–8. 51. Mao SS, Ahmadi N, Shah B, et  al. Normal thoracic aorta diameter on cardiac computed tomography in healthy asymptomatic adults: impact of age and gender. Acad Radiol. 2008;15:827–34.

407 52. Lu TL, Huber CH, Rizzo E, Dehmeshki J, von Segesser LK, Qanadli SD.  Ascending aorta measurements as assessed by ECG-gated multi-­ detector computed tomography: a pilot study to establish normative values for transcatheter therapies. Eur Radiol. 2009;19:664–9. 53. Devereux RB, de Simone G, Arnett DK, et al. Normal limits in relation to age, body size and gender of twodimensional echocardiographic aortic root dimensions in persons >/=15 years of age. Am J Cardiol. 2012;110:1189–94. 54. Gonda RL Jr, Gutierrez OH, Azodo MV.  Mycotic aneurysms of the aorta: radiologic features. Radiology. 1988;168:343–6. 55. Posniak HV, Olson MC, Demos TC, Benjoya RA, Marsan RE.  CT of thoracic aortic aneurysms. Radiographics. 1990;10:839–55. 56. Agarwal PP, Chughtai A, Matzinger FR, Kazerooni EA.  Multidetector CT of thoracic aortic aneurysms. Radiographics. 2009;29:537–52. 57. Kimura-Hayama ET, Melendez G, Mendizabal AL, Meave-Gonzalez A, Zambrana GF, Corona-­ Villalobos CP.  Uncommon congenital and acquired aortic diseases: role of multidetector CT angiography. Radiographics. 2010;30:79–98. 58. Warnes CA.  Transposition of the great arteries. Circulation. 2006;114:2699–709. 59. Squarcia U, Macchi C.  Transposition of the great arteries. Curr Opin Pediatr. 2011;23:518–22. 60. Baron RL, Gutierrez FR, Sagel SS, Levitt RG, McKnight RC.  CT of anomalies of the mediastinal vessels. AJR Am J Roentgenol. 1981;137:571–6. 61. Mustard W. Successful two-stage correction of transposition of the great vessels. Surgery. 1964;55:469–72. 62. Jatene AD, Fontes VF, Paulista PP, de Souza LC, Neger F, Galantier M, et al. Successful anatomic correction of transposition of the great vessels. A preliminary report. Arq Bras Cardiol. 1975;28:461–4. 63. Salerno TA, Ricci M.  Recognition of greatness: “the Jatene operation”. J Thorac Cardiovasc Surg. 2008;136:1404. 64. Takasugi JE, Godwin JD, Chen JT. CT in congenitally-­ corrected transposition of the great vessels. Comput Radiol. 1987;11:215–21. 65. Nielsen JC, Parness IA.  Anatomy of a criss-cross heart. Circulation. 2002;106:e41. 66. Ikeda M, Hirosawa K. Tetralogy of fallot. Circulation. 1968;38:21–34. 67. Shinebourne EA, Babu-Narayan SV, Carvalho JS.  Tetralogy of Fallot: from fetus to adult. Heart. 2006;92:1353–9. 68. Pigula FA, Khalil PN, Mayer JE, del Nido PJ, Jonas RA.  Repair of tetralogy of Fallot in neonates and young infants. Circulation. 1999;100:II157–61. 69. Park CS, Lee JR, Lim H-G, Kim W-H, Kim YJ. The long-term result of total repair for tetralogy of Fallot. Eur J Cardiothorac Surg. 2010;38:311–7. 70. Ferdman B, Singh G.  Persistent truncus arte riosus. Curr Treat Options Cardiovasc Med. 2003;5:429–38.

408 71. Collett RW, Edwards JE. Persistent truncus arteriosus; a classification according to anatomic types. Surg Clin North Am. 1949;29:1245–70. 72. Van Praagh R, Van Praagh S.  The anatomy of common aorticopulmonary trunk (truncus arteriosus communis) and its embryologic implications. A study of 57 necropsy cases. Am J Cardiol. 1965;16:406–25. 73. Tlaskal T, Chaloupecky V, Hucin B, Gebauer R, Krupickova S, Reich O, et  al. Long-term results after correction of persistent truncus arteriosus in 83 patients. Eur J Cardiothorac Surg. 2010;37:1278–84. 74. Mert M, Paker T, Akcevin A, Cetin G, Ozkara A, Saltik L, et  al. Diagnosis, management, and results of treatment for aortopulmonary window. Cardiol Young. 2004;14:506–11.

A. C. Lindsay et al. 75. Maeda E, Akahane M, Kato N, Hayashi N, Koga H, Yamada H, et  al. Assessment of major aortopulmonary collateral arteries with multidetector-row computed tomography. Radiat Med. 2006;24:378–83. 76. Cheng TO.  Marfan syndrome, not Marfan’s syn drome. Circulation. 1999;99:166–7. 77. Keane MG, Pyeritz RE.  Medical management of Marfan syndrome. Circulation. 2008;117:2802–13. 78. Yuan S-M, Jing H. Marfan’s syndrome: an overview. Sao Paulo Med J. 2010;128:360–6. 79. Bluemke DA. Pseudocoarctation of the aorta. Cardiol J. 2007;14:205–6. 80. Gaupp RJ, Fagan CJ, Davis M, Epstein NE. Pseudocoarctation of the aorta. J Comput Assist Tomogr. 1981;5:571–3.

Aortic Root Assessment with Computed Tomography in the Context of TAVR

26

Paul Schoenhagen, Lei Zhao, and Xiaohai Ma

26.1 Introduction Over the last few decades, cardiovascular computed tomography (CT) has evolved into an important preoperative tool for guiding surgical and interventional procedures. Initial studies described the relationship/proximity of cardiovascular structures to the sternum prior to re-operative open-heart surgery [1]. With the trend to less invasive procedures, the role of imaging has further evolved [2]. The relationship between innovative endovascular surgical procedures and CT imaging is exemplified by the experience with aorta endovascular stenting [3]. CT is attractive as a tool for planning and guiding transcatheter procedures, because it is widely available with standardized acquisition of 3-D volumetric data [4]. Subsequent analysis allows reconstruction along standard imaging planes but also additional oblique views, with high spatial resolution. Degenerative aortic stenosis has a high prevalence in elderly populations (Fig. 26.1) [5]. TAVR has developed into a viable treatment alternative in patient at high risk for conventional surgery

P. Schoenhagen (*) Cardiovascular Imaging, Imaging Institute and Heart and Vascular Institute, Cleveland, OH, USA e-mail: [email protected]

(Fig.  26.2) [6–9], and CT imaging plays an important role in the evaluation of patient evaluated for TAVR. It provides information about the aortic annulus/root, allows to predict appropriate fluoroscopic projections oriented orthogonal to the aortic valve plane, and is used to assess suitability of the peripheral access vessels. The CT images, similar to echocardiography and cardiac catheterization, have moved out of the “reading room” into the office of the surgeon/ interventionalist, and also into the operating room. This “point-of-care” data availability is critical and has become feasible with central server technology.

26.2 Scan Acquisition 26.2.1 Scanner Technology A detailed discussion about scanner technology is beyond the purpose of this chapter. For modern cardiovascular CT imaging, a scanner of at least the 64-detector technology and software allowing image acquisition with ECG synchronization is critical [10]. These systems support acquisition of 3-D volumetric data with minimal motion artifact.

L. Zhao · X. Ma Department of Radiology, Beijing Anzhen Hospital, Capital Medical University, Beijing, China © Springer-Verlag GmbH Austria, part of Springer Nature 2019 O. H. Stanger et al. (eds.), Surgical Management of Aortic Pathology, https://doi.org/10.1007/978-3-7091-4874-7_26

409

410

P. Schoenhagen et al.

Fig. 26.1  Intraoperative photograph of valve leaflets of patient with severe aortic stenosis (left panel) and intraoperative measurement of annular dimensions (right panel)

Fig. 26.2  This figure shows examples of CT scans after TAVR. Typically the stent frame is clearly identified, but the normal, thin leaflets are not well seen

26.2.2 Patient Preparation Including GFR-Based Approach to Contrast Administration The patient population currently considered for TAVR is characterized by high age and high frequency of comorbidities including renal dysfunc-

tion, limiting the use of iodine-based contrast material. Furthermore, the fluid challenge/shift associated with contrast administration and prehydration is often not well tolerated, in particular in those patients with reduced LV function. Therefore a careful, individualized approach to avoid potential renal injury as well as cardiac decompensation

26  Aortic Root Assessment with Computed Tomography in the Context of TAVR

is important. The choice of the imaging protocol, the associated amount of contrast material, as well as potential prehydration is tailored based on the GFR. Imaging of the root/annulus can be achieved with 50–60  mL, which often also allows assessment of the iliac arteries [11]. Limited iliac imaging can be performed after intra-arterial injection of about 15 ml contrast material [12]. While controlled, slow, and regular heart rate is an advantage for imaging of the aortic annulus, premedication with beta-blocker is contraindicated in patient with severe AS and increased heart rate. Similarly, the use of NTG to improve coronary visualization should be avoided. Radiation exposure associated with diagnostic CT should be minimized using dose-saving protocol options. The choice of protocol should balance required image quality and required radiation dose in individual patients [13, 14]. In the TAVR population with advanced age, radiation exposure is of lesser concern.

26.2.3 Acquisition Protocol Typical TAVR protocols include acquisition of the aortic annulus/root and the entire aorta including the iliac and common femoral arteries.

Coverage of this large volume is most frequently accomplished with two distinct acquisitions (“root” and “aorta/iliac arteries”) (Fig. 26.3). In order to minimize motion artifact, image acquisition of the aortic root must be synchronized to the electrocardiogram (ECG). This can be achieved either by “retrospective ECG gating” or “prospective ECG triggering.” Retrospective gating acquires image data through the entire cardiac cycle. Subsequently, data is reconstructed from the phase with minimal cardiac motion (typically diastole) for image analysis. For TAVR, multiple phases throughout the cardiac cycle are reconstructed. AVA and annulus measurements are typically performed in the systolic phase, which typically demonstrates slightly larger annular dimensions [15, 16]. While retrospective gating is more flexible, it is associated with higher radiation exposure than prospective triggering. Prospective triggering acquires images only in a pre-specified phase of the cardiac cycle and is used in order to reduce radiation exposure. For most patient evaluated for TAVR, retrospective image acquisition is currently recommended. Spatial resolution must be high to provide adequate imaging detail. It is important to acquire a dataset with similar spatial resolution “in-plane” and “through-plane” relative to the body axis.

Mode = spiral, retrospectively-gated - for 3-D and 4-D aortic root analysis - minimal sub-millimeter slice thickness - dose modulation centered in systole - alternative: MRI of the L VOT/root

Mode = spiral, non-gated - or high-pitch prospectively gated - for analysis of aorta and iliac arteries - slice thickness = 1 mm

Mode = spiral, non-gated - intra-arterial injection via pigtail-catheter - 15 ml contrast diluted 1:3

Dill KE, George E, Abbara S. et al. ACR Appropriateness Criterial Imaging for Transcatheter Aortic Valve Replacmement J Am Coll Radiol. 2013 Oct 31. [Epub ahead of print] Achenbach S, Delgado V, Hausleiter J, Schoenhagen P, Min JK, Leipsic JA. SCCT expert consensus document on computed tomography imaging before TAVI/TAVR. J Cardiovasc Comput Tomogr 2012;6:366–80

Fig. 26.3  Summary of CT acquisition protocol

411

412

Such “isotropic datasets” allow subsequent 3-D reconstruction along oblique planes without significant loss of detail in the resulting oblique images. Details vary and depend on the scanner system that is used, but typically an acquisition protocol that allows reconstruction at ≤1.0  mm slice thickness is preferred. Imaging of the aortic root and annulus is typically focused on systole. The remaining segments of the aorta and iliofemoral arteries are typically acquired without ECG synchronization. For these sections, non-­ gated acquisitions are preferred because of faster volume coverage that requires lower volumes of iodinated contrast medium. With 64-detector systems, an ECG-gated dataset that contains the heart and aortic root is typically acquired first (40–60 mL contrast), followed by a second non-gated acquisition to cover the aorta and iliac arteries (additional 50–60 mL bolus). With wide-detector and dual-­source systems, it is feasible to image the entire volume in a single acquisition [17].

26.3 Image Analysis 26.3.1 Aortic Annulus (Figs. 26.4 and 26.5) Measurement of the aortic annulus determines selection of appropriate prosthesis size. An undersized prosthesis increases the risk for embolization and paravalvular regurgitation, and oversized prosthesis may lead to root rupture [18–20]. The aortic “annulus” is a complex three-­ dimensional transition zone between the LVOT and aortic root with a crown-shaped configuration. Imaging defines the annular plane or “virtual basal ring” by the three lowest insertion points of the aortic valve cusps (“hinge points”) into the wall of the LVOT [21]. The annulus typically has an oval, rather than circular, shape. Therefore a single diameter, as acquired with two-dimensional (2-D) imaging techniques, is limited [22–24]. A 3-D reconstruction with measurements of circumference and area provides

P. Schoenhagen et al.

incremental precision, and CT-derived dimensions of the aortic annulus may improve outcome of the procedure [25–27]. With 3-D CT analysis, a cross-sectional image perpendicular to the centerline of the annulus is created at or immediately below the three lowest insertion points of the aortic cusps [9, 21]. The plane is often slightly tilted relative to the LVOT or valvular plane. Once the annular plane has been identified, measurements include minimal and maximal diameter (mean diameter is calculated), circumference (corresponding derived diameter is calculated), and area of the aortic annulus (corresponding diameter is calculated). Measurement of aortic annulus dimensions is typically performed in systole at 20–30% RR interval (similar to echocardiography), because the planimetered annular area and mean diameters are larger in systole than in diastole. The circumference may be more stable throughout the cardiac cycle, but annular area may provide better interobserver agreement [9, 28–30]. While initial recommendations have typically been based on 2-D echocardiographic measurements, the experience with CT has documented that 3-D-based dimensions allow better prediction of procedural outcome, e.g., incidence of paravalvular regurgitation [27, 31]. However, there is a lack of large clinical trials demonstrating that CT-derived dimensions permit improved prosthesis selection.

26.3.2 Aortic Valve Calcification (Fig. 26.6) Degenerative AS is characterized by thickening of the aortic valve cusps with typically extensive calcification. The presence and extent of valvular calcifications are likely important for prosthesis anchorage [32]. Lack of calcification may not provide a stable deployment zone, and excessive calcification may hamper the apposition of the prosthesis and may leave gaps between the prosthetic frame and the native aortic root with subse-

26  Aortic Root Assessment with Computed Tomography in the Context of TAVR

413

Fig. 26.4  Measurement of the aortic annulus as a plane at the lowest insertion points of the aortic valve leaflets. The three insertion points are displayed in the right upper panel. The image plane is perpendicular to the aortic root centerline (right lower panel)

quent occurrence of paravalvular aortic regurgitation [33, 34]. Aortic valve calcification may also be related to the risk of prosthesis dislodgement and annular rupture [35, 36]. The extent of aortic valve calcification appears related to the occurrence of embolism of calcified particles and stroke risk [37]. Quantification of aortic valve calcification in CT can be achieved with semiquantitative scores, for example, considering the circularity of cal-

cium or the number of affected cusps or a continuous scale using calcium scoring [34]. Quantification of aortic valve calcification also appears relevant in patient with suspected low-flow/low-gradient aortic stenosis [38, 39]. In these patients, mean gradient is discordant from AVA and is determined by multiple factors, including valvular anatomy and arterial compliance. The aortic valve calcium load as assessed by MDCT is associated with AS severity. The

414

majority of patients with discordant low gradient have significant aortic valve leaflet calcification reflective of severe calcified aortic valve disease, suggesting clinical value of calcium quantification in the management of these complex patients [40, 41]. Small studies have demonstrated that in patients with low-­ gradient low-flow AS, higher valvular calcium score predicts worse long-term mortality. AVR is associated with improved survival in patients with higher valve scores [42].

P. Schoenhagen et al.

26.3.3 Aortic Valve Function (Fig. 26.7) Careful description of the shape and symmetry of the valve opening area, including relationship to calcification, confirms the presence of severe stenosis and can differentiate between tricuspid and (partial) bicuspid valve anatomy. This is particularly valuable in patient with severe valvular calcification (when echocardiographic images are sometimes limited) and in patient with suspected low-flow/low-gradient aortic stenosis [38, 39]. The assessment is best made by review of the dynamic 4-D dataset, which allows planimetry of the aortic valve orifice area [43–45].

26.3.4 Height of the Coronary Ostia (Figs. 26.8 and 26.9)

Fig. 26.5  At the annulus, diameter, circumference, and area are measured

During TAVR, the diseased aortic cusps are displaced in the coronary sinuses. Displacement of calcified aortic cusp during TAVR implantation infrequently can cause occlusion of the coronary ostia. The distance between the annular plane and the coronary ostia relative to the height of the valve leaflets is therefore important in order to avoid coronary occlusion [46]. A low origin of the coronary arteries, long cusps, and shallow sinuses increases the risk of potential coronary occlusion. In a study of 100 patients with aortic stenosis undergoing CT, the average distance of

Fig. 26.6  “Calcium scoring” of the aortic valve provides a quantitative measure of leaflet calcification

26  Aortic Root Assessment with Computed Tomography in the Context of TAVR

415

Fig. 26.7  Description of the shape and symmetry of the valve opening area confirms the presence of severe stenosis and can differentiate between tricuspid and (partial) bicuspid valve anatomy. Planimetry of the aortic valve orifice area provides a measure of aortic valve stenosis severity but is typically not the primary indication to perform CT

Aortic valve area = 0.7 cm2

Fig. 26.8  Coronary ostial height is measured between the annular plane and ostia of the coronary arteries

left coronary ostium and right coronary ostium was found to be 15.5 ± 2.9 mm and 17.3 ± 3.6 mm, respectively [47]. A minimum distance values of

10–14 mm between the coronary ostia and leaflet insertion are usually suggested but depend on valve type [46].

416

P. Schoenhagen et al.

Fig. 26.9  In order to measure ostial height of the left main (LM, left panels) and right coronary artery (RCA, right panels), the images are rotated in the annular plane

Fig. 26.10 During catheter-based implantation, it is important to use fluoroscopic projections orthogonal to the aortic annular plane. By identifying the corresponding

LAO/RAO and cranial/caudal angulation, CT allow to predict appropriate projection angles

26.3.5 Root Angulation (Figs. 26.10, 26.11, and 26.12)

nal view onto the aortic annular plane without foreshortening. Standard angiographic views include a projection with the right coronary cusp in the center, and the left and non-coronary cusps positioned symmetrically to either side of the

During TAVR, it is important to use a fluoroscopic projection that provides an exact orthogo-

26  Aortic Root Assessment with Computed Tomography in the Context of TAVR

417

Fig. 26.11  Rotation of the images around the annular plane (left lower quadrant) results in orthogonal images (left upper and right lower quadrant). The angiographic

projections are displayed by most workstations (left lower aspect of each quadrant)

right coronary cusp, or LAO 20 and RAO 20 projections. By identifying the corresponding cranial/caudal angulation, CT allow to predict appropriate projection angles that will provide these orthogonal views onto the aortic valve plane [48–52]. Angulations predicted from pre-­ procedural CT have been shown to correlate well with 3-D rotational angiography at the time of the procedure. These measurements were initially

performed by manual image reconstruction, but dedicated, (semi)automated software programs allow root segmentation and provide the angulations automatically. It is important to remember that the patient must be positioned in similar fashion during CT acquisition and the procedure. If the position on the procedure table differs, corrections need to be made to account for the difference in patient orientation.

418

P. Schoenhagen et al.

Fig. 26.12  In this figure images orthogonal to the annular plane in a RAO 20 projection (left upper panel) and LAO 20 projection (right lower panel) are demonstrated

Fig. 26.13  Modern workstations allow to combine data reconstructed throughout the cardiac cycle into cine-loops for dynamic review for assessment of LV size and function. This data is used to confirm findings from echocardiography

26.3.6 Left Ventricular Assessment (Fig. 26.13) Acquisition of the aortic root and heart with retrospective gating allows to reconstruct images

throughout the entire cardiac cycle. These can be combined into cine-loops for dynamic review for assessment of LV size and function. These data are used to confirm findings from echocardiography, including LV size and function, and assess-

26  Aortic Root Assessment with Computed Tomography in the Context of TAVR

419

Fig. 26.14  CT assessment of coronary anatomy beyond ostial height (see above) is of limited value in the patient population evaluated for TAVR

ment of the LV apex (transapical access), including LV apical thrombi. A description of the position of the LV apex relative to the chest wall and alignment of the LV axis with LV outflow tract (LVOT) orientation may be useful information in the case of transapical access.

26.3.7 Coronary Arteries (Fig. 26.14) Beyond assessment of coronary ostial height (see above), the data can be used to assess coronary anatomy. However, because of the high prevalence of significant coronary artery disease in the patient population evaluated for TAVR and the known limitations of coronary CTA in the presence of significant calcification (“calcium bloom-

ing”), coronary CTA is limited in this patient population [53–55]. Furthermore, premedication with b-blocker and nitroglycerin is typically not possible. If coronary bypass grafts are present, their position and potential adhesion to the sternum may be of relevance if emergency conversion to open-heart surgery is required [1].

26.3.8 Root, Thoracic and Abdominal Aorta, and Iliac Arteries (Figs. 26.15, 26.16, and 26.17) Evaluation of aortic anatomy including the arch branch vessels and iliac arteries allows to plan access route and avoids unexpected complica-

420

P. Schoenhagen et al.

Fig. 26.15  Evaluation of aortic anatomy including the arch branch vessels and iliac arteries allows to plan access route and avoids unexpected complications during the procedure

tions during the procedure. Standard measurements of the aortic root, sinotubular junction (STJ), and aorta are performed. Assessment of plaque burden in the ascending aorta and arch may predict risk of stroke, as atheroembolism from the ascending aorta or aortic arch is assumed to be the most common cause of periprocedural stroke during TAVI/TAVR.  If a transaortic approach is considered, the position of the ascending aorta relative to the chest wall and aortic calcification is of importance [56]. Luminal dimensions and calcification of the infrarenal aorta and aortic bifurcation should be assessed. The subclavian and iliac arteries are assessed for luminal size, atherosclerotic changes, and tortuosity. Contraindications to peripheral access include small luminal size, tortuosity with kinking, dissection, or large thrombi protruding into the lumen. Vascular complications are a significant cause of mortality and morbidity in transfemoral TAVI/ TAVR [57–59]. Ongoing device development has

resulted in progressive reduction of the profile of the delivery systems for transfemoral TAVR.  Better patient selection and smaller sheaths have been associated with lower reported vascular complication rates. Risk factors for vascular complications are an external sheath diameter that exceeds the minimal artery diameter, moderate or severe calcification, and peripheral vascular disease. CT can consistently identify the presence of these risk factors [60].

26.4 Image Availability and Storage Selection of patients suitable for the TAVR procedure involves an interdisciplinary team including cardiologists, cardiothoracic surgeons, and imaging specialists/radiologists (heart team). Sharing of the huge amount of accumulation data (both clinical and imaging data) from a centralized archive with remote access and display, optimized for the workflow of individual groups of

26  Aortic Root Assessment with Computed Tomography in the Context of TAVR

421

Fig. 26.16  The subclavian and iliac arteries are assessed for luminal size, atherosclerotic changes, and tortuosity

clinicians, facilitates decision-making and clinical area [61, 62]. However, the need to share complex data creates new challenges for data

storage, post-processing, and display. It requires a complex infrastructure/network spanning across multiple locations in large hospital sys-

P. Schoenhagen et al.

422

Fig. 26.17  In patient with reduced kidney function, CT with intra-arterial injection can be considered (in this patient, a “pigtail” catheter is left in place at the end of diagnostic coronary angiography)

tems. This has been achieved by client-server solutions, where a single dataset is localized and modified on a powerful central server, while the local workstation used only the access point and command structure. These systems allow access to the data from multiple clinical workstations and also mobile access [63, 64]. These systems are a form of cloud computing, with the totality of the integrated systems being consistent with a “private medical-grade cloud.”

26.5 C  onclusion: Clinical Impact of Imaging CT imaging plays an important role in procedural planning for TAVI/TAVR and is an increasingly integrated component of TAVR work-up and clinical trials. The imaging specialists responsible for the interpretation of the CT examination should be integrated in the TAVI/TAVR team to

ensure appropriate incorporation into the patient selection process and procedure planning. Further standardization and eventual assessment of clinical impact will be necessary [65, 66].

References 1. Kamdar AR, Meadows TA, Roselli EE, Gorodeski EZ, Curtin RJ, Sabik JF, Schoenhagen P, White RD, Lytle BW, Flamm SD, Desai MY.  Multidetector computed tomographic angiography in planning of reoperative cardiothoracic surgery. Ann Thorac Surg. 2008;85:1239–45. 2. Moodley S, Schoenhagen P, Gillinov AM, Mihaljevic T, Flamm SD, Griffin BP, Desai MY.  Preoperative multidetector computed tomography angiography for planning of minimally invasive robotic mitral valve surgery: impact on decision making. J Thorac Cardiovasc Surg. 2013;146:262–8. 3. Johnson PT, Black JH, Zimmerman SL, Fishman EK.  Thoracic endovascular aortic repair: literature review with emphasis on the role of multidetector

26  Aortic Root Assessment with Computed Tomography in the Context of TAVR computed tomography. Semin Ultrasound CT MR. 2012;33:247–64. 4. Achenbach S, Delgado V, Hausleiter J, Schoenhagen P, Min JK, Leipsic JA.  SCCT expert consensus document on computed tomography imaging before transcatheter aortic valve implantation (TAVI)/ transcatheter aortic valve replacement (TAVR). J Cardiovasc Comput Tomogr. 2012;6:366–80. 5. Vahanian A, Alfieri O, Andreotti F, Antunes MJ, Barón-Esquivias G, Baumgartner H, Borger MA, Carrel TP, De Bonis M, Evangelista A, Falk V, Iung B, Lancellotti P, Pierard L, Price S, Schäfers HJ, Schuler G, Stepinska J, Swedberg K, Takkenberg J, Von Oppell UO, Windecker S, Zamorano JL, Zembala M, Joint Task Force on the Management of Valvular Heart Disease of the European Society of Cardiology (ESC); European Association for Cardio-Thoracic Surgery (EACTS). Guidelines on the management of valvular heart disease (version 2012). Eur Heart J. 2012;33:2451–96. 6. Leon MB, Smith CR, Mack M, Miller DC, Moses JW, Svensson LG, Tuzcu EM, Webb JG, Fontana GP, Makkar RR, Brown DL, Block PC, Guyton RA, Pichard AD, Bavaria JE, Herrmann HC, Douglas PS, Petersen JL, Akin JJ, Anderson WN, Wang D, Pocock S, PARTNER Trial Investigators. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med. 2010;363:1597–607. 7. Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG, Tuzcu EM, Webb JG, Fontana GP, Makkar RR, Williams M, Dewey T, Kapadia S, Babaliaros V, Thourani VH, Corso P, Pichard AD, Bavaria JE, Herrmann HC, Akin JJ, Anderson WN, Wang D, Pocock SJ, PARTNER Trial Investigators. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med. 2011;364:2187–98. 8. Reardon MJ, Adams DH, Coselli JS, Deeb GM, Kleiman NS, Chetcuti S, Yakubov SJ, Heimansohn D, Hermiller J Jr, Hughes GC, Harrison JK, Khabbaz K, Tadros P, Zorn GL 3rd, Merhi W, Heiser J, Petrossian G, Robinson N, Maini B, Mumtaz M, Lee JS, Gleason TG, Resar J, Conte J, Watson D, Chenoweth S, Popma JJ, CoreValve US Clinical Investigators. Self-­expanding transcatheter aortic valve replacement using alternative access sites in symptomatic patients with severe aortic stenosis deemed extreme risk of surgery. J Thorac Cardiovasc Surg. 2014;148(6):2869– 2876.e7. https://doi.org/10.1016/j.jtcvs.2014.07.020. 9. Holmes DR Jr, Mack MJ, Kaul S, Agnihotri A, Alexander KP, Bailey SR, Calhoon JH, Carabello BA, Desai MY, Edwards FH, Francis GS, Gardner TJ, Kappetein AP, Linderbaum JA, Mukherjee C, Mukherjee D, Otto CM, Ruiz CE, Sacco RL, Smith D, Thomas JD. 2012 ACCF/AATS/SCAI/STS expert consensus document on transcatheter aortic valve replacement. J Am Coll Cardiol. 2012;59:1200–54.

423

10. Abbara S, Arbab-Zadeh A, Callister TQ, Desai MY, Mamuya W, Thomson L, Weigold WG.  SCCT guidelines for performance of coronary computed tomographic angiography: a report of the Society of Cardiovascular Computed Tomography Guidelines Committee. J Cardiovasc Comput Tomogr. 2009;3:190–204. 11. Bischoff B, Meinel FG, Reiser M, Becker HC. Novel single-source high-pitch protocol for CT angiography of the aorta: comparison to high-pitch dualsource protocol in the context ofTAVI planning. Int J Cardiovasc Imaging. 2013;29:1159–65. 12. Joshi SB, Mendoza DD, Steinberg DH, Goldstein MA, Lopez CF, Raizon A, Weissman G, Satler LF, Pichard AD, Weigold WG.  Ultra-low-dose intra-­ arterial contrast injection for iliofemoral computed tomographic angiography. JACC Cardiovasc Imaging. 2009;2:1404–11. 13. Halliburton SS, Abbara S, Chen MY, Gentry R, Mahesh M, Raff GL, Shaw LJ, Hausleiter J, Society of Cardiovascular Computed Tomography. SCCT guidelines on radiation dose and dose-optimization strategies in cardiovascular CT. J Cardiovasc Comput Tomogr. 2011;5:198–224. 14. Schoenhagen P, Baker M.  Our pre-occupation with ‘ultra-low’ radiation dose radiation exposure, low contrast resolution, and cardiovascular CT imaging. J Cardiovasc Comput Tomogr. 2009;8:426. https://doi. org/10.1016/j.jcct.2014.09.004. 15. Willson AB, Webb JG, LaBounty TM, Achenbach S, Moss R, Wheeler M, Thompson C, Min JK, Gurvitch R, Norgard BL, Hague C, Toggweiler S, Binder RK, Freeman M, Poulson S, Wood DA, Leipsic J. 3-Dimensional aortic annular assessment by multidetector computed tomography predicts moderate or severe paravalvular regurgitation after transcatheter aortic valve replacement: implications for sizing of transcatheter heart valves. J Am Coll Cardiol. 2012;59:1287–94. 16. Blanke P, Reinöhl J, Schlensak C, Siepe M, Pache G, Euringer W, Geibel-Zehender A, Bode C, Langer M, Beyersdorf F, Zehender M.  Prosthesis oversizing in balloon-expandable transcatheter aortic valve implantation is associated with contained rupture of the aortic root. Circ Cardiovasc Interv. 2012;5:540–38. 17. Wuest W, Anders K, Schuhbaeck A, May MS, Gauss S, Marwan M, Arnold M, Ensminger S, Muschiol G, Daniel WG, Uder M, Achenbach S. Dual source multidetector CT-angiography before transcatheter aortic valve implantation (TAVI) using a high-pitch spiral acquisition mode. Eur Radiol. 2012;22:51–8. 18. Gilard M, Eltchaninoff H, Iung B, Donzeau-Gouge P, Chevreul K, Fajadet J, Leprince P, Leguerrier A, Lievre M, Prat A, Teiger E, Lefevre T, Himbert D, Tchetche D, Carrié D, Albat B, Cribier A, Rioufol G, Sudre A, Blanchard D, Collet F, Dos Santos P, Meneveau N, Tirouvanziam A, Caussin C, Guyon P, Boschat J, Lebreton H, Collart F, Houel R, Delpine S, Souteyrand G, Favereau X, Ohlmann P, Doisy

424 V, Grollier G, Gommeaux A, Claudel JP, Bourlon F, Bertrand B, Van Belle E, Laskar M, FRANCE 2 Investigators. Registry of transcatheter aortic-valve implantation in high-risk patients. N Engl J Med. 2012;366:1705–15. 19. Sinning JM, Hammerstingl C, Vasa-Nicotera M, Adenauer V, Lema Cachiguango SJ, Scheer AC, Hausen S, Sedaghat A, Ghanem A, Müller C, Grube E, Nickenig G, Werner N. Aortic regurgitation index defines severity of peri-prosthetic regurgitation and predicts outcome in patients after transcatheter aortic valve implantation. J Am Coll Cardiol. 2012;59:1134–41. 20. Kodali SK, Williams MR, Smith CR, Svensson LG, Webb JG, Makkar RR, Fontana GP, Dewey TM, Thourani VH, Pichard AD, Fischbein M, Szeto WY, Lim S, Greason KL, Teirstein PS, Malaisrie SC, Douglas PS, Hahn RT, Whisenant B, Zajarias A, Wang D, Akin JJ, Anderson WN, Leon MB, Trial Investigators PARTNER.  Two-year outcomes after transcatheter or surgical aortic-valve replacement. N Engl J Med. 2012;366:1686–95. 21. Piazza N, de Jaegere P, Schultz C, Becker AE, Serruys PW, Anderson RH. Anatomy of the aortic valvar complex and its implications for transcatheter implantation of the aortic valve. Circ Cardiovasc Interv. 2008;1:74–81. 22. Altiok E, Koos R, Schröder J, Brehmer K, Hamada S, Becker M, Mahnken AH, Almalla M, Dohmen G, Autschbach R, Marx N, Hoffmann R.  Comparison of two-dimensional and three-dimensional imaging techniques for measurement of aortic annulus diameters before transcatheter aortic valve implantation. Heart. 2011;97:1578–84. 23. Ng AC, Delgado V, van der Kley F, Shanks M, van de Veire NR, Bertini M, Nucifora G, van Bommel RJ, Tops LF, de Weger A, Tavilla G, de Roos A, Kroft LJ, Leung DY, Schuijf J, Schalij MJ, Bax JJ. Comparison of aortic root dimensions and geometries before and after transcatheter aortic valve implantation by 2- and 3-dimensional transesophageal echocardiography and multislice computed tomography. Circ Cardiovasc Imaging. 2010;3:94–102. 24. Gurvitch R, Webb JG, Yuan R, Johnson M, Hague C, Willson AB, Toggweiler S, Wood DA, Ye J, Moss R, Thompson CR, Achenbach S, Min JK, Labounty TM, Cury R, Leipsic J.  Aortic annulus diameter determination by multidetector computed tomography: reproducibility, applicability, and implications for transcatheter aortic valve implantation. JACC Cardiovasc Interv. 2011;4:1235–45. 25. Willson AB, Webb JG, Gurvitch R, Wood DA, Toggweiler S, Binder R, Freeman M, Madden M, Hague C, Leipsic J.  Structural integrity of balloon-­ expandable stents after transcatheter aortic valve replacement: assessment by multidetector computed tomography. JACC Cardiovasc Interv. 2012;5:525–32. 26. Jabbour A, Ismail TF, Moat N, Gulati A, Roussin I, Alpendurada F, Park B, Okoroafor F, Asgar A, Barker S, Davies S, Prasad SK, Rubens M, Mohiaddin

P. Schoenhagen et al. RH.  Multimodality imaging in transcatheter aortic valve implantation and post-procedural aortic regurgitation: comparison among cardiovascular magnetic resonance, cardiac computed tomography, and echocardiography. J Am Coll Cardiol. 2011;58:2165–73. 27. Jilaihawi H, Kashif M, Fontana G, Furugen A, Shiota T, Friede G, Makhija R, Doctor N, Leon MB, Makkar RR.  Cross-sectional computed tomographic assessment improves accuracy of aortic annular sizing for transcatheter aortic valve replacement and reduces the incidence of paravalvular aortic regurgitation. J Am Coll Cardiol. 2012;59:1275–86. 28. Hamdan A, Guetta V, Konen E, Goitein O, Segev A, Raanani E, Spiegelstein D, Hay I, Di Segni E, Eldar M, Schwammenthal E. Deformation dynamics and mechanical properties of the aortic annulus by 4-dimensional computed tomography: insights into the functional anatomy of the aortic valve complex and implications for transcatheter aortic valve therapy. J Am Coll Cardiol. 2012;59:119–27. 29. Masri A, Schoenhagen P, Svensson L, Kapadia SR, Griffin BP, Tuzcu EM, Desai MY.  Dynamic characterization of aortic annulus geometry and morphology with multimodality imaging: predictive value for aortic regurgitation after transcatheter aortic valve replacement. J Thorac Cardiovasc Surg. 2014;147:1847–54. 30. Bolen MA, Popovic ZB, Dahiya A, Kapadia SR, Tuzcu EM, Flamm SD, Halliburton SS, Schoenhagen P.  Prospective ECG-triggered, axial 4-D imaging of the aortic root, valvular, and left ventricular structures: a lower radiation dose option for preprocedural TAVR imaging. J Cardiovasc Comput Tomogr. 2012;6:393–8. 31. Schultz CJ, Tzikas A, Moelker A, Rossi A, Nuis RJ, Geleijnse MM, van Mieghem N, Krestin GP, de Feyter P, Serruys PW, de Jaegere PP.  Correlates on MSCT of paravalvular aortic regurgitation after transcatheter aortic valve implantation using the Medtronic CoreValve prosthesis. Catheter Cardiovasc Interv. 2011;78:446–55. 32. Van Mieghem NM, Schultz CJ, van der Boon RM, Nuis RJ, Tzikas A, Geleijnse ML, van Domburg RT, Serruys PW, de Jaegere PP.  Incidence, timing, and predictors of valve dislodgment during TAVI with the medtronic corevalve system. Catheter Cardiovasc Interv. 2012;79:726–32. 33. Delgado V, Ng AC, van de Veire NR, van der Kley F, Schuijf JD, Tops LF, de Weger A, Tavilla G, de Roos A, Kroft LJ, Schalij MJ, Bax JJ. Transcatheter aortic valve implantation: role of multi-detector row computed tomography to evaluate prosthesis positioning and deployment in relation to valve function. Eur Heart J. 2010;31:1114–23. 34. John D, Buellesfeld L, Yuecel S, Mueller R, Latsios G, Beucher H, Gerckens U, Grube E.  Correlation of device landing zone calcification and acute procedural success in patients undergoing transcatheter aortic valve implantations with the self-expanding CoreValve prosthesis. JACC Cardiovasc Interv. 2010;3:233–43.

26  Aortic Root Assessment with Computed Tomography in the Context of TAVR 35. Geisbusch S, Bleiziffer S, Mazzitelli D, Ruge H, Bauernschmitt R, Lange R.  Incidence and management of CoreValve dislocation during transcatheter aortic valve implantation. Circ Cardiovasc Interv. 2010;3:531–6. 36. Webb JG, Pasupati S, Humphries K, Thompson C, Altwegg L, Moss R, Sinhal A, Carere RG, Munt B, Ricci D, Ye J, Cheung A, Lichtenstein SV.  Percutaneous transarterial aortic valve replacement in selected high-risk patients with aortic stenosis. Circulation. 2007;116:755–63. 37. Athappan G, Gajulapalli RD, Sengodan P, Bhardwaj A, Ellis SG, Svensson L, Tuzcu EM, Kapadia SR.  Influence of transcatheter aortic valve replacement strategy and valve design on stroke after transcatheter aortic valve replacement: a meta-analysis and systematic review of literature. J Am Coll Cardiol. 2014;63:2101–10. 38. Ozkan A. Low gradient “severe” aortic stenosis with preserved left ventricular ejection fraction. Cardiovasc Diagn Ther. 2012;2:19–27. 39. Ozkan A, Hachamovitch R, Kapadia SR, Tuzcu EM, Marwick TH. Impact of aortic valve replacement on outcome of symptomatic patients with severe aortic stenosis with low gradient and preserved left ventricular ejection fraction. Circulation. 2013;128:622–31. 40. Clavel MA, Messika-Zeitoun D, Pibarot P, Aggarwal SR, Malouf J, Araoz PA, Michelena HI, Cueff C, Larose E, Capoulade R, Vahanian A, Enriquez-­ Sarano M.  The complex nature of discordant severe calcified aortic valve disease grading: new insights from combined Doppler echocardiographic and computed tomographic study. J Am Coll Cardiol. 2013;62:2329–38. 41. Clavel MA, Côté N, Mathieu P, Dumesnil JG, Audet A, Pépin A, Couture C, Fournier D, Trahan S, Pagé S, Pibarot P.  Paradoxical low-flow, lowgradient aortic stenosis despite preserved left ventricular ejection fraction: new insights from weights of operatively excised aortic valves. Eur Heart J. 2014;35:2655–62. 42. Aksoy O, Cam A, Agarwal S, Ige M, Yousefzai R, Singh D, Griffin BP, Schoenhagen P, Kapadia SR, Tuzcu ME.  Significance of aortic valve calcification in patients with low-gradient low-flow aortic stenosis. Clin Cardiol. 2014;37:26–31. 43. Schoenhagen P, Hausleiter J, Achenbach S, Desai MY, Tuzcu EM. Computed tomography in the evaluation for transcatheter aortic valve implantation (TAVI). Cardiovasc Diagn Ther. 2011;1:44–56. 44. Pflederer T, Achenbach S. Aortic valve stenosis: CT contributions to diagnosis and therapy. J Cardiovasc Comput Tomogr. 2010;4:355–64. 45. Ropers D, Ropers U, Marwan M, Schepis T, Pflederer T, Wechsel M, Klinghammer L, Flachskampf FA, Daniel WG, Achenbach S. Comparison of dual-­source computed tomography for the quantification of the aortic valve area in patients with aortic stenosis versus transthoracic echocardiography and invasive hemodynamic assessment. Am J Cardiol. 2009;4:1561–7.

425

46. Masson JB, Kovac J, Schuler G, Ye J, Cheung A, Kapadia S, Tuzcu EM, Kodali S, Leon MB, Webb JB.  Transcatheter aortic valve implantation: review of the nature, management, and avoidance of procedural complications. JACC Cardiovasc Interv. 2009;2:811–920. 47. Tops LF, Wood DA, Delgado V, Schuijf JD, Mayo JR, Pasupati S, Lamers FP, van der Wall EE, Schalij MJ, Webb JG, Bax JJ. Noninvasive evaluation of the aortic root with multislice computed tomography. Implications for transcatheter aortic valve replacement. JACC Cardiovasc Imaging. 2008;1:321–30. 48. Binder RK, Leipsic J, Wood D, Moore T, Toggweiler S, Willson A, Gurvitch R, Freeman M, Webb JG.  Prediction of optimal deployment projection for transcatheter aortic valve replacement: angiographic 3-dimensional reconstruction of the aortic root versus multidetector computed tomography. Circ Cardiovasc Interv. 2012;5:247–52. 49. Arnold M, Achenbach A, Pfeiffer I, Ensminger S, Marwan M, Einhaus F, Pflederer T, Ropers D, Schuhbaeck A, Anders K, Lell M, Uder M, Ludwig J, Weyand M, Daniel WG, Feyrer R.  A method to determine suitable fluoroscopic projections for transcatheter aortic valve implantation by computed tomography. J Cardiovasc Comput Tomogr. 2012;6(6):422–8. 50. Tzikas A, Schultz C, Van Mieghem NM, de Jaegere PP, Serruys PW.  Optimal projection estimation for transcatheter aortic valve implantation based on contrast-aortography: validation of a Prototype Software. Catheter Cardiovasc Interv. 2010;76:602–7. 51. Gurvitch R, Wood DA, Leipsic J, Tay E, Johnson M, Ye J, Nietlispach F, Wijesinghe N, Cheung A, Webb JG.  Multislice computed tomography for prediction of optimal angiographic deployment projections during transcatheter aortic valve implantation. JACC Cardiovasc Interv. 2010;3:1157–65. 52. Kurra V, Kapadia SR, Tuzcu EM, Halliburton SS, Svensson L, Roselli EE, Schoenhagen P.  Pre-­ procedural imaging of aortic root orientation and dimensions: comparison between X-ray angiographic planar imaging and 3-dimensional multidetector row computed tomography. JACC Cardiovasc Interv. 2010;3:105–13. 53. Goel SS, Ige M, Tuzcu EM, Ellis SG, Stewart WJ, Svensson LG, Lytle BW, Kapadia SR.  Severe aortic stenosis and coronary artery disease-implications for management in the transcatheter aortic valve replacement era: a comprehensive review. J Am Coll Cardiol. 2013;62:1–10. 54. Salhab KF, Al Kindi AH, Lane JH, Knudson KE, Kapadia S, Roselli EE, Tuzcu ME, Svensson LG. Concomitant percutaneous coronary intervention and transcatheter aortic valve replacement: safe and feasible replacement alternative approaches in high-­ risk patients with severe aortic stenosis and coronary artery disease. J Card Surg. 2013;28:481–3. 55. Andreini D, Pontone G, Mushtaq S, Bartorelli AL, Ballerini G, Bertella E, Segurini C, Conte E, Annoni

426 A, Baggiano A, Formenti A, Fusini L, Tamborini G, Alamanni F, Fiorentini C, Pepi M.  Diagnostic accuracy of multidetector computed tomography coronary angiography in 325 consecutive patients referred for transcatheter aortic valve replacement. Am Heart J. 2014;168:332–9. 56. Bruschi G, de Marco F, Botta L, Cannata A, Oreglia J, Colombo P, Barosi A, Colombo T, Nonini S, Paino R, Klugmann S, Martinelli L.  Direct aortic access for transcatheter self-expanding aortic bioprosthetic valves implantation. Ann Thorac Surg. 2001;94:497–503. 57. Kurra V, Schoenhagen P, Roselli EE, Kapadia SR, Tuzcu EM, Greenberg R, Akhtar M, Desai MY, Flamm SD, Halliburton SS, Svensson LG, Sola S. Prevalence of significant peripheral artery disease in patients evaluated for percutaneous aortic valve insertion: preprocedural assessment with multidetector computed tomography. J Thorac Cardiovasc Surg. 2009;137:1258–64. 58. Hayashida K, Lefevre T, Chevalier B, Hovasse T, Romano M, Garot P, Mylotte D, Uribe J, Farge A, Donzeau-Gouge P, Bouvier E, Cormier B, Morice MD.  Transfemoral aortic valve implantation new criteria to predict vascular complications. JACC Cardiovasc Interv. 2011;4:851–8. 59. Toggweiler S, Gurvitch R, Leipsic J, Wood DA, Willson AB, Binder RK, Cheung A, Ye J, Webb JB.  Percutaneous aortic valve replacement: vascular outcomes with a fully percutaneous procedure. J Am Coll Cardiol. 2012;59:113–8. 60. Krishnaswamy A, Parashar A, Agarwal S, Modi DK, Poddar KL, Svensson LG, Roselli EE, Schoenhagen P, Tuzcu EM, Kapadia SR. Predicting vascular com-

P. Schoenhagen et al. plications during transfemoral transcatheter aortic valve replacement using computed tomography: a novel area-based index. Catheter Cardiovasc Interv. 2014;84:844–51. 61. Schoenhagen P, Zimmermann M, Falkner J.  Advanced 3-D analysis, client-server systems, and cloud computing-integration of cardiovascular imaging data into clinical workflows of transcatheter aortic valve replacement. Cardiovasc Diagn Ther. 2013;3:80–92. 62. Schoenhagen P, Falkner J, Piraino D.  Transcatheter aortic valve repair, imaging, and electronic imaging health record. Curr Cardiol Rep. 2013;15:319. 63. Cowie MR, Chronaki CE, Vardas P. E-Health innovation: time for engagement with the cardiology community. Eur Heart J. 2012;34(25):1864–8. 64. Fernandez-Bayó J.  IHE profiles applied to regional PACS. Eur J Radiol. 2011;78:250–2. 65. Leon MB, Piazza N, Nikolsky E, Blackstone EH, Cutlip DE, Kappetein AP, Krucoff MW, Mack M, Mehran R, Miller C, Morel MA, Petersen J, Popma JJ, Takkenberg JJ, Vahanian A, van Es GA, Vranckx P, Webb JG, Windecker S, Serruys PW. Standardized endpoint definitions for transcatheter aortic valve implantation clinical trials: a consensus report from the Valve Academic Research Consortium. J Am Coll Cardiol. 2011;57:253–69. 66. Genereux P, Head SJ, Van Mieghem NM, Kodali S, Kirtane AJ, Xu K, Smith C, Serruys PW, Kappetein AP, Leon MB.  Clinical outcomes after transcatheter aortic valve replacement using valve academic research consortium definitions: a weighted meta-­ analysis of 3,519 patients from 16 studies. J Am Coll Cardiol. 2012;59:2317–26.

Three-Dimensional Rotational Angiography

27

Konstantin von Aspern and Lukas Lehmkuhl

27.1 Introduction

27.2 Technical Aspects

Due to the complex nature of vascular pathologies, accurate anatomical understanding of a target vessel in relation to surrounding structures is crucial for treatment planning and execution [1]. Since images produced through routine two-­dimensional fluoroscopy may provide an incomplete geometrical representation of cardiovascular structures [2], new three-­dimensional imaging techniques were developed to overcome these limitations. Threedimensional rotational angiography (3D-RA) produces cross-­ sectional images comparable to computed tomography (CT), which can be used for further 3D reconstruction and could help to overcome these limitations. Rotational angiography is an X-ray-based imaging technique using a C-arm-­ mounted flat-panel detector to generate fluoroscopic images from multiple angles around the patient. It is also known as flat-panel CT or cone-beam CT. Its versatility makes it an attractive modality applicable also intraprocedurally in a multidisciplinary setting [3].

So far a variety of different products using 3D-RA and their respective reconstruction software are commercially available, such as the syngo DynaCT™ (Siemens), Innova CT HD™ (GE Healthcare), INFX+CT™ (Toshiba), XperCT™ (Philips), and Safire 3D-C™ (Shimadzu). Image acquisition by 3D-RA is performed using intra-arterial contrast agent application in most clinical scenarios. A set of digital images is acquired at equiangular intervals along a circular arc of the C-arm around a fixed reference point (isocenter). The raw data provides cross-sectional submillimeter images consisting of isotropic voxels. These voxels are the basis for further multiplanar or three-dimensional reconstruction. Each image slice has comparable features to CT images but with different resolution properties (Fig.  27.1). Usually 3D-RA focuses on angiographic imaging with strong delineation of the contrast medium but a reduced visualization of low-contrast tissue and calcified structures due to the susceptibility to X-ray scatter of area detectors and cone-beam geometry and different settings of tube voltage and current. Various strategies are employed to optimize the reduced low-contrast visibility including anti-scatter grids, air gaps, wedge-shaped beam compensating filters, and software corrections [2].

K. von Aspern Department of Cardiac Surgery, University Heart Center Leipzig, Leipzig, Germany L. Lehmkuhl (*) Clinic for Radiology, Cardiovascular Center Bad Neustadt, Bad Neustadt, Germany e-mail: [email protected]; [email protected]

© Springer-Verlag GmbH Austria, part of Springer Nature 2019 O. H. Stanger et al. (eds.), Surgical Management of Aortic Pathology, https://doi.org/10.1007/978-3-7091-4874-7_27

427

K. von Aspern and L. Lehmkuhl

428

a

b

Fig. 27.1 Coronal multiplanar image reconstruction of the aortic root and ascending aorta in the same patient with severe aortic stenosis. (a) 3D-RA with 20  mL diluted contrast medium administration via pigtail catheter in the ascending

aorta. (b) CT with intravenous administration of 70 mL nondiluted contrast agent. Note the different visualizations of adjacent structures (left ventricle, right atrium, pulmonary artery) and of aortic and coronary artery calcification

The amount and dilution ratio of contrast medium, the overall numbers of projections, and the tube current necessary to generate satisfactory image quality vary considerably depending on the area of interest and planned procedure. Typically a bolus of diluted contrast medium is administered via a catheter positioned in the vicinity of the area of interest and followed by a 180°–240° circular rotation of the C-arm at 30–60 degrees per second. Raw data is then transferred onto a dedicated 3D workstation equipped with reconstruction software. While standard angiographic equipment is available in most interventional centers, not every system and protocol is equally suitable for all possible applications.

crucial for accurate valve positioning and to avoid side effects such as severe paravalvular leakage, valve dislocation, coronary artery obstruction, annulus rupture, or aortic dissection [4]. The use of 3D-RA during TAVI focuses mainly on correct perpendicular angulation of the C-arm and correct positioning of the valve to avoid repetitive contrast medium application through online overlay of reconstructed images [4, 5].

27.3 Clinical Application in Cardiac and Vascular Procedures 27.3.1 3D-RA in Transcatheter Aortic Valve Implantation (TAVI) Transcatheter aortic valve implantation (TAVI) strongly depends on reliable imaging. Best knowledge of the individual anatomic morphology is

27.3.2 Procedural Protocol for TAVI During 3D-RA for TAVI, the patient is placed with the arms to the side of the body. Typically a 6 Fr pigtail catheter for contrast medium application is positioned in the non-coronary cusp of the aortic root, and a balloon-tipped pacing catheter is placed into the right ventricle. Injection of diluted contrast medium (0.3–0.8 per kg diluted 1:2–3, flow rate 15–25 mL per second) followed by a 180°–220° circular rotation of the C-arm (1  s prescan delay) is performed. Alternatively, undiluted contrast agent at slower injection rates may be administered (e.g., 30  mL at 8  mL/s). Images are acquired during inspiratory breathhold to reduce respiratory motion and subsequent artifacts. Immediately before and during contrast

27  Three-Dimensional Rotational Angiography

injection, rapid ventricular pacing (RVP) at 180– 200 beats per minute is instituted. Image acquisition is completed within 5  s (60  frames/s, ≈70–125  kV), generating 240–320 projections. Depending on the reconstruction software abilities, the aortic valve cusps are automatically depicted, and a circle is generated at the nadir to indicate the aortic valve plane. A volume-rendered 3D model of the aortic root may be generated and used as an overlay to guide the procedure. Typically a simplified delineation of the aortic root is used as a real-time mask for perpendicular angulation. A suitable cusp alignment, favored by many interventionists, is where the non-coronary cusp appears on the left and the left coronary cusp on the right (Fig. 27.2).

27.3.3 Measurement Accuracy and Annulus Sizing Apart from achieving the ideal C-arm angulation for prosthesis implantation, 3D-RA has also been used for measuring aortic structures, particularly the aortic valve annulus diameter and its distance to the coronary ostia [4–6]. Few

a

Fig. 27.2 (a) 3D-RA of the aortic root and ascending aorta (maximum intensity projection). (b) Software-based semiautomatic evaluation of the 3D-RA data in the same

429

studies have compared agreement and reproducibility of measurements obtained by 3D-RA with multislice computed tomography (MSCT). Results showed that measurements of supraannular structures—like the sinotubular or ascending aortic diameters—were without significant difference between both methods; however, measurements of the aortic annulus by 3D-RA differed significantly and were less reproducible in comparison with MSCT [4]. A reason contributing to this difference is the method of contrast application during 3D-RA via a catheter in the aortic sinus. With the contrast medium entirely situated above the aortic valve, adequate depiction of the left ventricular outflow tract is difficult, leading to a lack of visible reference points and subsequently vague bordering of the aortic annulus. Recently the method of left ventricular contrast injection has been investigated and found feasible with measurement accuracy and reliability in good agreement with MSCT [7]. Besides choosing a suitable prosthesis size— taking into account native annulus dimensions, shape, and adjacent calcification—the correct implantation angle and plane play a pivotal role for

b

patient provides the aortic annulus size and its angulation prior to a TAVI procedure

430

a successful TAVI procedure. It has been shown that paravalvular leakage following TAVI is frequent and has a negative impact on survival [8, 9]. Utilization of 3D-RA in combination with additional add-on software, providing automatic root segmentation and anatomical landmark indication, has been found to be beneficial (Fig. 27.3). Analysis has shown that an excellent implant angle is significantly more likely to be achieved with the use of (semi)automatic reconstruction software. Furthermore, non-­excellent implant angles were more often associated with postprocedural paravalvular leakage, concluding that optimizing implant angles may be important in reducing paravalvular leakage after TAVI [10, 11].

27.3.4 Practical Considerations and Limitations Transcatheter aortic valve implantation has been shown to improve hospitalization rates and mortality in patients deemed inoperable [12]. Among others, an important aspect for a patients’ eligibility for TAVI—and the decision whether to choose a retrograde or antegrade approach—is dependent on preoperative imaging to assess vascular anatomy, dimensions, and calcifications.

Fig. 27.3 Simulation of an aortic valve prosthesis implantation based on an individual 3D-RA dataset to ensure a correct prosthesis choice

K. von Aspern and L. Lehmkuhl

In theory, provided that a modern interventional hybrid operating room is available, 3D-RA offers the advantage to intraprocedurally plan the arterial access site, perform aortic annulus sizing, and correct C-arm angulation and prosthesis implantation without the need for transferring the patient to a dedicated radiology unit. A recent study concluded that 3D-RA is equal to MSCT in accurately depicting the iliofemoral arterial caliber and tortuosity; however, it is inferior in the assessment of calcification [13]. A study examining feasibility and accuracy of 3D-RA in pretreatment evaluation of aortic aneurysms found that assessment of the iliac arteries was suboptimal due to a limited imaging volume size [14]. Regarding the abovementioned influence preoperative imaging has on patient selection ­ prior to TAVI and the limitations this imaging modality poses, intraprocedural 3D-RA alone may not sufficiently answer all key questions necessary for decision-making. When using 3D-RA during a standard TAVI procedure, radiation protection represents an issue. Usually scatter radiation coming from the bottom is shielded against by radiopaque covering at the side of the operating table. This protection, however, needs to be dismantled during circular image acquisition around the patient. After 3D-RA is performed, the covering needs to be reattached, which can pose tremendous difficulties, given that the operating field is already covered and sterile. Another issue concerning 3D-RA image quality is associated with breathing motion of the non-intubated patient when in horizontal position. Although breath-hold image acquisition is preferred, most patients are not capable to comply due to respiratory distress caused by the underlying pathology. Modern hybrid operating rooms have standard imaging equipment readily available, including C-arm-mounted flat-panel detectors for angiography and 3D-RA.  Depending on the equipment used, detector panel sizes and ultimately the effective maximum field of view (FOV) vary considerably. A too small FOV may result in inadequate depiction of the entire heart and the ascending aorta and should be considered when planning a hybrid OR.

27  Three-Dimensional Rotational Angiography

27.4 3D-RA in Endovascular Aneurysm Repair (EVAR) Accurate assessment of anatomical structures is mandatory for successful planning and execution of endovascular aneurysm repair (EVAR). To date multislice computed tomography (MSCT) is the preferred imaging modality and emerged as the gold standard prior to EVAR [15]. Only recently ontable 3D-RA has been used as an alternative and/or complementary to MSCT for EVAR of the abdominal aorta with promising initial results [14, 16]. Particularly in emergency EVAR due to ruptured aneurysms, periprocedural 3D-RA may avoid the diagnostic delay of pre-procedural MSCT, which needs to be performed outside the intervention lab or hybrid operating room. In comparison to conventional open aortic aneurysm repair, EVAR is reported to be associated with an increased frequency of secondary interventions, mandating regular follow-up imaging [15]. It has been suggested that early re-interventions may be a reflection of inadequate detection and immediate treatment of intraprocedural complications [15]. Whether 3D-RA during EVAR has the potential to improve this issue is subject to ongoing research with promising initial results [15, 17, 18].

27.4.1 Procedural Protocol for EVAR Depending on the extent and individual features of the aortic aneurysm and the respective center experience, 3D-RA protocols vary with regard to catheter placement, contrast medium, and detector configuration. Typically a 4 Fr diagnostic catheter is introduced via a femoral artery and positioned at the level of or slightly above the renal arteries. Between 90 and 140 ml, diluted contrast medium (dilution ratio 1:1) is injected at 6–9 mL per second. After a prescan delay of 3–4 s, images are acquired during breath-hold. Within 6–8 s, a circular C-arm rotation of 180–220 degrees with 15–50 frames per second is performed, registering between 120 and 420 projections. These ­cross-­sectional images are then sent to a dedi-

431

cated workstation for reconstruction [17].The three-­ dimensional volumetric reconstructions can then be used as a roadmap overlay on live fluoroscopy and synchronized with C-arm and table positions to reduce repetitive contrast agent application and total radiation dose [17].

27.4.2 Feasibility and Accuracy Choosing a suitable aortic stent graft as well as its correct positioning is an important aspect for successful EVAR.  Absence of procedural complications such as endoleaks and graft kinking, verified by an uneventful follow-up CT, is associated with improved outcome [15]. Initial experience with 3D-RA during EVAR has shown that it provides sufficient information for determining the correct treatment and selecting the proper stent graft before EVAR [14]. It represents a feasible intraoperative adjunct to completion angiography, which improves intraoperative quality control during endovascular repair of abdominal aortic aneurysms. To date MSCT represents the gold standard in planning EVAR as well as follow-up examination. Recently 3D-RA in conjunction with the completion angiogram has been trailed against early follow-up MSCT with regard to detection of procedural complications. In view of the results, it was suggested that 3D-RA, as a feasible imaging method both in EVAR planning and as completion imaging to detect complications missed by conventional angiography, could replace early follow-up MSCT, potentially reducing overall radiation and contrast use [15, 19]. It remains unclear whether the 3D-RA is sufficiently accurate for the detection of low-flow endoleaks compared to modern scanning protocols that allow a dynamic MSCT [20].

27.4.3 Limitations and Practical Improvements Although initial experience using 3D-RA during EVAR is promising, certain practical limitations have been described. A major issue poses the lim-

432

ited FOV resulting from the detector panel size and orientation. Typically multiple roadmaps of the entire aorta and target vessels are needed for planning and execution [17]. The orthogonal orientation of the detector (horizontal mode) results in a further reduction of the effective FOV so that the stent graft usually is not completely included [15]. In order to tackle this issue, a group of researchers updated the 3D-RA software, switching from horizontal to vertical detector mode (90° rotation), thus being able to depict a larger portion of the region of interest [15]. Imaging artifacts due to scatter have also been described to be a limitation of 3D-RA. Typically radiopaque guide wires are used during the procedure. Exchanging these for regular catheters further reduces image artifacts and should be considered whenever possible prior to 3D-RA image acquisition.

27.5 Summary Three-dimensional rotational angiography has evolved to be an additional valuable imaging tool to optimize procedural success also in endovascular treatment and cardiothoracic surgery. It can help to overcome the limitations of two-dimensional fluoroscopy and the hereby associated incomplete geometrical representation of cardiovascular structures. Its versatility makes it an attractive modality applicable intraprocedurally in a multidisciplinary setting. Despite a promising perspective for the 3D-RA, there are numerous limitations, particularly with regard to the size of the scan field and for reasons of practicality in a sterile environment.

References 1. Namba K, Niimi Y, Song JK, Berenstein A.  Use of dyna-CT angiography in neuroendovascular decision-­ making. A case report. Intervent Neuroradiol. 2009;15:67–72. 2. Lanzer P.  Catheter-based cardiovascular interventions. Berlin: Springer; 2013. 3. Pamir MN, Seifert V, Kiris T. Intraoperative imaging. In: Tonn JC, editor. Intraoperative computed tomography, vol. 168. Wien: Springer; 2011.

K. von Aspern and L. Lehmkuhl 4. Lehmkuhl LH, von Aspern K, Foldyna B, Grothoff M, Nitzsche S, Kempfert J, Rastan A, et al. Comparison of aortic root measurements in patients undergoing transapical aortic valve implantation (TA-AVI) using three-dimensional rotational angiography (3D-RA) and multislice computed tomography (MSCT): differences and variability. Int J Cardiovasc Imaging. 2013;29:417–24. 5. Kempfert J, Falk V, Schuler G, Linke A, Merk D, Mohr FW, Walther T.  Dyna-CT during minimally invasive off-pump transapical aortic valve implantation. Ann Thorac Surg. 2009;88:2041. 6. Zheng Y, John M, Liao R, Boese J, Kirschstein U, Georgescu B, Zhou SK, Kempfert J, Walther T, Brockmann G, Comaniciu D.  Automatic aorta segmentation and valve landmark detection in C-arm CT: application to aortic valve implantation. Med Image Comput Comput Assist Interv. 2010;13:476–83. 7. Balzer JC, Boering YC, Mollus S, Schmidt M, Hellhammer K, Kroepil P, Westenfeld R, Zeus T, Antoch G, Linke A, Steinseifer U, Merx MW, Kelm M.  Left ventricular contrast injection with rotational C-arm CT improves accuracy of aortic annulus measurement during cardiac catheterisation. EuroIntervention. 2014;10:347–54. 8. Athappan G, Patvardhan E, Tuzcu EM, Svensson LG, Lemos PA, Fraccaro C, et al. Incidence, predictors, and outcomes of aortic regurgitation after transcatheter aortic valve replacement: meta-analysis and systematic review of literature. J Am Coll Cardiol. 2013;61:1585–95. 9. Kodali S, Pibarot P, Douglas PS, Williams M, Xu K, Thourani V, et  al. Paravalvular regurgitation after transcatheter aortic valve replacement with the Edwards sapien valve in the PARTNER trial: characterizing patients and impact on outcomes. Eur Heart J. 2015;36:449–56. 10. Gurvitch R, Wood DA, Leipsic J, Tay E, Johnson M, Ye J, Nietlispach F, Wijesinghe N, Cheung A, Webb JG.  Multislice computed tomography for prediction of optimal angiographic deployment projections during transcatheter aortic valve implantation. JACC Cardiovasc Interv. 2010;3:1157–65. 11. Poon KK, Crowhurst J, James C, Campbell D, Roper D, Chan J, et al. Impact of optimising fluoroscopic implant angles on paravalvular regurgitation in transcatheter aortic valve replacements—utility of three-dimensional rotational angiography. EuroIntervention. 2012;8:538–45. 12. Kodali SK, Williams MR, Smith CR, Svensson LG, Webb JG, Makkar RR, et  al. Investigators PT.  Two-­ year outcomes after transcatheter or surgical aortic-­ valve replacement. N Engl J Med. 2012;366:1686–95. 13. Crowhurst JA, Campbell D, Raffel OC, Whitby M, Pathmanathan P, Redmond S, et  al. Using DynaCT for the assessment of Ilio-femoral arterial calibre, calcification and tortuosity index in patients selected for trans-catheter aortic valve replacement. Int J Cardiovasc Imaging. 2013;29:1537–45. 14. Eide KR, Odegard A, Myhre HO, Hatlinghus S, Haraldseth O. DynaCT in pre-treatment evaluation of

27  Three-Dimensional Rotational Angiography aortic aneurysm before EVAR. Eur J Vasc Endovasc Surg. 2011;42:332–9. 15. Tornqvist P, Dias N, Sonesson B, Kristmundsson T, Resch T. Intra-operative cone beam computed tomography can help avoid reinterventions and reduce CT follow up after infrarenal EVAR. Eur J Vasc Endovasc Surg. 2015;49:390–5. 16. Eide KR, Odegard A, Myhre HO, Lydersen S, Hatlinghus S, Haraldseth O.  DynaCT during EVAR—a comparison with multidetector CT.  Eur J Vasc Endovasc Surg. 2009;37:23–30. 17. Tacher V, Lin M, Desgranges P, Deux JF, Grunhagen T, Becquemin JP, Luciani A, Rahmouni A, Kobeiter H.  Image guidance for endovascular repair of complex aortic aneurysms: comparison of two-dimensional and three-dimensional angiography and image fusion. J Vasc Interv Radiol. 2013;24:1698–706.

433 18. Kim DH, Rha SW, Kook H, Kim W, Lee SK, Oh SK, Choi CU, Oh DJ.  Three-dimensional angiography-­ guided percutaneous transluminal angioplasty for distal aorta and bi-iliac chronic total occlusion. Korean Circ J. 2013;43:261–4. 19. Biasi L, Ali T, Ratnam LA, Morgan R, Loftus I, Thompson M.  Intra-operative DynaCT improves technical success of endovascular repair of abdominal aortic aneurysms. J Vasc Surg. 2009;49:288–95. 20. Lehmkuhl L, Andres C, Lucke C, Hoffmann J, Foldyna B, Grothoff M, et  al. Dynamic CT angiography after abdominal aortic endovascular aneurysm repair: influence of enhancement patterns and optimal bolus timing on endoleak detection. Radiology. 2013;268:890–9.

4D Flow MR: Insights into Aortic Blood Flow Characteristics

28

Florian von Knobelsdorff-Brenkenhoff and Alex J. Barker

28.1 What Is 4D Flow MR? 4D flow MRI is an extension of the well-­established 2D phase contrast sequence commonly used to measure blood flow velocities in the vasculature. In its simplest form, the 2D flow approach is used clinically to measure unidirectional blood flow through imaging planes placed perpendicular to the long axis of a vessel. 4D flow MRI is a further extension of the technique that does not only collect velocity data though, or in the imaging plane, but rather in all three principal directions (x, y, z) over a volumetric field of view. The volumetric acquisition, which includes a spatially and temporally resolved three-­directional velocity field, is commonly referred to in the literature as 4D flow MRI [1]. Regardless of the implementation (4D or 2D flow), phase-­contrast MRI is rooted in the use of a bipolar magnetic gradient to cause proton spin phase shifts proportional to their displacement over time (in the case of blood, the phase shifts are caused by water protons). Given that these phase F. von Knobelsdorff-Brenkenhoff (*) Department of Cardiology, Clinic Agatharied, Academic Teaching Hospital, Ludwig-Maximilians-University Munich, Norbert-Kerkel-Platz, Munich, Germany e-mail: [email protected] A. J. Barker Department of Radiology, Feinberg School of Medicine, Northwestern University, Chicago, IL, USA e-mail: [email protected]

shifts are acquired during a specific time window, knowledge of the bulk proton displacement within a voxel can be used to obtain a spatially resolved velocity field of the corresponding fluid. When combined with ECG gating, time-resolved velocities can be derived on a voxel-by-voxel basis, and their temporal evolution can be represented over one (virtual) heart cycle. Integration of the temporal and spatial velocity field over the cross section of the target vessel allows for the calculation of blood flow rate and volume. This approach is commonly used in clinical routine, for example, with 2D flow, to quantify aortic regurgitation volumes or the ratio of pulmonary to systemic stroke volume to diagnose a shunt [2]. The same approach as that used with 2D flow can be used with a 4D flow acquisition, but with additional coverage of the adjacent vasculature. Similar results have been found at both 1.5 and 3 T [3], although the signal-to-noise ratio is higher at 3 T with visibly better image contrast. As a general rule of thumb, contrast agents are not required but are known to enhance the intraluminal signal and improve image quality and velocity to noise ratio [4]. For planning, a 3D field of view is typically prescribed to cover the targeted area, such as one covering the thoracic aorta. During acquisition, data are collected with ECG gating to correct for cardiac movement and to order the k-space lines with respect to the cardiac cycle. Respiratory motion is commonly compensated using a navigator technique. One approach is to continuously

© Springer-Verlag GmbH Austria, part of Springer Nature 2019 O. H. Stanger et al. (eds.), Surgical Management of Aortic Pathology, https://doi.org/10.1007/978-3-7091-4874-7_28

435

F. von Knobelsdorff-Brenkenhoff and A. J. Barker

436

monitor the movement of the diaphragm at end diastole and acquire image data at a predefined acceptance window for the diaphragm location. As a result, respiratory motion is minimized by limiting the acquisition to a small range of acceptable positions, which will be used to reconstruct the final images, while the remainder of the data is rejected. Depending on the size of the field of view, efficiency of the navigator, heart rate, and acceleration mode of the sequence, image acquisition generally takes 8–20  min. The resulting image dataset typically contains magnitude images depicting the anatomy, and three phase -contrast datasets that represent the principal velocity directions, which when combined, represents the 3D flow field (Fig. 28.1). Typical imaging parameters for the thoracic aorta include a spatial resolution of 2–2.5  mm3 and a temporal resolution of 35–50  ms [5]. To visualize and extract quantitative flow information, the dataset requires further post-processing, which is accomplished with a diverse set of software solutions, and is dependent on user requirements. For exam-

ple, most post-processing solutions allow for flow visualization of both the blood flow velocity and direction using color-coded pathlines or streamlines (Fig.  28.2). Qualitatively, streamline and pathline visualization aids in the ability to localize complex flow phenomena, such as identifying the presence of helices and vortices caused by abnormalities in vascular structure (Fig.  28.3) [6]. Furthermore, the quantification of flow velocity as well as a variety of derived parameters like flow volumes, wall shear stress, pulse wave velocity, and energy loss is possible; however some of these parameters may require specialized software or programming expertise. Nonetheless, most analysis software will allow for the user to take advantage of the main benefit of 4D flow MRI, that is, the ability to retrospectively prescribe imaging measurement planes in the 3D dataset. In contrast, the traditional approach of 2D flow is limited to the original predefined imaging planes (Fig.  28.2). While image acquisition and data post-processing of 4D flow data could be time-consuming in the past, recent software and

Vx

Vy

Vz

velocity [cm/s] 100 50 0 -50 -100

Raw data

Fig. 28.1  Acquisition planning. A 3D volume (field of view) is selected to cover the target vessel. The 4D flow raw data comprises of magnitude images reflecting the

anatomy and phase images reflecting flow velocities along all three spatial dimensions, three velocity directions, and time in the cardiac cycle (adapted from Markl et al. [5])

28  4D Flow MR: Insights into Aortic Blood Flow Characteristics

a

2D Flow

437

b 4D Flow

magnitude (anatomy)

** **

**

Velocity [m/s] 1.20 0.90 ** 0.60 0.30 0.00

**

phase (through-plane velocity)

** **

* ** Fig. 28.2 (a) Visualization of the output from a standard 2D flow acquisition in comparison to (b) a 4D flow acquisition in a normal healthy volunteer (the 2D flow acquisition plane is illustrated on the 4D flow volume with the “*” label). The additional cross-sectional planes (labeled “**”) illustrate the ability to retrospectively interrogate the velocity field anywhere in the imaging field of view.

The color-coded velocity streamlines represent the velocity magnitude of the characteristic systolic flow pattern in the thoracic aorta. For the healthy aorta and aortic valve, the streamlines are cohesive and oriented in the longitudinal direction of the vessel during systole. Little evidence of vortical or helical flow is visible at this time point (adapted from Barker et al. [12])

hardware developments are quickly enabling the workup of 4D flow data in a time window appropriate for the clinical setting. In this context, the 2014 European Society for Cardiology guidelines for aortic diseases comment that 4D flow provides the unique opportunity to visualize and measure blood flow patterns and highlights that quantitative parameters such as pulse wave velocity and wall shear stress can be determined [7].

ture, converging slightly toward the outer wall. During early diastole, retrograde flow occurs along the inner curvatures of both the ascending and proximal descending arch, which may contribute to diastolic filling of the coronary arteries [9]. In preliminary examinations of flow in the normal aorta, estimates of wall shear stress (the tangential frictional force at the vessel wall) have been reported. This force, which is computed using the blood velocity gradient at the wall as obtained from 4D flow MRI and an assumption of constant blood viscosity, is a known mechanotransduction factor affecting atherosclerosis, cellular signaling, and vascular remodeling [10]. Of particular interest is the range of normal wall shear stress expressed in the healthy individual as a function of vascular region. For example, in one such study, the mean absolute time-averaged wall shear stress ranged between 0.25 ± 0.04 N/m2 and 0.33  ±  0.07  N/m2 and incorporated a substantial circumferential component (−0.05  ±  0.04 to

28.2 4D Flow in the Normal Aorta During systole, the high-velocity ejection jet tends to migrate to the outer curvature, as visualized by peak velocity streamlines. Later in systole, the streamlines curve posterolaterally, back toward the inner curvature in a right-handed helix in the ascending aorta and arch [8]. In the proximal descending aorta, velocities increase where streamlines tend to separate from the inner curva-

438

F. von Knobelsdorff-Brenkenhoff and A. J. Barker

Fig. 28.3  3D streamline visualization of thoracic aortic systolic blood flow as assessed by 4D flow in a patient with a bicuspid aortic valve with fusion of the right- and left-coronary leaflets and aortic coarctation at the proximal descending aorta. There is a posteriorly directed, high-velocity flow jet in the ascending aorta (AAo) with

associated right-handed helix formation. Complex aortic geometry near the coarctation results in vortex formation proximal to the coarctation, a right-handed helix distal to the coarctation, and flow acceleration through the aortic narrowing (adapted from Markl et al. [5])

0.07 ± 0.02 N/m2) [11]. At each level of the thoracic aorta, a regional location was identified with the lowest absolute wall shear stress and highest oscillatory shear index, which differed significantly from the mean values within the plane. Interestingly, the observed distribution of atherogenic low wall shear stress and high oscillatory shear index closely resembled typical locations of atherosclerotic lesions at the inner aortic curvature and supraaortic branches. Additional independent 4D flow studies involving healthy individuals have corroborated these insightful wall shear stress findings [12–15]. As a result, data collected with the 4D flow technique have provided compelling evidence to corroborate the hemodynamic hypothesis that low and oscillating wall shear stress promotes the development and progression of regional atherosclerotic lesions in the aorta. Additional 4D flow post-processing techniques have demonstrated the ability to measure the propagation of the systolic flow wavefront along the length of the aorta. The velocity of the flow wavefront,

known as pulse wave velocity, is a known marker for global vascular compliance and has shown to be feasibly estimated using 4D flow data [16, 17]. The 4D flow approach to estimate pulse wave velocity is a promising method to investigate the change of vascular compliance in the presence of aging, atherosclerosis, and cardiovascular disease.

28.3 4D Flow in the Dilated Aorta The guidelines for aortic disease and valvular disease cite aortic dilatation as a risk factor for aortic dissection [18, 19]. The concept of aorta size as a risk factor for dissection is somewhat controversial since, on an individual basis, aortic dissection occurs in subjects with normal aorta sizes, and the majority of patients with aortic dilatation will never suffer from aortic dissection. Therefore, personalized measures to improve the understanding of aortic remodeling and to risk-­ stratify beyond diameter thresholds are desired to improve

28  4D Flow MR: Insights into Aortic Blood Flow Characteristics

patient care. With this in mind, 4D flow has been explored in patients with aortic dilatation to identify abnormal flow patterns and wall shear stress distribution. For example, Bieging et al. compared patients with aortic dilatation to healthy controls and found that ascending aortic dilatation was associated with increased diastolic wall shear stress, decreased systolic to diastolic wall shear stress ratio, and delayed onset of peak wall shear stress. In addition, temporally averaged wall stress was increased, and peak systolic wall shear stress was decreased. The maximum wall shear stress in patients with aortic dilatation was on the anterior wall of the ascending aorta. Vortical flow with highest velocities along the anterior wall and

Blood flow velocity

a

v (m/s) 2

439

increased helical flow during diastole were observed in patients [13]. Bürk et al. also studied patients with dilated ascending aorta and compared them to age-matched subjects. Thereby, the incidence and strength of supraphysiologic-­helix and vortex flow in the ascending aorta were significantly higher in patients with dilated ascending aorta than in controls. Interestingly, the extent and incidence of ascending aortic helix and vortex flow were associated with significant differences in ascending aortic diameters. Peak systolic wall shear stress in the ascending aorta and aortic arch was significantly lower in patients with dilated ascending aorta (Fig. 28.4). The ascending aortic diameter positively correlated to time to peak sys-

v (m/s) 2

b

v (m/s)

c

4

1.5

1.5

3

1

1

2

0.5

0.5

1

0

0

0

d

e

f

Velocity profiles Wall shear stress WSS (Pa) 1.5

g

CONTROL

WSS (Pa) 1.5

WSS (Pa) 3

1

1

2

0.5

0.5

1

0

h

DILATION

Fig. 28.4  Visualization of blood flow velocity vectors in a typical control aorta (a), dilated aorta (b), and aorta with severe valve stenosis (c). Cross-sectional velocity profiles are shown at the level of the black lines in (d), (e), and (f). The

0

i

STENOSIS

0

corresponding 3D WSS pattern is shown below these subjects (g), (h), and (i). Note that the velocity and WSS color bar for the stenosis subject are two times higher than for the other two examples (adapted from van Ooij et al. [22])

440

tolic velocities and oscillatory shear index and inversely correlated to peak systolic wall shear stress. Peak systolic wall shear stress was significantly lower in ascending aortic aneurysms at the right and outer curvature within the ascending aorta and proximal arch [15]. These findings indicate that specific hemodynamic alterations exist in the dilated aorta compared to the normal aorta. In the future, these insights may help elucidate specific flow patterns that increase risk for adverse cardiovascular events, such as dissection.

28.4 4D Flow in Aortic Stenosis Post-stenotic aortic dilation has been linked to altered mechanical stress on the wall of the ascending aorta caused by flow disturbance downstream of a stenotic lesion [20]. Normal systolic flow in the ascending aorta is cohesive, with fastest flow in the vessel center, and shear stress evenly distributed around the aortic circumference. In patients with post-stenotic dilation, however, systolic flow is often eccentric and displaced from the centerline toward the vessel wall and follows a helical path through the ascending aorta. Consequently, wall shear stress is asymmetrically elevated. This pattern was observed in subjects with aortic stenosis both with normal and with dilated ascending aorta, which supports the argument that the abnormal flow is caused by the valve itself and not by the dilated aorta (Fig. 28.4) [21, 22]. The preceding descriptions focus on blood flow downstream from the aortic valve and provide general information regarding the impact of aortic stenosis on flow pattern abnormalities and flow environment near the ascending aorta wall. On the ventricle side of the valve, blood flow caused by aortic stenosis will have an upstream impact on function and cardiac burden. One parameter that may gain importance in this regard is the measurement of irreversible energy losses caused by frictional effects across a valvular obstruction and altered downstream flow. These losses exist in both laminar and turbulent flow regimes, with the turbulent regime typically being an order of magnitude greater compared to

F. von Knobelsdorff-Brenkenhoff and A. J. Barker

the laminar regime. As a result, recent efforts have attempted to estimate turbulent kinetic energy from 4D flow datasets noninvasively to detect regions of elevated flow turbulence and thus irreversible energy loss. This is important as irreversible energy loss (manifested as pressure loss) in post-stenotic flow is an important determinant of the hemodynamic significance of aortic stenosis and cardiac afterload. Post-stenotic energy loss is largely caused by dissipation of turbulent kinetic energy into heat and manifests in the clinic as a measurable pressure loss during invasive catheterization. In a recent study, Dyverfeldt et  al. measured the turbulent kinetic energy in patients with aortic stenosis. It was significantly higher in stenosis patients than in normal volunteers, and the peak total turbulent kinetic energy in the ascending aorta was strongly correlated to indexed pressure loss as obtained by echocardiography [23]. Similarly, using 4D flow datasets, Barker et al. quantified the laminar component of viscous energy loss in the ascending aorta and found elevated losses in patients with aortic stenosis compared with healthy volunteers. In addition, those patients with dilated aortas and no valve disease displayed significantly higher viscous losses than healthy individuals, although lower than aortic stenosis patients. These data reinforce the concept that cardiac afterload is increased due to abnormal flow, aortic size, and valve morphology in these subjects [24].

28.5 4  D Flow in Bicuspid Aortic Valve Bicuspid aortic valve disease is associated with ascending aortic dilatation and increased risk of aortic dissection [7]. This association is attributed to a genetic predisposition leading to malformation of the valve and the aortic wall. Furthermore, unfavorable shear forces near the vessel wall can change endothelial function and possibly create areas at risk for vascular remodeling. The permanent hemodynamic burden due to the abnormal valve geometry is thought to be a contributor to aortic abnormalities in patients with bicuspid aortic valve [25]. The latter argument has recently

28  4D Flow MR: Insights into Aortic Blood Flow Characteristics

441

been supported by studies using 4D flow to visualize and analyze the blood flow in the ascending aorta. Barker et al. measured the impact of bicuspid aortic valve disease on the distribution of regional aortic wall shear stress compared with age-/aorta size-controlled cohorts with tricuspid valves. Wall shear stress patterns in the ascending aorta of bicuspid aortic valve patients were significantly elevated, independent of stenosis severity (Fig.  28.5). The observation of right-anterior ascending aorta wall/jet impingement in patients corresponded to regions with elevated wall shear stress. Alternative jetting patterns were observed depending on the fusion type of the bicuspid aortic valve [12]. Meierhofer et  al. confirmed these results in another series [14]. In a study with almost a hundred of patients with bicuspid aortic

valve disease by Bissell et al., patients with bicuspid aortic valve had predominantly abnormal right-handed helical flow in the ascending aorta, larger ascending aortas, and higher rotational (helical) flow, systolic flow angle, and systolic wall shear stress compared with healthy volunteers. Bicuspid aortic valve with right-handed flow and right-­noncoronary cusp fusion showed more severe flow abnormalities and larger aortas than right-­left cusp fusion. Patients with bicuspid aortic valve with normal flow patterns had similar aortic dimensions and wall shear stress when compared to healthy volunteers. Younger patients with bicuspid aortic valve showed abnormal flow patterns, but no aortic dilation. Both of these observations support the potential importance of flow pattern in the pathogenesis of aortic dilation [26].

Fig. 28.5  Distribution of peak wall shear stress projected on a 3D segmentation of the aorta (left column) and wall shear stress vectors (right column) in planes placed at the regions of maximum wall shear stress for (a) bicuspid aortic valve (BAV) as compared to the same location in a healthy control (b) (adopted from [22]). Eccentric WSS at the sinus and proximal arch is illustrated for two phenotypes of BAV patients (a, c) as compared to healthy controls (b, d). One example is a BAV patient with a fusion of the right-left (BAV RL) coronary leaflets, and the other is a BAV patient with a fusion of the right and noncoronary (BAV RN) leaflets

a

WSS (Pa) 1.5 1.0 0.5 0

b

WSS (Pa) 1.5 1.0 0.5

c

0 WSS (Pa) 1.5 1.0 0.5 0 WSS (Pa) 1.5

d

1.0 0.5 0

442

Hope et  al. addressed the effect of abnormal blood flow in the ascending aorta on the aortic growth rate in patients with bicuspid aortic valve. They analyzed serial MR or CT studies. The growth rates of patients with bicuspid aortic valve were significantly higher than those of controls. Those patients with abnormal flow patterns demonstrated significantly higher growth rates than those with normal flow. The authors state that imaging biomarkers such as these could be used to identify and risk-stratify patients in whom clinically significant aortic disease is likely to develop [27]. Figure 28.3 shows a 4D flow case of bicuspid aortic valve disease that illustrates the potential of 4D flow to capture the impact of localized pathologies (bicuspid aortic valve disease, here in association with coarctation) on complex changes in aortic hemodynamics affecting the entire thoracic aorta. In addition, the complete volumetric coverage provides the user with the ability to identify the optimal location for retrospective quantification of clinically relevant parameters such as peak jet flow velocities distal to the bicuspid aortic valve and within the coarctation [5].

28.6 4  D Flow in Aortic Coarctation 4D flow can help evaluate collateral blood flow as a potential measure of hemodynamic significance in patients with aortic coarctation. Additionally, distorted flow patterns in the descending aorta after coarctation repair such as marked helical and vortical flow in regions of post-stenotic dilation were reported [28, 29]. This distorted pattern becomes even more marked in the presence of BAV.  In a recent study, 4D flow data were used to calculate pressure fields in patients with aortic coarctation, which showed a close agreement to catheterization as the clinical gold standard, and may eventually replace this invasive procedure in the future [30].

28.7 4D Flow After Aortic Surgery Postoperative 4D flow MRI can provide information regarding the ability of the surgeon to restore blood flow in the aorta to that representing a

F. von Knobelsdorff-Brenkenhoff and A. J. Barker

healthy physiologic flow pattern. Along these lines, 4D flow has been performed to analyze blood flow in the thoracic aorta of patients after valve-sparing aortic root replacement. In a study by Markl et al., 12 patients after David reimplantation using a cylindrical tube graft (T. David-I) and two versions of neosinus recreation (T.  David-V and T.  David-V-Smod) were included. Systolic vortices were seen in both coronary sinuses of all volunteers. Comparable coronary vortices were detected in all operated patients. Vorticity was minimal in the noncoronary cusp in T. David-I repairs but was prominent in T. David-V noncoronary graft pseudosinuses. Retrograde flow and helicity were found in all patients but were not distinguishable from normal values in the T. David-V-Smod patients [31]. Another study applied 4D flow to characterize the aortic blood flow in patients following valve-­ sparing aortic root replacement compared with presurgical cohorts matched by tricuspid and bicuspid valve morphology, age, and presurgical aorta size. They found that after valve-sparing aortic root replacement, the helical flow was reduced and less eccentric compared to presurgical control subjects, but there was a trend toward higher systolic flow acceleration as a surrogate measure of reduced aortic compliance [32].

28.8 4  D Flow After Aortic Valve Surgery Aortic remodeling after aortic valve replacement (AVR) may be influenced by postoperative blood flow patterns in the ascending aorta. The feasibility of 4D flow in the proximity of heart valve prosthesis has been demonstrated to perform reliably in in vitro flow phantoms [33]. In an in vivo pilot study, 4D flow was applied to describe ascending aortic flow characteristics after various types of AVR: mechanical prostheses, stented bioprostheses, stentless bioprostheses, and autografts. The study demonstrated that the flow characteristics in the ascending aorta after each type of AVR were different from native aortic valves and moreover differed between the various types of AVR.  Additionally, mechanical prostheses showed the most distinct vorticity compared to

28  4D Flow MR: Insights into Aortic Blood Flow Characteristics

443

Anterior (Outer Curvature) 2

2.5

12

2.0

10 5

Left

9 6

1.5

2.5

11

4

Right

Peak WSS (N/m2)

1

3

7

8

1.5

Posterior (Inner Curvature)

* * *

1.0

* * *

2.0

1.0 0.5

0.5

0.0

0.0 1

2

3

4

5

6

7

8

9 10 11 12

1

2

3

4

5

6

7

8

9 10 11 12

Fig. 28.6  Distribution of peak wall shear stress in the mid-ascending aorta after various types of aortic valve surgery (adapted from [34])

controls, while stented bioprostheses exhibited the most distinct helicity. Instead of physiologic central flow, all stented, stentless, and mechanical prostheses showed eccentric flow jets mainly directed toward the right-anterior aortic wall. Stented and stentless prostheses showed an asymmetric distribution of peak wall shear stress along the aortic circumference, with significantly increased local wall shear stress where the flow jet impinged on the aortic wall (Fig. 28.6). Local wall shear stress was higher in stented and stentless compared to autografts and controls. Autografts exhibited lower wall shear stress than controls [34]. A similar approach has been used in recipients of a transcatheter aortic valve implantation (TAVI). Thereby, TAVI and stented bioprostheses exhibited a similar pattern of asymmetric wall shear stress distribution in the ascending aorta, while the appearance of vortices and helices was stronger in those with stented bioprosthesis ­compared to TAVI [35].

ECG trigger time will cause signal loss or artifact in the corresponding regions. The presence of signal loss or artifacts is especially relevant for regions with high degrees of turbulence and strong intravoxel velocity variations (which can manifest as intravoxel dephasing). Besides turbulent flow, the typical temporal resolution of a 4D flow sequence is on the order of 40 ms, which can miss local peaks or fluctuations of wall shear stress. However, this disadvantage (compared to echocardiography) is balanced with the added ability to visualize the full magnitude of eccentric velocity jets, contrary to echocardiography, which is limited to resolving velocities collinear with the beamline. Finally, one must consider that the resolution of the acquisition is finite. Thus, the spatial resolution of 4D flow data is insufficient to resolve small-scale boundary layers or arterial velocity profiles in small vessels (1 PWRR and RRED in asymptomatic AAAs is associated with higher rupture risk

Significant positive correlation between WS and growth rate Significant positive correlation between shoulder WS and growth rate PWS moderately correlates to FDG uptake, which is a significant predictor of 2-year event-free survival NA

Outcomes NA

472 A. Nchimi et al.

30  Emerging Tools to Assess the Risk of Rupture in AAA: Wall Stress and FDG PET

473

the local diameter, (2) the local thickness of the ILT layer, and (3) local wall stress.

PET and its variants to be established as cornerstones of molecular imaging.

30.4.3 Limitations

30.5.1 Technique

Even though the findings regarding biomechanical parameters in the context of AAA are promising, there are some inherent limitations. A recent meta-analysis underlines the heterogeneity in retrospective patient selection, as some studies made a distinction between symptomatic non-­ ruptured and ruptured AAA, whereas other didn’t [79]. In studies evaluating FES before a clinical event occurs, there remain heterogeneities in the timing of CT relative to the time of rupture. Technically, FES relies on some standard assumptions, such as biomechanical properties of the AAA wall components on ex  vivo experiments. For example, the AAA wall thickness is set homogeneous and isotropic with 2 mm thickness with linear elastic properties in most simulations, whereas the wall properties have shown to be variable, especially in the presence of calcifications or ILT [48, 74]. Finally, the reproducibility of FES is a concern, owing to the existence of methods variability and several (commercially available) semiautomated softwares. Though good interand intra-observer reproducibilities of PWS and PWRI have been reported using a semiautomatic FES model [90, 91], data evaluating the inter-­ software variability remain missing. All these sources of variability need to be standardized before the thresholds provided to clinically determine the risk of rupture are widely used and implemented in the AAA repair surgery guidelines.

Positron emission tomography produces images from an internal source, by detecting via a circular gamma camera, the two opposite photons resulting from positron annihilation, after ­disintegration of a radionuclide, also known as a tracer. This tracer is often a molecule with a specific biodistribution, coupled with a positron emitter that enables imaging of some physiological or functional processes. Fludeoxyglucose F18 (18F-FDG, FDG) is one of the most popular tracers containing the radionuclide fluorine F-18 that has a half-life of 109.77  min. Similarly to glucose, FDG enters cells by the same membrane transporter and undergoes phosphorylation by hexokinase before being metabolized in FDG-6-­ phosphate. The latter cannot enter the cycle of hexoses (glycolysis) nor exit the cells as dephosphorylation by glucose-6-phosphatase is minimal. FDG-6-phosphate therefore accumulates into the cells [92]. This accumulation identifies sites of increased uptake, corresponding to increased glycolysis such as in neoplastic and inflammatory cells (macrophages, B and T lymphocytes, mast cells) [93, 94]. An easy quantitative measure of uptake is the standardized uptake value (SUV), measurable in a region of interest (ROI), which corresponds to the uptake in a given tissue divided by the total dose injected to the patient: SUVROI  =  [FDG uptakeROI (MBq/g) x body weight (g)] /injected activity (MBq). To compensate for background noise, SUVROI is often normalized by the vascular or the liver SUVs that are sensitive to most input or output bias. Even so, use of SUV remains subject to bias, although it is the most common quantitative uptake descriptor in clinical practice [95]. Finally, partial volume effects have to be considered, especially when the objects’ size are less than two times the full width at half maximum of the point spread function (the spatial resolution), even though a recent study has found their effect negligible in the assessment of AAA [96].

30.5 The Role of FDG PET Imaging The first human whole-body PET was commercialized in 1978 and the first hybrid PET-CT and PET-MRI scanners, respectively, in 2000 and 2011. Improving the spatial resolution down to 4–7 mm, availability of cyclotrons and radiopharmaceuticals suited to this type of imaging allowed

474

Combination of PET to CT has allowed coupling metabolic and morphological information. Practically, the PET data acquisition is immediately followed by CT image acquisition that in turn serves to correct the attenuation of the PET raw data by the tissue density. Both PET and CT are united within the same gantry, and software empower the co-registration of PET and CT data, with an adjustable level of fusion, simplifying accurate anatomical localization of tracer uptake. Of note, recent years have seen the development of PET-MRI, a promising alternative to PET-CT to which it might be superior but needs future research to assess the most appropriate applications before becoming clinically widespread [97].

30.5.2 Results 30.5.2.1 R  elationship Between FDG Uptake and Biological Activities In 2002, Sakalihasan et al. conducted a pioneering study with the underlying rationale that association of AAA prone to rupture (i.e., with abundant inflammatory cell infiltrates) could be detected by increased signaling on FDG PET [98]. All 10 of the 26 patients included had a short-term clinical event, whereas only 2 of the 16 remaining showed rapid growth, suggesting a possible association between increased FDG uptake in AAA and clinical outcomes. Since then, several studies have been performed to confirm the association between FDG uptake and the physiopathological events precluding AAA rupture. It has been widely shown that FDG uptake is associated with inflammatory and phagocytic cells infiltrates [56, 99–102], proteolytic activity by matrix metalloproteases [96, 99], and cellular and molecular signaling prefacing rupture [103]. However, there were arguments that FDG uptake is a nonspecific [104] and uncommon [105, 106] figure in AAAs since they are characterized by a cell density decrease [102]. This subchapter addresses the relationship between FDG uptake and patient/aneurysm outcomes.

A. Nchimi et al.

30.5.2.2 F  DG Uptake and AAA Growth and Rupture Prediction Animal models of AAA support the link between FDG uptake, growing size, and biological activities [60] (Figs. 30.6 and 30.7). English et al. used rats that were exposed to intra-aortic porcine pancreatic elastase and underwent daily subcutaneous injection of β-aminopropionitrile (an inducer of AAA rupture) in one group and saline in the remaining [107]. All the rats underwent sequential FDG micro-PET examinations, and rupture was monitored by radiotelemetry. FDG uptake was associated with inflammation in the AAA wall and focally increased at future sites of rupture, compared to control rats (Fig.  30.8). To date, there have been few studies evaluating aneurysm outcomes relative to FDG uptake in humans (Table 30.2). In the studies reporting rupture as the endpoint in patients with FDG PET/ CT, the site of rupture almost always spatially co-­ localizes with an area of FDG uptake [83, 98] (Fig.  30.9). Studying 53 patients with aortic aneurysms, we found that a higher rate of a composite outcome (rupture, dissection, or growth >1 cm) occurred in patients within 2 years after a PET/CT examination showing a visually increased 18F-FDG uptake [108] (Fig.  30.10). This replicates smaller series reported earlier by Sakalihasan et al. observing 26 patients [98] and Xu et al. observing 5 patients [83]. When the outcome is the aneurysm growth rate, the results seems to be more conflicting, according to a recent meta-analysis [109]. Kotze et  al. found no significant correlation between FDG uptake and AAA growth [101] in 14 AAAs under surveillance by using regular ultrasound. A study performed later by the same group in 25 patients similarly confirmed a negative correlation between FDG uptake and ultrasound expansion 1 year later [110]. Morel et  al. designed a different study by evaluating 39 patients with medically treated AAA who underwent an FDG PET/CT at baseline and 9 months later. The nine patients showing a significant increase in maximal diameter (≥2.5  mm) during the 9-month period had (1) a lower SUVmax in the AAA at

30  Emerging Tools to Assess the Risk of Rupture in AAA: Wall Stress and FDG PET

475

6

Aortic maximal diameter (mm)

5

11

10

10

7

7

7

4

4

14 4

e

15

19

3 22

18

15

19

d

2

c b

1

a 0 2

4

6

8

10

12

14

16

18

Days post surgery

Fig. 30.6 Post-elastase infusion (surgical) diameter growth curve of the aorta. Each point represents the number of rats, mean abdominal aortic diameter, and standard deviation. Images (a–e) are timeline inserts of selected transverse MRI. The curve is characterized by four different phases separated by vertical dotted lines. There is a short central quiescent phase (red circle) in-between day 3–6 post surgery, where a macroscopic ILT appears (b).

This phase is surrounded by two growth phases, the first of which is characterized by a subtle wall thickening and the remaining by progressive thickening and stratification of the ILT. Images (f–g) are inserts of ×20 magnification hematoxylin-eosin histological views of the normal aortic wall (f) and the aneurismal wall after appearance of the ILT (g, asterisk), showing inflammatory infiltrates. Abbreviations as in the text (Adapted From [60])

baseline than the other patients, (2) a higher variability of SUVmax between the two ­examinations, and (3) similar values of SUVmax compared to the other patients on the second observation. These results suggest a lower level of FDG uptake before a growth phase and more interestingly a pattern of cyclic metabolic changes in the AAA wall [111].

types present in the aneurysm. For example, there are two antagonist subtypes of macrophages within vascular lesions, which could not be distinguished by FDG PET: one pro-atherogenic and pro-inflammatory type, called M1, and the opposite, called M2 [112]. It makes nevertheless little doubt that FGD uptake tracks inflammation but the counterhypothesis that inflammatory activity in the aortic wall would be the consequence of expansion rather than the cause remains to be tested. Another part of the conflicting results with the use of FDG PET to predict AAA-related outcomes is attributable to the small study groups, mostly including large aneurysms and some descending thoracic aneurysms [83, 108]. Indeed, there is no evidence that FDG uptake is similar in large vs small aneurysms or in thoracic vs abdominal aneurysms. In addition, the aneurysms near the surgical thresholds prevent long-­

30.5.3 Limitations One of the main limitations of FDG PET is technical. Partial volume effect observed in small targets, like AAA wall, and spillover of lumen activity may alter the measures’ accuracy. Uptake quantification could be more accurate by implementing volume effect correction. The FDG uptake is not necessarily associated with a deleterious outcome, depending on the cells or cell sub-

A. Nchimi et al.

476

a

d

b

e

c

f

Fig. 30.7  ILT positive (left panel) and ILT negative (right panel) prone transverse FDG PET (a, d), CT (b, e) and fused PET-CT (c, e) images of the aorta in rats, respectively, 13 and 5 days after infusion of elastase. On the left panel, the aorta (arrow) is largely dilated. The ILT is seen as a ventral thickening of the aortic wall, containing two layers of different densities on CT. The luminal

part of the ILT has a low density and exhibits low FDG uptake (red arrows), while the external part of the thrombus has a higher CT density and exhibits stronger FDG uptake (yellow arrows). On the right panel, the aorta (open arrow) is undilated, and neither intraluminal thrombus nor increased FDG uptake is seen. Abbreviations as in the text (Adapted from [60])

30  Emerging Tools to Assess the Risk of Rupture in AAA: Wall Stress and FDG PET

m

0.009 0.001

0.01

2 1 0

(N =7 6d A

6d

(N =3

)

(N =8 14 A

RA A

l

d

(N =5

k

CA A

6d A

j

CA A

i

)

)

h

0.03

3

A

g

d

NR AA

f

c

)

e

b

18F-FDG SUVmax

a

477

Fig. 30.8  Micro-PET FDG uptake maps demonstrating increased focal uptake at the site of ultimate AAA rupture. Early phase represents the first 90 s, and late phase represents the last 30 min, of a 90 min micro-PET scan. (a–d) Coronal cuts for early phase images in control AAA 6  days (a) and 14 (b) after infusion of elastase, non-­ ruptured AAA (c), and ruptured (d) AAA 6  days after infusion of elastase and then rupture induction by daily subcutaneous β-aminopropionitrile. (e–h) Coronal cuts

for late phase images in the same animals, showing diffuse FDG uptake, with decreased uptake in the left anterolateral AAA wall, in all non-ruptured AAAs, and focal FDG uptake in the left lateral wall of the ruptured AAA. (i–l) Represent harvest photographs for the respective animals, magnification ×0.75, scale bar 5 mm: Rupture site correlated with the focal 18F-FDG uptake noted in image H. Yellow arrows identify the left kidney. Abbreviations as in the text (Adapted from [107])

term follow-ups. More importantly, these studies are based on time-point observations, which probably tell little about the dynamic nature of the processes ongoing in the aortic wall under the influence of other risk factors such as increases in blood pressure, infection, etc. Considering hypothesis of a cyclic pattern of metabolic changes in the AAA wall, a significantly increased FDG uptake may nevertheless prove to have a high positive predictive value and prompt AAA surgery repair after due confirmation by further research overcoming the current limitations.

sue characterization properties to the functional information provided by PET. In other cases, this confrontation may result in discrepancies, raising new insights for missing links between models. In our study, 68 PET and CT examinations were performed in 57 patients with aortic aneurysms among whom 6 patients with descending thoracic aortic aneurysms and 43 with AAAs [108]. Both quantitative FDG uptake and FES raised a total number of 163 hot spots (i.e., with either elevated wall tress or FDG uptake). Correlating the values of wall stress estimates and FDG uptake on these points resulted in a statistically significant, though moderate, correlation (r = 0.3, p = 0.0014), indicating a potentially comparable value for risk management. More interestingly, FDG uptake values were different depending on aneurysm location, higher in thoracic aneurysms than in AAAs; patients with history of AAA, or other arterial aneurysms; or angina pectoris, whereas it was lower in patients with claudication and chronic obstructive pulmonary disease. On a clin-

30.6 Insights from the Relationship Between Wall Stress and FDG Uptake Merging imaging modalities with distinct aims has the potential of compensating for each other’s limits, such as in the case of PET/CT where CT brings in spatial resolution, localization, and tis-

A. Nchimi et al.

478 Table 30.2  Summary of the results FDG PET with regard to patient and aneurysm outcomes Author name (reference) Sakalihasan et al. [98]

Publication Patient year N° Study design 2002 26 Case study follow-ups after initial FES and FDG PET

Reeps et al. [130] Kotze et al. [101] Xu et al. [83]

2008

15

2009

14

2010

5

Kotze et al. [110]

2011

25

Nchimi et al. [108]

2014

53

Morel et al. [111]

2015

39

Growth rate assessment after initial FDG PET Growth rate assessment after initial FDG PET Multiple case report follow-ups after initial FES and FDG PET Growth rate assessment after initial FDG PET 2-year event-free (rupture, accelerated growth, symptoms) survival after FDG PET assessment Growth rate assessment after initial FDG PET

FDG uptake analysis Study finding(s) Visual Increased FDG uptake was followed by symptoms, rapid expansion, pain, rupture SUVmax No correlation between FDG uptake and growth SUVmax No correlation between FDG uptake and growth Visual Potential link between high PWS and high SUV in AAA wall and rupture site SUVmax Inverse correlation between baseline FDG uptake and future growth Visual, Significantly higher rate of SUVmax clinical events in aneurysms with increased FDG uptake SUVmax

Inverse correlation between baseline FDG uptake and future growth

Abbreviations as in the text. NA not applicable

ical view, the two imaging models interplay and probably have a complimentary value in evaluating the risk of AAA rupture. The fact that wall metabolic activity and its correlation to wall stress were influenced by patient-specific factors points to a potential alteration of the biological responses to wall stress in AAA and provide possible clues for the increased risk of rupture [113] and poor outcomes after endovascular repair [114] associated with a familial history of AAA.  In short, FDG uptake could be at least partially related to genetic or acquired alterations of the arterial wall response to stress. An integrated patient-specific risk assessment strategy that would include imaging parameters and personal and heritable risk factors is therefore becoming increasingly suitable. Such an approach is tentatively included in PWRI estimates, where estimation of the aortic wall strength accounts for ILT thickness and gender [115]. Inclusion of further variables with established effect on wall strength such as increased FDG uptake and tobacco smoking will probably refine the current models. The gain expected by these refinements

has nevertheless to be weighted by cost-effectiveness and the evolving epidemiology currently reporting a steadily declining incidence of AAA in economically developed countries [116–118]. In the current state of the art, some diameterbased clinical scenarios could be yet improved by the impact of both FDG PET and FES in AAA risk of rupture assessment, including decisionmaking in patients with a large AAAs, elder or in poor condition, for whom procedural and postprocedural repair risks account for comorbidity.

30.7 Future Imaging Tools to Assess the Risk of Rupture in AAA Future imaging tools for assessing the risk of rupture include progress of current clinical imaging techniques and widespread use of other technical concepts. Among the promises raised from novel assessment of routinely used techniques is textural image analysis and artificial intelligence. In a recent study involving 50 patients with AAA,

30  Emerging Tools to Assess the Risk of Rupture in AAA: Wall Stress and FDG PET

479

A

c

b B

correlating point of max stress

point of max stress

SMOOTHED EFFECTIVE STRESS RST CALC TIME 2.500

156000. 132000. 108000. 84000. 60000. 36000. 12000.

MAXIMUM ∆ 166848. MINIMUM 2300.

a

*

Fig. 30.9  Upper panel (A): CT image of a large AAA (a), transverse images of fused PET-CT (b) showing increased 18-FDG update in the aneurysm neck, and the predicted wall stress (c). Lower panel (B). Top view of the

predicted wall stress in the AAA (a) and the corresponding FDG PET image (b) showing good correspondence between locations of high wall stress and 18-FDG uptake. Abbreviations as in the text (Adapted from [83])

among whom 40 underwent ultrasound follow-­up for 1 year after initial PET/CT, it was reported that the signal heterogeneity of the aneurismal wall components on CT correlated well with FDG uptake, thus reflecting the metabolic activity [119]. Further, CT signal heterogeneity and FDG uptake were strong predictors for expansion. MRI

also offers many opportunities for rupture risk assessment in AAAs, one of which evaluates the movement of water molecules in tissues: the diffusion-weighted imaging (DWI). DWI doesn’t require ionizing radiations, and even if the causes of signaling differ from that of FDG PET, it is similarly sensitive to cellular density. Despite

A. Nchimi et al.

0.8 0.6 0.4

PETPET+

0.0

0.2

Event-free survival

Fig. 30.10  A 2-year event-free survival curves in PET-positive and PET-negative patients according to increased FDG uptake. Abbreviations as in the text (from [108])

1.0

480

No. at risk 53 0

36 5

30

21

10

15

10 20

3 25

Time (months)

being an established technique for clinically important issues like acute stroke and inflammatory diseases [120, 121], the similarity between DWI and FDG PET in aortic aneurysm has so far been reported only once in a patient with an aortic arch aneurysm [122]. Likewise, tissue perfusion is another prospect of MRI, aiming at evaluating periaortic neoangiogenesis a marker of instability, as shown in preliminary reports [123, 124]. New contrast agents and tracers with organ and/or physiological process-specific affinity are highly suitable. The use of iron oxide particles that have affinity for the reticuloendothelial ­system has shown promises in evaluating macrophage adsorption at the luminal surface of the ILT [125] and outcomes in AAAs [126] before these contrast agents became commercially unavailable. A recent comparison of FDG and iron oxide uptake, respectively, on PET and MRI in 15 patients with AAA, has suggested that targets of the two techniques may be different cellular groups [127]. The future of metabolic and molecular imaging therefore lies in the combination of contrast agents and radiotracers to specific ligands, such as membrane components of macrophages, platelets, activated endothelia, and oxidized low-density lipoproteins [128, 129].

References 1. Lozano R, Naghavi M, Foreman K, Lim S, Shibuya K, Aboyans V, et al. Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet. 2012;380(9859):2095–128. 2. Lederle FA, Johnson GR, Wilson SE, Ballard DJ, Jordan WD Jr, Blebea J, et  al. Rupture rate of large abdominal aortic aneurysms in patients refusing or unfit for elective repair. JAMA. 2002;287(22):2968–72. 3. Mortality results for randomised controlled trial of early elective surgery or ultrasonographic surveillance for small abdominal aortic aneurysms. The UK Small Aneurysm Trial Participants. Lancet. 1998;352(9141):1649–55. 4. Fillinger MF, Marra SP, Raghavan ML, Kennedy FE.  Prediction of rupture risk in abdominal aortic aneurysm during observation: wall stress versus diameter. J Vasc Surg. 2003;37(4):724–32. 5. Moll FL, Powell JT, Fraedrich G, Verzini F, Haulon S, Waltham M, et  al. Management of abdominal aortic aneurysms clinical practice guidelines of the European society for vascular surgery. Eur J Vasc Endovasc Surg. 2011;41(Suppl 1):S1–S58. 6. Brown LC, Powell JT.  Risk factors for aneurysm rupture in patients kept under ultrasound surveillance. UK Small Aneurysm Trial Participants. Ann Surg. 1999;230(3):289–96; discussion 96–7. 7. Darling RC, Messina CR, Brewster DC, Ottinger LW. Autopsy study of unoperated abdominal aortic aneurysms. The case for early resection. Circulation. 1977;56(3 Suppl):II161–4.

30  Emerging Tools to Assess the Risk of Rupture in AAA: Wall Stress and FDG PET 8. Venkatasubramaniam AK, Fagan MJ, Mehta T, Mylankal KJ, Ray B, Kuhan G, et  al. A comparative study of aortic wall stress using finite element analysis for ruptured and non-ruptured abdominal aortic aneurysms. Eur J Vasc Endovasc Surg. 2004;28(2):168–76. 9. Lederle FA, Johnson GR, Wilson SE, Chute EP, Littooy FN, Bandyk D, et al. Prevalence and associations of abdominal aortic aneurysm detected through screening. Aneurysm Detection and Management (ADAM) Veterans Affairs Cooperative Study Group. Ann Intern Med. 1997;126(6):441–9. 10. Sakalihasan N, Limet R, Defawe OD.  Abdominal aortic aneurysm. Lancet. 2005;365(9470):1577–89. 11. Nicholls SC, Gardner JB, Meissner MH, Johansen HK. Rupture in small abdominal aortic aneurysms. J Vasc Surg. 1998;28(5):884–8. 12. Stonebridge PA, Draper T, Kelman J, Howlett J, Allan PL, Prescott R, et al. Growth rate of infrarenal aortic aneurysms. Eur J Vasc Endovasc Surg. 1996;11(1):70–3. 13. Vardulaki KA, Prevost TC, Walker NM, Day NE, Wilmink AB, Quick CR, et al. Growth rates and risk of rupture of abdominal aortic aneurysms. Br J Surg. 1998;85(12):1674–80. 14. Santilli SM, Littooy FN, Cambria RA, Rapp JH, Tretinyak AS, d'Audiffret AC, et  al. Expansion rates and outcomes for the 3.0-cm to the 3.9-cm infrarenal abdominal aortic aneurysm. J Vasc Surg. 2002;35(4):666–71. 15. Limet R, Sakalihassan N, Albert A.  Determination of the expansion rate and incidence of rupture of abdominal aortic aneurysms. J Vasc Surg. 1991;14(4):540–8. 16. Stenbaek J, Kalin B, Swedenborg J.  Growth of thrombus may be a better predictor of rupture than diameter in patients with abdominal aortic aneurysms. Eur J Vasc Endovasc Surg. 2000;20(5):466–9. 17. Martufi GLLM, Sakalihasan N, Panuccio G, Hultgren R, Roy J, Gasser TC. Local diameter, wall stress and thrombus thickness influence the local growth of abdominal aortic an-eurysms. J Endovasc Ther. 2016;23(6):957–66. 18. Liljeqvist MHR, Gasser TC, Roy J.  Volume growth of abdominal aortic aneurysms correlates with baseline volume and increasing finite element analysis-derived rupture risk. J Vasc Surg. 2016;63(6):1434–1442.e3. 19. Kurvers H, Veith FJ, Lipsitz EC, Ohki T, Gargiulo NJ, Cayne NS, et al. Discontinuous, staccato growth of abdominal aortic aneurysms. J Am Coll Surg. 2004;199(5):709–15. 20. Sakalihasan N, Heyeres A, Nusgens BV, Limet R, Lapiere CM.  Modifications of the extracellular matrix of aneurysmal abdominal aortas as a function of their size. Eur J Vasc Surg. 1993;7(6):633–7. 21. Busuttil RW, Abou-Zamzam AM, Machleder HI. Collagenase activity of the human aorta. A comparison of patients with and without abdominal aortic aneurysms. Arch Surg. 1980;115(11):1373–8.

481

22. Dobrin PB, Schwarcz TH, Baker WH. Mechanisms of arterial and aneurysmal tortuosity. Surgery. 1988;104(3):568–71. 23. Sakalihasan N, Delvenne P, Nusgens BV, Limet R, Lapiere CM.  Activated forms of MMP2 and MMP9 in abdominal aortic aneurysms. J Vasc Surg. 1996;24(1):127–33. 24. Defawe OD, Colige A, Lambert CA, Munaut C, Delvenne P, Lapiere CM, et  al. TIMP-2 and PAI-1 mRNA levels are lower in aneurysmal as compared to athero-occlusive abdominal aortas. Cardiovasc Res. 2003;60(1):205–13. 25. Defawe OD, Colige A, Lambert CA, Delvenne P, Lapiere Ch M, Limet R, et al. Gradient of proteolytic enzymes, their inhibitors and matrix proteins expression in a ruptured abdominal aortic aneurysm. Eur J Clin Investig. 2004;34(7):513–4. 26. Aimes RT, Quigley JP.  Matrix metalloproteinase-2 is an interstitial collagenase. Inhibitor-free enzyme catalyzes the cleavage of collagen fibrils and soluble native type I collagen generating the specific 3/4- and 1/4-length fragments. J Biol Chem. 1995;270(11):5872–6. 27. Busti C, Falcinelli E, Momi S, Gresele P.  Matrix metalloproteinases and peripheral arterial disease. Intern Emerg Med. 2010;5(1):13–25. 28. Egeblad M, Werb Z.  New functions for the matrix metalloproteinases in cancer progression. Nat Rev Cancer. 2002;2(3):161–74. 29. Thompson RW, Parks WC. Role of matrix metalloproteinases in abdominal aortic aneurysms. Ann N Y Acad Sci. 1996;800:157–74. 30. Choke E, Cockerill GW, Dawson J, Wilson RW, Jones A, Loftus IM, et al. Increased angiogenesis at the site of abdominal aortic aneurysm rupture. Ann N Y Acad Sci. 2006;1085:315–9. 31. Holmes DR, Liao S, Parks WC, Thompson RW.  Medial neovascularization in abdominal aortic aneurysms: a histopathologic marker of aneurysmal degeneration with pathophysiologic implications. J Vasc Surg. 1995;21(5):761–71; discussion 71–2. 32. Paik DC, Fu C, Bhattacharya J, Tilson MD. Ongoing angiogenesis in blood vessels of the abdominal aortic aneurysm. Exp Mol Med. 2004;36(6):524–33. 33. Reeps C, Pelisek J, Seidl S, Schuster T, Zimmermann A, Kuehnl A, et  al. Inflammatory infiltrates and neovessels are relevant sources of MMPs in abdominal aortic aneurysm wall. Pathobiology. 2009;76(5):243–52. 34. Sakalihasan N, Pincemail J, Defraigne JO, Nusgens B, Lapiere C, Limet R.  Decrease of plasma vitamin E (alpha-tocopherol) levels in patients with abdominal aortic aneurysm. Ann N Y Acad Sci. 1996;800:278–82. 35. Pincemail J, Defraigne JO, Cheramy-Bien JP, Dardenne N, Donneau AF, Albert A, et  al. On the potential increase of the oxidative stress status in patients with abdominal aortic aneurysm. Redox Rep. 2012;17(4):139–44.

482 36. Darling RC 3rd, Brewster DC, Darling RC, LaMuraglia GM, Moncure AC, Cambria RP, et  al. Are familial abdominal aortic aneurysms different? J Vasc Surg. 1989;10(1):39–43. 37. Verloes A, Sakalihasan N, Koulischer L, Limet R. Aneurysms of the abdominal aorta: familial and genetic aspects in three hundred thirteen pedigrees. J Vasc Surg. 1995;21(4):646–55. 38. Verloes A, Sakalihasan N, Limet R, Koulischer L.  Genetic aspects of abdominal aortic aneurysm. Ann N Y Acad Sci. 1996;800:44–55. 39. Derubertis BG, Trocciola SM, Ryer EJ, Pieracci FM, McKinsey JF, Faries PL, et  al. Abdominal aortic aneurysm in women: prevalence, risk factors, and implications for screening. J Vasc Surg. 2007;46(4):630–5. 40. Brown PM, Zelt DT, Sobolev B. The risk of rupture in untreated aneurysms: the impact of size, gender, and expansion rate. J Vasc Surg. 2003;37(2):280–4. 41. Powell JT, Worrell P, MacSweeney ST, Franks PJ, Greenhalgh RM.  Smoking as a risk factor for abdominal aortic aneurysm. Ann N Y Acad Sci. 1996;800:246–8. 42. Lindeman JH, Abdul-Hussien H, van Bockel JH, Wolterbeek R, Kleemann R. Clinical trial of doxycycline for matrix metalloproteinase-9 inhibition in patients with an abdominal aneurysm: doxycycline selectively depletes aortic wall neutrophils and cytotoxic T cells. Circulation. 2009;119(16):2209–16. 43. Nyberg A, Skagius E, Englund E, Nilsson I, Ljungh A, Henriksson AE. Abdominal aortic aneurysm and the impact of infectious burden. Eur J Vasc Endovasc Surg. 2008;36(3):292–6. 44. Cheuk BL, Ting AC, Cheng SW.  Detection of C. pneumoniae by polymerase chain reaction-enzyme immunoassay in abdominal aortic aneurysm walls and its association with rupture. Eur J Vasc Endovasc Surg. 2005;29(2):150–5. 45. Marques da Silva R, Caugant DA, Lingaas PS, Geiran O, Tronstad L, Olsen I.  Detection of Actinobacillus actinomycetemcomitans but not bacteria of the red complex in aortic aneurysms by multiplex polymerase chain reaction. J Periodontol. 2005;76(4):590–4. 46. Nakano K, Nemoto H, Nomura R, Inaba H, Yoshioka H, Taniguchi K, et al. Detection of oral bacteria in cardiovascular specimens. Oral Microbiol Immunol. 2009;24(1):64–8. 47. Vorp DA, Raghavan ML, Webster MW. Mechanical wall stress in abdominal aortic aneurysm: influence of diameter and asymmetry. J Vasc Surg. 1998;27(4):632–9. 48. Kazi M, Thyberg J, Religa P, Roy J, Eriksson P, Hedin U, et al. Influence of intraluminal thrombus on structural and cellular composition of abdominal aortic aneurysm wall. J Vasc Surg. 2003;38(6):1283–92. 49. Vollmar JF, Paes E, Pauschinger P, Henze E, Friesch A. Aortic aneurysms as late sequelae of above-knee amputation. Lancet. 1989;2(8667):834–5.

A. Nchimi et al. 50. Biasetti J, Hussain F, Gasser TC.  Blood flow and coherent vortices in the normal and aneurysmatic aortas: a fluid dynamical approach to intra-­ luminal thrombus formation. J R Soc Interface. 2011;8(63):1449–61. 51. Biasetti J, Spazzini PG, Swedenborg J, Gasser TC. An integrated fluid-chemical model toward modeling the formation of intra-luminal thrombus in abdominal aortic aneurysms. Front Physiol. 2012;3:266. 52. Hans SS, Jareunpoon O, Balasubramaniam M, Zelenock GB.  Size and location of thrombus in intact and ruptured abdominal aortic aneurysms. J Vasc Surg. 2005;41(4):584–8. 53. Li ZY, U-King-Im J, Tang TY, Soh E, See TC, Gillard JH. Impact of calcification and intraluminal thrombus on the computed wall stresses of abdominal aortic aneurysm. J Vasc Surg. 2008;47(5):928–35. 54. Thubrikar MJ, Robicsek F, Labrosse M, Chervenkoff V, Fowler BL. Effect of thrombus on abdominal aortic aneurysm wall dilation and stress. J Cardiovasc Surg. 2003;44(1):67–77. 55. Touat Z, Lepage L, Ollivier V, Nataf P, Hvass U, Labreuche J, et al. Dilation-dependent activation of platelets and prothrombin in human thoracic ascending aortic aneurysm. Arterioscler Thromb Vasc Biol. 2008;28(5):940–6. 56. Sarda-Mantel L, Coutard M, Rouzet F, Raguin O, Vrigneaud JM, Hervatin F, et al. 99mTc-annexin-V functional imaging of luminal thrombus activity in abdominal aortic aneurysms. Arterioscler Thromb Vasc Biol. 2006;26(9):2153–9. 57. Houard X, Rouzet F, Touat Z, Philippe M, Dominguez M, Fontaine V, et al. Topology of the fibrinolytic system within the mural thrombus of human abdominal aortic aneurysms. J Pathol. 2007;212(1):20–8. 58. Martinez-Pinna R, Madrigal-Matute J, Tarin C, Burillo E, Esteban-Salan M, Pastor-Vargas C, et al. Proteomic analysis of intraluminal thrombus highlights complement activation in human abdominal aortic aneurysms. Arterioscler Thromb Vasc Biol. 2013;33(8):2013–20. 59. Fontaine V, Jacob MP, Houard X, Rossignol P, Plissonnier D, Angles-Cano E, et al. Involvement of the mural thrombus as a site of protease release and activation in human aortic aneurysms. Am J Pathol. 2002;161(5):1701–10. 60. Nchimi A, Courtois A, El Hachemi M, Touat Z, Drion P, Withofs N, et  al. Multimodality imaging assessment of the deleterious role of the intraluminal thrombus on the growth of abdominal aortic aneurysm in a rat model. Eur Radiol. 2016;26(7):2378–86. 61. Vorp DA, Lee PC, Wang DH, Makaroun MS, Nemoto EM, Ogawa S, et  al. Association of intraluminal thrombus in abdominal aortic aneurysm with local hypoxia and wall weakening. J Vasc Surg. 2001;34(2):291–9. 62. Behr-Rasmussen C, Grondal N, Bramsen MB, Thomsen MD, Lindholt JS. Mural thrombus and the progression of abdominal aortic aneurysms: a large

30  Emerging Tools to Assess the Risk of Rupture in AAA: Wall Stress and FDG PET population-based prospective cohort study. Eur J Vasc Endovasc Surg. 2014;48(3):301–7. 63. Remus EW, O’Donnell RE Jr, Rafferty K, Weiss D, Joseph G, Csiszar K, et al. The role of lysyl oxidase family members in the stabilization of abdominal aortic aneurysms. Am J Physiol Heart Circ Physiol. 2012;303(8):H1067–75. 64. Abedin M, Tintut Y, Demer LL.  Vascular calcification: mechanisms and clinical ramifications. Arterioscler Thromb Vasc Biol. 2004;24(7):1161–70. 65. O'Leary SA, Mulvihill JJ, Barrett HE, Kavanagh EG, Walsh MT, McGloughlin TM, et al. Determining the influence of calcification on the failure properties of abdominal aortic aneurysm (AAA) tissue. J Mech Behav Biomed Mater. 2015;42:154–67. 66. Zienkiewicz OC.  The finite elements method. The basis. 5th ed. Oxford: Butterworth Heinemann; 2000. 67. Leung JH, Wright AR, Cheshire N, Crane J, Thom SA, Hughes AD, et  al. Fluid structure interaction of patient specific abdominal aortic aneurysms: a comparison with solid stress models. Biomed Eng Online. 2006;5:33. 68. Raghavan ML, Vorp DA.  Toward a biomechanical tool to evaluate rupture potential of abdominal aortic aneurysm: identification of a finite strain constitutive model and evaluation of its applicability. J Biomech. 2000;33(4):475–82. 69. Polzer S, Gasser TC, Swedenborg J, Bursa J.  The impact of intraluminal thrombus failure on the mechanical stress in the wall of abdominal aortic aneurysms. Eur J Vasc Endovasc Surg. 2011;41(4):467–73. 70. Auer M, Gasser TC.  Reconstruction and finite element mesh generation of abdominal aortic aneurysms from computerized tomography angiography data with minimal user interactions. IEEE Trans Med Imaging. 2010;29(4):1022–8. 71. Vande Geest JP, Schmidt DE, Sacks MS, Vorp DA. The effects of anisotropy on the stress analyses of patient-specific abdominal aortic aneurysms. Ann Biomed Eng. 2008;36(6):921–32. 72. Gasser TC, Auer M, Labruto F, Swedenborg J, Roy J. Biomechanical rupture risk assessment of abdominal aortic aneurysms: model complexity versus predictability of finite element simulations. Eur J Vasc Endovasc Surg. 2010;40(2):176–85. 73. Maier A, Gee MW, Reeps C, Pongratz J, Eckstein HH, Wall WA.  A comparison of diameter, wall stress, and rupture potential index for abdominal aortic aneurysm rupture risk prediction. Ann Biomed Eng. 2010;38(10):3124–34. 74. Gasser TC, Nchimi A, Swedenborg J, Roy J, Sakalihasan N, Bockler D, et al. A novel strategy to translate the biomechanical rupture risk of abdominal aortic aneurysms to their equivalent diameter risk: method and retrospective validation. Eur J Vasc Endovasc Surg. 2014;47(3):288–95. 75. Xenos M, Labropoulos N, Rambhia S, Alemu Y, Einav S, Tassiopoulos A, et  al. Progression of

483

abdominal aortic aneurysm towards rupture: refining clinical risk assessment using a fully coupled fluid-­ structure interaction method. Ann Biomed Eng. 2015;43(1):139–53. 76. Vande Geest JP, Di Martino ES, Bohra A, Makaroun MS, Vorp DA. A biomechanics-based rupture potential index for abdominal aortic aneurysm risk assessment: demonstrative application. Ann N Y Acad Sci. 2006;1085:11–21. 77. Fillinger MF, Raghavan ML, Marra SP, Cronenwett JL, Kennedy FE. In vivo analysis of mechanical wall stress and abdominal aortic aneurysm rupture risk. J Vasc Surg. 2002;36(3):589–97. 78. Heng MS, Fagan MJ, Collier JW, Desai G, McCollum PT, Chetter IC.  Peak wall stress measurement in elective and acute abdominal aortic aneurysms. J Vasc Surg. 2008;47(1):17–22; discussion 79. Khosla S, Morris DR, Moxon JV, Walker PJ, Gasser TC, Golledge J. Meta-analysis of peak wall stress in ruptured, symptomatic and intact abdominal aortic aneurysms. Br J Surg. 2014;101(11):1350– 7. discussion 7 80. Erhart P, Hyhlik-Durr A, Geisbusch P, Kotelis D, Muller-Eschner M, Gasser TC, et al. Finite element analysis in asymptomatic, symptomatic, and ruptured abdominal aortic aneurysms: in search of new rupture risk predictors. Eur J Vasc Endovasc Surg. 2015;49(3):239–45. 81. Erhart P, Roy J, de Vries JP, Liljeqvist ML, Grond-­ Ginsbach C, Hyhlik-Durr A, et  al. Prediction of rupture sites in abdominal aortic aneurysms after finite element analysis. J Endovasc Ther. 2016;23(1):115–20. 82. Doyle BJ, McGloughlin TM, Miller K, Powell JT, Norman PE.  Regions of high wall stress can predict the future location of rupture of abdominal aortic aneurysm. Cardiovasc Intervent Radiol. 2014;37(3):815–8. 83. Xu XY, Borghi A, Nchimi A, Leung J, Gomez P, Cheng Z, et  al. High levels of 18F-FDG uptake in aortic aneurysm wall are associated with high wall stress. Eur J Vasc Endovasc Surg. 2010;39(3):295–301. 84. Erhart P, Grond-Ginsbach C, Hakimi M, Lasitschka F, Dihlmann S, Bockler D, et al. Finite element analysis of abdominal aortic aneurysms: predicted rupture risk correlates with aortic wall histology in individual patients. J Endovasc Ther. 2014;21(4):556–64. 85. Malkawi A, Pirianov G, Torsney E, Chetter I, Sakalihasan N, Loftus IM, et  al. Increased expression of Lamin A/C correlate with regions of high wall stress in abdominal aortic aneurysms. Aorta (Stamford). 2015;3(5):152–66. 86. Georgakarakos E, Ioannou C, Kostas T, Katsamouris A. Inflammatory response to aortic aneurysm intraluminal thrombus may cause increased 18F-FDG uptake at sites not associated with high wall stress: comment on “high levels of 18F-FDG uptake in aortic aneurysm wall are associated with high wall

484 stress”. Eur J Vasc Endovasc Surg. 2010;39(6):795; author reply –6 87. Speelman L, Hellenthal FA, Pulinx B, Bosboom EM, Breeuwer M, van Sambeek MR, et al. The influence of wall stress on AAA growth and biomarkers. Eur J Vasc Endovasc Surg. 2010;39(4):410–6. 88. Li ZY, Sadat U, U-King-Im J, Tang TY, Bowden DJ, Hayes PD, et  al. Association between aneurysm shoulder stress and abdominal aortic aneurysm expansion: a longitudinal follow-up study. Circulation. 2010;122(18):1815–22. 89. Martufi G, Lindquist Liljeqvist M, Sakalihasan N, Panuccio G, Hultgren R, Roy J, et  al. Local diameter, wall stress, and thrombus thickness influence the local growth of abdominal aortic aneurysms. J Endovasc Ther. 2016;23(6):957–66. 90. Hyhlik-Durr A, Krieger T, Geisbusch P, Kotelis D, Able T, Bockler D.  Reproducibility of deriving parameters of AAA rupture risk from patient-­ specific 3D finite element models. J Endovasc Ther. 2011;18(3):289–98. 91. Teutelink A, Cancrinus E, van de Heuvel D, Moll F, de Vries JP.  Preliminary intraobserver and interobserver variability in wall stress and rupture risk assessment of abdominal aortic aneurysms using a semiautomatic finite element model. J Vasc Surg. 2012;55(2):326–30. 92. Rudd JH. The role of 18F-FDG PET in aortic dissection. J Nucl Med. 2010;51(5):667–8. 93. Kadoglou NP, Liapis CD. Matrix metalloproteinases: contribution to pathogenesis, diagnosis, surveillance and treatment of abdominal aortic aneurysms. Curr Med Res Opin. 2004;20(4):419–32. 94. Jacob T, Ascher E, Hingorani A, Gunduz Y, Kallakuri S.  Initial steps in the unifying theory of the pathogenesis of artery aneurysms. J Surg Res. 2001;101(1):37–43. 95. Buvat I.  Les limites du SUV.  Méd Nucl. 2007;21:165–72. 96. Reeps C, Bundschuh RA, Pellisek J, Herz M, van Marwick S, Schwaiger M, et al. Quantitative assessment of glucose metabolism in the vessel wall of abdominal aortic aneurysms: correlation with histology and role of partial volume correction. Int J Cardiovasc Imaging. 2013;29(2):505–12. 97. Torigian DA, Zaidi H, Kwee TC, Saboury B, Udupa JK, Cho ZH, et al. PET/MR imaging: technical aspects and potential clinical applications. Radiology. 2013;267(1):26–44. 98. Sakalihasan N, Van Damme H, Gomez P, Rigo P, Lapiere CM, Nusgens B, et  al. Positron emission tomography (PET) evaluation of abdominal aortic aneurysm (AAA). Eur J Vasc Endovasc Surg. 2002;23(5):431–6. 99. Defawe OD, Hustinx R, Defraigne JO, Limet R, Sakalihasan N.  Distribution of F-18 fluorodeoxyglucose (F-18 FDG) in abdominal aortic aneurysm: high accumulation in macrophages seen on PET imaging and immunohistology. Clin Nucl Med. 2005;30(5):340–1.

A. Nchimi et al. 100. Truijers M, Kurvers HA, Bredie SJ, Oyen WJ, Blankensteijn JD.  In vivo imaging of abdominal aortic aneurysms: increased FDG uptake suggests inflammation in the aneurysm wall. J Endovasc Ther. 2008;15(4):462–7. 101. Kotze CW, Menezes LJ, Endozo R, Groves AM, Ell PJ, Yusuf SW.  Increased metabolic activity in abdominal aortic aneurysm detected by 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET/CT). Eur J Vasc Endovasc Surg. 2009;38(1):93–9. 102. Marini C, Morbelli S, Armonino R, Spinella G, Riondato M, Massollo M, et al. Direct relationship between cell density and FDG uptake in asymptomatic aortic aneurysm close to surgical threshold: an in vivo and in vitro study. Eur J Nucl Med Mol Imaging. 2012;39(1):91–101. 103. Courtois A, Nusgens BV, Hustinx R, Namur G, Gomez P, Somja J, et al. 18F-FDG uptake assessed by PET/CT in abdominal aortic aneurysms is associated with cellular and molecular alterations prefacing wall deterioration and rupture. J Nucl Med. 2013;54(10):1740–7. 104. Barwick TD, Lyons OT, Mikhaeel NG, Waltham M, O'Doherty MJ. 18F-FDG PET-CT uptake is a feature of both normal diameter and aneurysmal aortic wall and is not related to aneurysm size. Eur J Nucl Med Mol Imaging. 2014;41(12):2310–8. 105. Palombo D, Morbelli S, Spinella G, Pane B, Marini C, Rousas N, et  al. A positron emission tomography/computed tomography (PET/CT) evaluation of asymptomatic abdominal aortic aneurysms: another point of view. Ann Vasc Surg. 2012;26(4):491–9. 106. Tegler G, Ericson K, Sorensen J, Bjorck M, Wanhainen A. Inflammation in the walls of asymptomatic abdominal aortic aneurysms is not associated with increased metabolic activity detectable by 18-fluorodeoxyglucose positron-emission tomography. J Vasc Surg. 2012;56(3):802–7. 107. English SJ, Piert MR, Diaz JA, Gordon D, Ghosh A, D'Alecy LG, et  al. Increased 18F-FDG uptake is predictive of rupture in a novel rat abdominal aortic aneurysm rupture model. Ann Surg. 2015;261(2):395–404. 108. Nchimi A, Cheramy-Bien JP, Gasser TC, Namur G, Gomez P, Seidel L, et al. Multifactorial relationship between 18F-fluoro-deoxy-glucose positron emission tomography signaling and biomechanical properties in unruptured aortic aneurysms. Circ Cardiovasc Imaging. 2014;7(1):82–91. 109. Timur UT, van Herwaarden JA, Mihajlovic D, De Jong P, Mali W, Moll FL. (18)F-FDG PET scanning of abdominal aortic aneurysms and correlation with molecular characteristics: a systematic review. EJNMMI Res. 2015;5(1):76. 110. Kotze CW, Groves AM, Menezes LJ, Harvey R, Endozo R, Kayani IA, et  al. What is the relationship between (18)F-FDG aortic aneurysm uptake on PET/CT and future growth rate? Eur J Nucl Med Mol Imaging. 2011;38(8):1493–9.

30  Emerging Tools to Assess the Risk of Rupture in AAA: Wall Stress and FDG PET 111. Morel O, Mandry D, Micard E, Kauffmann C, Lamiral Z, Verger A, et  al. Evidence of cyclic changes in the metabolism of abdominal aortic aneurysms during growth phases: (1)(8)F-FDG PET sequential observational study. J Nucl Med. 2015;56(7):1030–5. 112. Riou LM, Vanzetto G, Broisat A, Fagret D, Ghezzi C. Equivocal usefulness of FDG for the noninvasive imaging of abdominal aortic aneurysms. Eur J Nucl Med Mol Imaging. 2014;41(12):2307–9. 113. Sakalihasan N, Defraigne JO, Kerstenne MA, Cheramy-Bien JP, Smelser DT, Tromp G, et  al. Family members of patients with abdominal aortic aneurysms are at increased risk for aneurysms: analysis of 618 probands and their families from the Liege AAA Family Study. Ann Vasc Surg. 2014;28(4):787–97. 114. van de Luijtgaarden KM, Bastos Goncalves F, Hoeks SE, Valentijn TM, Stolker RJ, Majoor-­ Krakauer D, et  al. Lower atherosclerotic burden in familial abdominal aortic aneurysm. J Vasc Surg. 2014;59(3):589–93. 115. Vande Geest JP, Wang DH, Wisniewski SR, Makaroun MS, Vorp DA.  Towards a noninvasive method for determination of patient-specific wall strength distribution in abdominal aortic aneurysms. Ann Biomed Eng. 2006;34(7):1098–106. 116. Svensjo S, Bjorck M, Gurtelschmid M, Djavani Gidlund K, Hellberg A, Wanhainen A.  Low prevalence of abdominal aortic aneurysm among 65-year-old Swedish men indicates a change in the epidemiology of the disease. Circulation. 2011;124(10):1118–23. 117. Anjum A, Powell JT.  Is the incidence of abdominal aortic aneurysm declining in the 21st century? Mortality and hospital admissions for England & Wales and Scotland. Eur J Vasc Endovasc Surg. 2012;43(2):161–6. 118. Sandiford P, Mosquera D, Bramley D.  Trends in incidence and mortality from abdominal aortic aneurysm in New Zealand. Br J Surg. 2011;98(5):645–51. 119. Kotze CW, Rudd JH, Ganeshan B, Menezes LJ, Brookes J, Agu O, et  al. CT signal heterogeneity of abdominal aortic aneurysm as a possible predictive biomarker for expansion. Atherosclerosis. 2014;233(2):510–7. 120. Padhani AR, Koh DM, Collins DJ.  Whole-body diffusion-weighted MR imaging in cancer: current status and research directions. Radiology. 2011;261(3):700–18. 121. Jacobs MA, Ibrahim TS, Ouwerkerk R.  AAPM/ RSNA physics tutorials for residents: MR imag-

485

ing: brief overview and emerging applications. Radiographics. 2007;27(4):1213–29. 122. Nchimi A, Couvreur T, Meunier B, Sakalihasan N. Magnetic resonance imaging findings in a positron emission tomography-positive thoracic aortic aneurysm. Aorta (Stamford). 2013;1(3):198–201. 123. Nguyen VL, Backes WH, Kooi ME, Wishaupt MC, Hellenthal FA, Bosboom EM, et  al. Quantification of abdominal aortic aneurysm wall enhancement with dynamic contrast-enhanced MRI: feasibility, ­reproducibility, and initial experience. J Magn Reson Imaging. 2014;39(6):1449–56. 124. Nguyen VL, Kooi ME, Backes WH, van Hoof RH, Saris AE, Wishaupt MC, et  al. Suitability of pharmacokinetic models for dynamic contrast-enhanced MRI of abdominal aortic aneurysm vessel wall: a comparison. PLoS One. 2013;8(10):e75173. 125. Nchimi A, Defawe O, Brisbois D, Broussaud TK, Defraigne JO, Magotteaux P, et  al. MR imaging of iron phagocytosis in intraluminal thrombi of abdominal aortic aneurysms in humans. Radiology. 2010;254(3):973–81. 126. Richards JM, Semple SI, MacGillivray TJ, Gray C, Langrish JP, Williams M, et  al. Abdominal aortic aneurysm growth predicted by uptake of ultrasmall superparamagnetic particles of iron oxide: a pilot study. Circ Cardiovasc Imaging. 2011;4(3):274–81. 127. McBride OM, Joshi NV, Robson JM, MacGillivray TJ, Gray CD, Fletcher AM, et  al. Positron emission tomography and magnetic resonance imaging of cellular inflammation in patients with abdominal aortic aneurysms. Eur J Vasc Endovasc Surg. 2016;51(4):518–26. 128. Kitagawa T, Kosuge H, Chang E, James ML, Yamamoto T, Shen B, et al. Integrin-targeted molecular imaging of experimental abdominal aortic aneurysms by (18)F-labeled Arg-Gly-Asp positron-­ emission tomography. Circ Cardiovasc Imaging. 2013;6(6):950–6. 129. Tegler G, Estrada S, Hall H, Wanhainen A, Bjorck M, Sorensen J, et al. Autoradiography screening of potential positron emission tomography tracers for asymptomatic abdominal aortic aneurysms. Ups J Med Sci. 2014;119(3):229–35. 130. Reeps C, Essler M, Pelisek J, Seidl S, Eckstein HH, Krause BJ.  Increased 18F-fluorodeoxyglucose uptake in abdominal aortic aneurysms in positron emission/computed tomography is associated with inflammation, aortic wall instability, and acute symptoms. J Vasc Surg. 2008;48(2):417–23. d­ iscussion 24

Exploring the Thoracic Aorta with Advanced Magnetic Resonance Imaging Beyond Routine Diameter Measurements

31

Alban Redheuil

31.1 Introduction If aorta, the main trunk of the cardiovascular system, appears by name in Medieval Latin in the 1590s, it is Aristotle, in ancient Greece, who first named artere the “great artery of the heart.” The initial meaning of the word is to “hold suspended.” So it seems that the initial understanding of the purpose of the aorta was to hold the heart suspended in the chest. Later, if William Harvey and René Descartes disagreed as to the mechanics of the heart, they both acknowledged the central role of the aorta as the prime conduit for blood to circulate from the heart to peripheral arteries. For the aorta to be truly considered a “great artery,” it’s crucial cushioning function of pulsatile blood flow needed to be recognized. This came with several authors, and William Osler in The Principles and Practice of Medicine in 1892 distinguished arteriosclerosis related to physiological aging from atheroma and put forA. Redheuil (*) Department of Cardiovascular Imaging DICVRIT, Institute of Cardiology, Pitié-Salpêtrière Hospital, Paris, France Sorbonne Université, Pierre et Marie Curie School of Medicine, UPMC, Paris, France LIB Biomedical Imaging Laboratory, INSERM/ CNRS/UPMC, Paris, France ICAN Institute of Cardiometabolism and Nutrition, Imaging Core Lab, Paris, France e-mail: [email protected]

ward the importance of aortic elasticity by referring to the aortic wall as the “vital rubber.” He also added the notion of individual variability in the quantity and quality of aortic material and pointed to possible congenital differences in aortic compliance. Since these historic times, a large body of work led primarily by physiologists and hypertension experts led to the recognition of the aorta and foremost the ascending portion of the thoracic aorta. However, in daily practice this has remained largely overlooked as the quantification of the effects of aging on the arterial tree remains largely based on cuff measurements of brachial blood pressure and calculation of pulse pressure. Furthermore, the concept of an acute aortic syndrome including dissection, mural hematoma, ulcer, and trauma is fairly recent and only beginning to be considered as a purveyor of unexplained sudden deaths and contributor to overall mortality in aging populations. Paradoxically, aortic surgery and endovascular interventions have done tremendous progress in the last decades and so have noninvasive imaging techniques toward 3D or 4D (dynamic 3D), but routine decisions and guidelines for prophylactic surgery remain based on a maximal diameter. The striking fact that must call into question this status quo is that half of all aortic dissections happen in patients with aortic diameters below the recommended threshold for preventive ­intervention [1].

© Springer-Verlag GmbH Austria, part of Springer Nature 2019 O. H. Stanger et al. (eds.), Surgical Management of Aortic Pathology, https://doi.org/10.1007/978-3-7091-4874-7_31

487

488

Our aim in this chapter will not be to extensively review imaging techniques of the aorta but rather to focus on selected issues, concerns, and hopes raised by the most recent imaging technology to increase our ability to explore aortic aging and diseases and potentially contribute new imaging biomarkers to assess individual aortic risk.

31.2 The Importance of the Proximal Aorta The increasing burden of cardiovascular disease in aging world populations has been related to two main factors: arterial stiffness associated with increased systolic and pulse pressures. Aortic stiffness has emerged clinically as an important integrated marker of large arterial vessel damage over a lifetime, whereas measures such as blood pressure or biological markers are indirect and instantaneous measures of aortic stiffness furthermore highly modifiable by drugs [2]. Furthermore, age-related changes in blood pressure are only indirect measures of aortic stiffness and begin to be significant only later in life. Central to the arterial tree, the proximal aorta is responsible for most of the buffering and initial conduit function of the pulsatile systolic flow from the heart to the peripheral vasculature and end organs. Alterations of these two components of aortic function lead to inefficiency of the circulatory system and potential vascular-­ventricular uncoupling eventually deleterious to the heart through increased workload. One of the issues concerning aortic diseases promoting imaging at the forefront of aortic exploration lies in the fact that most aortic diseases are slow progressive alterations of aortic shape and size leading to late-onset symptoms often announcing life-threatening complications. The main features of aortic disease remain based on changes in overall vascular shape with increased size and tortuosity and to a lesser degree to increased wall thickness in inflammatory or infectious disease. Historically, the aortic contours have been observed in standard radiography with important inherent drawbacks, and

A. Redheuil

the lumen was seen on aortography images using intravascular contrast agents. Ultrasounds have allowed to visualize both vascular wall and lumen with high temporal and spatial resolution, essentially in 2D, but are technically limited to available acoustic windows making the exploration of the whole aorta difficult. However, exploration of the aortic root is constantly available, and large cross-sectional and longitudinal study data are available. In clinical routine, aortic root diameter is commonly measured from M-mode tracings with the leading-­edge-to-leading-edge technique, as recommended by the American Society of Echocardiography [3]. Changes in aortic diameter throughout the cardiac cycle represent 1D measures of aortic strain that reflect local aortic function. Reference values for aortic diameter by two-dimensional (2D) echocardiography have been reported in children and young adults [4] over a broad range of age [5, 6]. Echocardiography also allows for estimation of aortic flow using Doppler, albeit in narrow windows often insufficient for the comprehensive assessment of complex flow patterns and with the necessity to study the principal flow direction. Computed tomography, thanks to isotropic spatial resolution and recent advances in spatial coverage, has enabled complete imaging of the aorta in a few seconds combining high-resolution luminal imaging with imaging of the calcified and noncalcified wall components. If CT provides accurate aortic area or volume instead of simple diameters, radiation exposure and inability to quantify flow represent the main drawbacks. CT is the most frequently used comprehensive imaging technique to assess ­aortic dissection and other entities of the aortic syndrome in an emergency setting [7]. Conversely, MRI, a noninvasive technique, allows for direct and accurate measurement of different aortic characteristics including structural measures such as aortic area or volume, aortic length, aortic curvature, and aortic wall thickness, as well as functional measures such as aortic deformation (strain) and stiffness, distensibility, and pulse wave velocity (PWV) [8]. The intrinsic sensitivity of magnetic resonance

31  Exploring the Thoracic Aorta with Advanced Magnetic Resonance Imaging Beyond Routine Diameter… 489

to motion and flow can be used to produce velocity and flow maps using phase contrast acquisition sequences. One of the main disadvantages of MRI is the relatively lower temporal resolution (10–20 ms) compared to ultrasounds or applanation tonometry, although recent MRI technical developments and post-processing techniques have the potential to overcome such disadvantages. Newer techniques such as MRI 4D flow can provide a comprehensive assessment of aortic structure, function, and flow simultaneously in the three spatial dimensions and through time [9].

31.3 Aortic Structure and Geometry MRI, a reference technique to evaluate cardiac structure and function, also uniquely provides noninvasive depiction of aortic anatomy, wall motion, and flow in any given anatomical localization or spatial direction. ECG gating and segmented acquisition techniques over several heartbeats are used to generate anatomic images corresponding to different phases of the cardiac cycle. Monophasic diastolic or systolic images can be acquired with nulling of the blood-­ generating black-blood images useful for arterial wall and wall thickness assessment. Dynamic cine acquisitions can be used to track aortic wall motion with high spatial (0.7 mm) and temporal resolution (10–20  ms) depending on the heart rate and breath-holding ability (Fig. 31.1).

31.3.1 Aortic Wall Thickness MRI allows to visualize the aortic wall in any plane orientation (Fig.  31.1). Most studies are based on black-blood spin echo T1-weighted acquisitions which generate a sharp contrast between the signal from the wall and the blood. An age-related increase in average wall thickness of the descending aorta was reported in 1053 general population individuals of the MESA (Multiethnic Study of Atherosclerosis) aged 45–85 years (average 2.35 ± 0.5 mm) but was not

found for the ascending aorta (average 2.8 mm). In this study, SBP and hypertension were also independent correlates of increased wall thickness. However, this study also reported that aortic wall thickness was more strongly associated with distensibility than with PWV among participants aged 55–90  years without hypertension [10]. This interesting finding suggests that increases in aortic wall thickness are an associate of local increased stiffness seen in late adulthood when biodynamics shift from predominantly elastic to mostly stiff and fibrotic wall properties related to physiological aging. Good reproducibility for this measurement has been reported [11]. However, with the spatial resolution of most acquisitions (0.7–1 mm) being about half of the expected thickness for the aortic wall (1.5–2.5  mm), we believe that partial volume remains a major drawback that will require higher spatial resolution and refined automated quantification methods to be accurate. No large MRI studies are available on the wall thickness of young and healthy individuals.

31.3.2 Cross-Sectional Aortic Dimensions One of the strengths of MRI is to provide a 2D or 3D approach allowing to accurately measure aortic area or volume instead of relying on a 1D measure of diameter prone to higher measurement bias such as in echocardiography. It should be noted that the cross-sectional area of the aorta can also be used to calculate a mean diameter value which is more robust than a single 1D ­measurement and can be readily interpreted by physicians, as the culture in the field remains largely based on diameter values. The distribution of the mean diameter of the ascending aorta was reported in 3573 general population individuals aged 45–85 years according to age and gender by Turkbey et al. who measured the luminal diameter of the ascending aorta on magnitude images of a phase contrast sequence in a large number of healthy subjects of the MESA [12]. A synthetic view of expected mean ascending aortic diameters is given in Table 31.1. Voges et al. [13]

A. Redheuil

490

a

b

c

d

e

Fig. 31.1 Assessing anatomy of the thoracic aorta including lumen, wall, and arch geometry in MRI. Note: (a) Cross-sectional black-blood T1-weighted image used to measure aortic wall thickness and detect intramural hematoma (absent here); (b) cross-sectional SSFP cine (diastolic image) used to measure aortic strain or intimal flap dynamics in aortic dissection; (c) long-axis

p­arasagittal view of the thoracic aorta in black-blood T1-weighted imaging; (d) sectional view showing central aortic regurgitation (*dark linear signal void from the aortic valve leaflet tips toward the anterior mitral valve leaflet); and (e) 3D view of the thoracic aorta from a 3D SSFP cine ECG-­triggered and respiratory-gated sequence

Table 31.1  Reference median values for the ascending aorta diameter Ascending aorta diameter (mm) Mena Womena Children Kaiser et al.b Voges et al.c

45–54 years 32 29

55–64 years 33 30

65–74 years 34 31

75– 84 years 35 31

Expected diameter irrespective of age and gender in children

−1.33 + 18.6*BSA0.5 [4*(−0.0386 + 2.913*BSA)/π]0.5

Note: BSA body surface area Summary from aTurkbey et al.: mean diastolic diameter calculated from cross-sectional area segmentation in the MESA sub-cohort without cardiovascular risk factors; bKaiser et al. reported expected luminal aortic diameters after reconstructions from 3D MR angiographies; cVoges et al. reported external systolic aortic areas from which a mean diameter is calculated for comparability from SSFP cine images

31  Exploring the Thoracic Aorta with Advanced Magnetic Resonance Imaging Beyond Routine Diameter… 491

and Burmann et  al. [14] performed measurements of the aortic root (cusp-commissure and cusp-cusp) in diastole and systole on SSFP cine images. In the MESA study by Turkbey et  al., the median value of AA luminal diameter increased by 1.1  mm per each decade for both men and women. Overall, age was the strongest associate of increased ascending aortic diameter along with gender and body surface area. The subgroup study of the 1612 “healthy” individuals without risk factors showed a mean and 95th percentile (upper limit) of AA diameter of 31 mm (38 mm) vs. 32  mm (39  mm) in the global cohort. The inter-reader intraclass correlation for the measurement of mean ascending aortic area on modulus images of MRI phase contrast acquisitions on 100 individuals was 0.98 (95% CI: 0.98–0.99) with a mean difference of 0.14  mm (limits of agreement −1.1–1.4  mm). Interestingly, in the Dallas Heart study, increased abdominal aortic wall thickness was associated with a high lifetime predicted risk of cardiovascular disease in individuals from 30 to 50 years with a low short-­ term risk [15]. This rapid overview concerning MRI data of the ascending aorta demonstrates the dramatic heterogeneity of imaging sequences, ECG gating (static or dynamic imaging), cardiac cycle phase, inclusion or exclusion of vessel wall, and vascular segmentation: manual, semiautomated, and rarely automated. This absence of standardiza-

16 yo

30 yo

42 yo

Fig. 31.2  Morphological evolution of the human aorta with age. Note that the early modifications of the aorta include widening of the aorta, particularly the ascending segment accompanied by widening of the arch and finally marked tortuosity. Automated segmentation of aortic vol-

tion is currently undermining the comparability of available data and the definition of the standard of care for future guidelines. The constitutional decrease in aortic diameter along the aorta or aortic “tapering” has been well documented by MRI [16, 17]. However, the age-­ related widening of the aortic lumen and rate of diameter increase for each respective location is not homogeneous along the aorta. Indeed, the progression of aortic diameters markedly predominates in the ascending compared to descending aorta. Secondly, the thoracic aorta elongates with increasing age [17, 18]. We have recently shown that the age-related elongation of the aorta is also much more marked in the ascending aorta and associated with widening of the arch and unfolding of the aorta [17] as illustrated in Fig.  31.2. Reasons for this inhomogeneity in regional structural alterations likely include the relative content in elastin fibers, higher in the ascending aorta and decreasing along the aorta leading to the more muscular and resistive peripheral arteries. This constitutional difference between the ascending aorta and more peripheral locations is well known but not often functionally explored. We have shown aortic arch unfolding due to widening and elongation with increased tortuosity to be independently related to increased peripheral and central blood pressure components as well as to cardiac hypertrophic ­remodeling, i.e., increased left ventricular mass and mass-to-volume ratio in healthy individuals [18].

59 yo

64 yo

69 yo

ume on isotropic 3D-SSFP MRI ECG and respiratory-­ gated acquisitions as in Dietenbeck T. et al. JCMR 2017. Images courtesy of Thomas Dietenbeck, LIB Laboratory of Biomedical Imaging INSERM/CNRS/UPMC, Paris (courtesy of T. Dietenbeck, LIB, Paris)

A. Redheuil

492

Although CT and MRI provide 3D images, diameters are still widely used, and the potential of these imaging techniques is overlooked. Reasons for this underuse include only recent development of reliable high-resolution gated sequences, lack of fast and simple post-­processing methods, lack of available volumetric data, and standardization. The newer 3D anatomic sequences such as dual-gated SSFP should be preferred over classic non-ECG-gated 3D MR angiography as they provide motion-free images of the aortic root and proximal aorta and allow more precise measurement. Furthermore, they can be performed without contrast injection. Ideally, these images should be analyzed using a semiautomated method with standardized anatomical site labeling providing regional aortic volumes which will enable accurate patient follow-­up and inter-patient comparisons. Using this approach, Dietenbeck et al. show using MRI that aortic volume changes in hypertension and aortic diseases are more significant than diameter changes independent of BSA.  Furthermore, the volume increase in hypertension is shown to reflect a homogeneous dilatation of the ascending and descending aorta, whereas aneurysmal patients predominantly have ascending aorta dilatation and elongation. This pattern was found to be further pronounced in bicuspid patients (unpublished data). The main advantage of using aortic volumes instead of diameters is that volumetric measurements by definition take into account both the cross-sectional dilatation and longitudinal elongation processes. The main drawback that remains today is the lack of large population data and nomograms for aortic volumes necessary for routine use. Of similar importance, proof of the clinical added value of using volumes instead of diameters is also lacking.

31.4 A  ortic Function: Aortic Stiffness and Flow The noninvasive assessment of aortic stiffness schematically relies on the assessment of either pulse wave velocity or local cross-sectional changes in vessel area driven by local pulse pres-

sure (distensibility) as summarized in Fig.  31.3. The reference method to assess global aortic stiffness is the measurement of carotid to femoral pulse wave velocity by applanation tonometry (cfPWV) [19]. This technique benefits from a large experience, wide cohort applications, and reported distribution in the general population and patients, particularly in hypertension. Furthermore, cfPWV is a recognized independent risk factor for mortality and hard cardiovascular events in the general population. More recently, magnetic resonance imaging (MRI) has been proposed to provide a comprehensive noninvasive study of the aorta. One of the strengths of this method is indeed to provide simultaneously a thorough three-dimensional exploration of aortic anatomy as well as local wall dynamics and flow. Assessment of cardiac structure and function or peripheral arteries and organs can also be performed in the same setting. Local and direct indices of aortic stiffness may have increased relevance as they may be more specifically related to aging vs. a disease process and also because they may be more sensitive to early infraclinical alterations than more global or ­surrogate markers of arterial stiffness. However, the challenge remains to determine the predictive value of such markers for adverse aortic events beyond available routine measurements such as diameter.

31.4.1 Aortic Wall Dynamics (Fig. 31.3) MRI allows complete dynamic exploration of the thoracic aorta in any given plane. Most static cross-sectional aortic area measures are done in diastole and therefore only represent the “resting” state of the aorta. Dynamic acquisitions providing circumferential aortic lumen area changes across the cardiac cycle may have greater value as they quantify the buffer component of aortic function dampening pulsatile flow from the heart. Circumferential aortic strain may be defined as the relative change in aortic area during the cardiac cycle, expressed in percent (AS  =  ΔA/ Amin). Aortic distensibility can be measured as AD = AS/PP, PP being the pulse pressure driving the aortic area changes.

31  Exploring the Thoracic Aorta with Advanced Magnetic Resonance Imaging Beyond Routine Diameter… 493

Fig. 31.3  Assessing aortic stiffness in MRI: aortic distensibility and pulse wave velocity. Note: PWV pulse wave velocity; cf-PWV carotid-femoral pulse wave veloc-

ity using applanation tonometry; PP pulse pressure; Amax, Amin maximal and minimal aortic areas; ΔA Amax−Amin

Accurate measurement of aortic strain requires high spatial, temporal, and contrast resolution. These features are achieved by current SSFP (steady-state free precession) cine acquisitions which can achieve 0.7 mm of spatial resolution, less than 10 ms of temporal resolution, and high contrast between wall and blood with constant signal. Such images allow to perform robust automated segmentation of the aortic wall [20, 21]. Most aortic distensibility studies rely on brachial cuff pressure measured during the MRI exam to normalize aortic strain. However, because of the amplification phenomenon in the periphery this may lead to overestimate central

PP and therefore underestimate distensibility, especially in younger individuals, we proposed to use central PP instead of brachial PP to calculate distensibility [8]. Central pressures can be estimated either using carotid tonometry or radial tonometry and a transfer function.

31.4.2 Aortic Wall Dynamics in Aging and Disease Aortic distensibility decreases nonlinearly with advancing age. Marked age-related decrease in aortic distensibility has been shown to occur in

A. Redheuil

494

asymptomatic general population individuals from 20 to 50  years of age with a continued decrease thereafter albeit at a slower pace [8]. Ascending aortic distensibility measured using central pulse pressure is on average 74 ± 23 kPa−1 × 10−3 between 20 and 30 years, 18  ±  7  kPa−1  ×  10−3 between 50 and 60, and 10 ± 6 kPa−1 × 10−3 after 70 years with an average decrease of 14  ±  1  kPa−1  ×  10−3 per decade. Importantly, this change is due to a decline in aortic strain as pulse pressure is not significantly increased in this age group. In multivariate analysis including cf-PWV, carotid distensibility, and Aix, ascending aortic distensibility was the strongest correlate of age making it a powerful independent and early infraclinical marker of functional aortic dysfunction. Regional heterogeneity is also found for functional indices such as distensibility as only the youngest individuals before the age of 30 have ascending aorta distensibility exceeding descending aorta distensibility [15, 16]. This early reversal in proximal to distal distensibility in healthy individuals could be modified by early-onset cardiovascular disease. Recent studies from MESA showed that decreased ascending aortic distensibility assessed by MRI predicted all-cause mortality in a multiethnic population free of overt CV disease [22]. The hazard ratio for death over 8.5 years was 2.3 for the 20% of individuals with the stiffest aorta independent of all established CV risk factors, LV mass, and common measures of subclinical atherosclerosis (coronary artery calcium, carotid intima-media thickness, ankle-brachial index). Ascending aortic distensibility predicted CV events independent of age and CVD risk factors in those with low-to-moderate individual risk. There are only few data on the predictive value of aortic distensibility for aortic events. Teixedo et al. studied 80 consecutive patients with Marfan syndrome using MRI compared with 36 age- and sex-matched controls [23]. Aortic distensibilities were abnormal in the entire aorta in Marfan patients. Moreover, Marfan patients without dilated aortic root showed clear impairment of aortic biomechanics, which suggests that such imaging functional parameters may be used as early biomarkers of aortic involvement in these

patients. Another study by Prakash et al. showed in 83 patients with connective tissue disorder a reduction in ascending aorta distensibility in comparison with published normative values. Over a median follow-up period of 2.7  years, there were no aortic dissections or deaths, but 16 of 83 (19%) patients underwent surgical aortic root replacement. In multivariable analysis, lower aortic root strain (P  =  0.05) was independently associated with aortic root replacement. Lower values of ascending aorta strain (P = 0.02) were associated with a higher rate of aortic root dilation [24].

31.5 Aortic Flow 31.5.1 Regional Aortic Pulse Wave Velocity (Fig. 31.3) Phase contrast MRI allows to acquire spatially registered velocities simultaneously with morphological data throughout the cardiac cycle thus allowing to obtain local flow values in any aortic location in 2D. Velocity mapping is usually performed in one through-plane direction (perpendicular to the vessel of interest) but can also be performed in-plane or in 3D at the cost of lower spatial and temporal resolutions and higher scanning and post-processing times [9, 25]. A single acquisition plane, typically placed perpendicular to the ascending aorta at the level of the mid-right pulmonary artery, simultaneously provides flow in the ascending and descending thoracic aorta during the same heartbeats. Pulse wave velocity is measured as the transit vascular length divided by the blood flow transit time. Flow curves should be generated using a robust automated vascular segmentation method reducing manual bias. Transit time measurement also requires caution as the temporal resolutions used in most 2D MRI phase contrast methods (T) has been shown to exert a paradoxical vasoprotective effect in repaired TOF associated with smaller diameters at the sinotubular junction level and decreased aortic stiffness [13]. In addition, exonic fibrillin-1 gene variants have been associated with an eightfold increased risk of aortic dilatation in a wide spectrum of TOF cases, including extreme variants of double-outlet right ventricle and pulmonary atresia [14]. The presence of 22q11.2 deletion in TOF children has also been associated with increased aortic annular and aortic sinus diameters [15]. Clinical risk factors for progressive aortic dilatation in repaired TOF include male sex, longer time interval between palliative shunt and repair, presence of pulmonary atresia, and right aortic arch [7]. Increased aortic wall stiffness has also

been associated with aortic root dilatation in both children and adults with repaired TOF [16]. Despite increased observation of aortopathy in TOF, the rate of aortic dissection remains low and has only been reported in isolated cases (Table 32.1) [11, 17–19]. On the other hand, the greatest aortic root diameter in TOF without dissection has been reported in a patient who underwent complete repair at age 43 with subsequent progressive dilatation to 85  mm [12]. Thus, it may be that TOF patients are less predisposed to aortic dissection compared to BAV and Marfan patients, despite similar underlying histopathological changes in their aortas.

32.3 Common Arterial Trunk Common arterial trunk, also known as truncus arteriosus, is a rare CHD lesion often associated with DiGeorge syndrome (22q11 deletion) [20]. It is characterised by a single arterial vessel arising from the base of the heart with a diameter larger than the normal aorta. Histopathological analyses are sparse although moderate to severe medial wall degeneration has been found in four neonates undergoing initial repair and a young child undergoing conduit replacement [21]. Similar to other CHD lesions, aortopathy in this setting may be inherent with superimposed

32  Congenital Aortopathy

505

Table 32.1  Reported cases of aortic dissection and rupture in congenital heart disease lesions. The anatomic site of aortic dilatation and associated risk factors are also outlined with relevant references included in the main text Lesion Site of dilatation Tetralogy of Fallot Ao root, ascending aorta (lesser extent)

Common arterial Ao root trunk Eisenmenger VSD – Ross procedure

Arterial switch operation Hypoplastic left heart syndrome

Risk factors • Male sex • Longer time interval between palliative shunt and repair • Pulmonary atresia • Right aortic arch • Severe truncal valve regurgitation –

Aortic dissectiona Four cases in adults with repaired TOF (aortic diameter ranging from 53 to 93 mm) [1–4]

One dissection [5], 1 rupture [6] (diameter not defined) Single case (normal-sized ascending aorta) [7] Four cases (autograft diameter 55 mm Ao root (predominantly • Preoperative aortic regurgitation and dilatation in two cases, not defined in remaining sinus level and two) [8–11] • Surgical technique sinotubular junction) Ao root (predominantly • Previous pulmonary artery Two TGA cases [5] (no detailed history/diameter available) sinus level) banding • Older age at operation Ao root (predominantly • Neo-aortic regurgitation Single case (neo-aortic root diameter sinus level) (any degree) 78 mm) [12]

Abbreviations: Ao aortic, TGA transposition of the great arteries, TOF tetralogy of Fallot Maximal aortic diameter given in brackets when available

a

v­olume and pressure overload contributing to arterial dilatation prior to surgical repair [21]. Progressive aortic root dilatation has been reported in a small number of DiGeorge cases, although these were only associated with minor cardiac defects rather than common arterial trunk [22]. Aortic root z-scores following surgical repair were ≥2 in 96% of patients with truncus arteriosus in a series of 76 young children and adults (7.1 years median age at follow-up, range 0.2– 39.5) [23]. Aortic z-scores remained stable as age increased, with only six patients necessitating root surgery, predominantly for truncal valve regurgitation [23]. To date, there is a single report of aortic dissection in the literature [24] and another case of aortic root rupture in a 35-year-­old man with unrepaired truncus arteriosus and secondary pulmonary arterial hypertension (Table 32.1) [25]. Although the common arterial trunk diameter is not reported in either, microscopic studies of the truncal wall revealed elastic fragmentation with increased mucoid content in the latter case. Interestingly, severe medial wall abnormalities have been described in necropsy specimens from three adult patients

with truncus arteriosus and associated pulmonary vascular disease [21]. Despite the possible protective effect of surgical repair from exposure to increased arterial pressures, younger age at initial intervention has not been associated with smaller truncal root z-scores [23]. Other negative associations included gender, DiGeorge syndrome, presence of interrupted aortic arch, and use of angiotensin-converting enzyme (ACE) inhibitor with only severe truncal valve regurgitation correlating with higher root z-scores [23].

32.4 Other Rare Lesions Structural abnormalities of the aortic wall have been reported in a number of other CHD lesions, such as pulmonary stenosis with single ventricle, double aortic arch, and Eisenmenger ventricular septal defect (VSD) [21]. The clinical significance of the above findings remains to be seen as aortic dilatation in the above lesions is rarely observed. Importantly, aortic rupture has been reported in a single case with Eisenmenger VSD; despite a normal diameter aorta in this patient,

506

severe medial wall abnormalities were present on microscopy in both the ascending aorta and the pulmonary trunk [21].

32.4.1 Postoperative Neo-aortic Dilatation Aortic dilatation in CHD may occur postoperatively in a number of corrective and palliative procedures where the autologous pulmonary valve is transferred either functionally or anatomically into the systemic circulation. In all of the above situations, the reconstructed systemic outflow tract, also known as the “neo-aorta”, is introduced to a high-pressure left-sided physiology, resulting in arterial remodelling and neo-aortic dilatation.

32.5 Ross Procedure Since its introduction in 1967, Ross procedure has gained popularity as the preferred operation for aortic valve replacement in the young, allowing for somatic growth and obviating the need for permanent anticoagulation [26]. Sixty-five to 81% of patients undergoing the Ross procedure have a BAV [27, 28]. Pulmonary autograft dilatation is a common complication with increased incidence when the root method is applied (pulmonary autograft implanted as a full root replacement) compared to the sub-coronary technique (pulmonary autograft inserted as an inclusion cylinder) [29]. Neo-aortic dilatation is associated with valve insufficiency and is noted primarily at the sinus level and sinotubular junction [30]. Enlargement of the neo-aortic root occurs rapidly within 10 days of the procedure and continues to progress during the first year with similar growth patterns in operated patients and control subjects thereafter [31, 32]. Freedom from pulmonary autograft dilatation at 7  years is approximately 45% [33, 34]. Severe medial degenerative changes have been observed in both the ascending aorta and main pulmonary artery of BAV patients undergoing the Ross procedure [35]. Perturbations in the

M. Prapa and M. A. Gatzoulis

common embryologic origin of the great arteries may be fundamental to this process and raise the question of whether pulmonary autograft autotransplantation should be performed in this setting [35]. However, in a large meta-analysis of 39 studies focused on Ross procedure outcomes, only in a single report there was a possible association between neo-aortic valve regurgitation and presence of a BAV [36, 37]. Thus, it may be that exposure to systemic pressures is the main driver of neo-aortic remodelling with intrinsic abnormalities of the pulmonary artery in BAV having a secondary role. The above is further supported by common histological findings of medial fragmentation and fibrosis in explanted pulmonary autografts for various reasons, with no difference in severity depending on the absence or presence of a native BAV [38]. Preoperative aortic regurgitation and dilatation appear to be the main risk factors for decreased autograft durability, with the impact of BAV disease remaining debatable as mentioned above [36]. Surgical technique is also important with individual variations of the root method influencing results; as postoperative autograft enlargement occurs predominantly at the level of the sinus and sinotubular junction rather than the annulus, it may be that minimising autograft root length results in less overall dilatation [36]. Freedom from autograft reoperation is between 65 and 82% at 15 years [39, 40]. To date, a few cases of pulmonary autograft dissection have been reported in the literature (Table 32.1), with all three patients having previously undergone the Ross procedure for severe BAV insufficiency [41–43]. Histological findings have been described in a different study including a single case of a dissected autograft [38]; smooth muscle cell hypertrophy and realignment were a uniform finding in both aneurysmal and non-­ aneurysmal autograft tissues, suggesting an active cell state even 12 years after the Ross operation. A similar state has been described in neo-­ aortic valvular interstitial cells, with matrix metalloproteinase-13 levels exceeding those of normal control valves, up to 6  years after autograft implantation [44, 45]. Thus, a common mode of adaptive remodelling may take place in

32  Congenital Aortopathy

the pulmonary autograft whilst exceeding its biologic allowance to act as systemic outflow tract and ultimately leading to autograft wall degeneration and failure [38].

32.6 Arterial Switch Operation Over the last two to three decades, arterial switch operation (ASO) has been established as the preferred surgical technique for anatomic correction of transposition of the great arteries (TGA) [46]. ASO offers the important advantage of a systemic left ventricle compared to atrial switch procedure but is not without late complications, including disproportionate neo-aortic dilatation and valve insufficiency [47]. Dilatation occurs predominantly at the level of the sinuses and, to a lesser extent, the annulus and sinotubular junction [48]. On echocardiographic assessment of 335 ASO patients with a median postoperative follow-up of 5 years, 33.4% had neo-aortic root z-scores of ≥3 [49]. Enlargement of the neo-aorta is rapid over the first postoperative year with gradual growth occurring thereafter, in pace with the normal population [50, 51]. Moderate to severe medial wall abnormalities have been described in the normal-sized ascending aorta of eight neonates with complete TGA undergoing ASO [21]. In a different study, 20% out of 28 neonates had cystic medial necrosis in both their aorta and pulmonary artery following one-stage ASO [52]. Three phases of arterial remodelling have been highlighted in the context of ASO: the initial preoperative phase, where the transposed aorta is exposed to 60% of the combined cardiac output passing the right ventricle; the immediate postoperative phase, with exposure of the neo-aorta to systemic pressures; and the third phase of gradual neo-aortic growth [50]. Posterior translocation of the ascending aorta during ASO also has long-term effects leading to sharper angulation of the aortic arch with greater pulse wave reflection and dilatation of the ascending aorta [53]. Schwartz et  al. reported 95% freedom from reoperation on the neo-aortic valve or root 10  years following ASO [49]. Risk factors for

507

enlargement of the neo-aorta included previous pulmonary artery banding and older age at operation [49]. Interestingly, aortic pathology has also been reported in TGA following atrial switch procedure [54] with a reported case of a 7 cm aortic aneurysm arising from the right ventricle leading to aortic dissection in a 42-year-old female patient [55]. To current date, only two further cases of aortic dissection in TGA have been reported in a population-based study, although their surgical history is not specified (Table 32.1) [24]. However, it is worth noticing that “time” appears to be the most important factor for the anatomical and functional state of the neo-aortic root [51]. In the largest published series of 1,200 newborns and infants undergoing ASO at a single centre, only a subset of 1,095 survivors (operated in 1982–1999) is currently reaching adulthood, and hence, many of the procedure’s late complications still remain to be seen [47].

32.7 H  ypoplastic Left Heart Syndrome Following staged palliative reconstruction for hypoplastic left heart syndrome (HLHS), the pulmonary valve functions as the systemic valve with reported dilatation of the neo-aortic valve complex [56]. In contrast to the pattern of dilatation seen in ASO and Ross procedure, neo-aortic enlargement in HLHS occurs progressively and out of proportion to normal growth in children and adolescents [56]. In a study of 53 HLHS patients who had the Fontan procedure, neo-­ aortic z-scores were normal after stage 1 reconstruction but increased to >2 in the vast majority (98%) following the volume unloading procedure [56]. Dilatation occurred at all levels from the neo-aortic valve annulus to the sinotubular junction but was more pronounced at the level of the sinuses. The only variable correlating with severity of neo-aortic dilatation in the above study was the presence of any degree of neo-­aortic regurgitation at most recent follow-up. The underlying histopathological changes in the HLHS neo-aorta are largely unknown although medial wall degenerative changes have

M. Prapa and M. A. Gatzoulis

508

been described in the great arteries of other single-­ventricle physiology lesions, such as pulmonary atresia [21]. Interestingly, in a case report of a 14-year-old boy with HLHS who developed significant aneurysm of the neo-aorta and proximal aortic arch following completed, staged palliation, microscopy of the resected native pulmonary artery and ascending aorta exhibited elastic fragmentation and mucopolysaccharide-­ rich areas [57]. However, there was additional fibrosis and dystrophic calcification of the ascending aorta with the boy testing positive for a novel SMAD3 mutation on a background family history of aneurysms-osteoarthritis syndrome. Despite the likely effect of the SMAD3 mutation in the above case contributing to aneurysm formation [58], there are a few cases of HLHS patients in the literature with neo-aortic root dilatation necessitating surgical intervention [59–61] including a case of a 26-year-old male with a dissected 7.8 cm neo-aortic root (Table 32.1) [62].

32.8 Management 32.8.1 Presentation In most patients with repaired CHD, the aorta is incidentally found to be dilated during follow-up evaluation of other abnormalities [63]. Essentially a silent disease, aortic dilatation can manifest with symptoms of aortic regurgitation, when associated with significant annular dilatation, or chest/back pain, in the incidence of acute expansion or dissection. Importantly, dissection of the denervated autograft in cases such as the Ross procedure can occur without any symptoms and, thus, go undetected for long periods of time [43].

32.8.2 Diagnosis Transthoracic echocardiography (TTE) is usually the first-line investigation of the above aortic pathologies, able to detect and grade aortic insufficiency, measure aortic dimensions, and assess systolic and diastolic ventricular function. TTE can also assess elastic properties of the aorta with

a positive association between increased stiffness and aortic root z-scores in repaired TOF [64]. However, TTE is not preferable for surveillance of patients with aortic disease as it fails to visualise and measure consistently the tubular portion of the ascending aorta [65]. To obtain a complete thoracic aortic view, computed tomography (CT) or magnetic resonance imaging (MRI) is superior. MRI is particularly useful in the evaluation of conotruncal anomalies, as it can accurately assess right ventricular size and function in the presence of pulmonary regurgitation [63]. There is by and large lack of guidance focused on CHD aortopathy with the exception of connective tissue disorders and BAV disease. Imaging of the aorta is generally recommended on an annual basis when aortic diameters exceed 45 mm; there is progressive dilatation or a family history of dissection [66, 67]. MRI or CT imaging of the aorta is also an important part of preoperative planning in these lesions; repeat median sternotomy can injure anteriorly positioned aneurysmal segments such as the ascending aorta in repaired TOF. Additionally, coronary artery anatomy can be delineated by the above imaging modalities and contribute to surgical planning, when anomalies are present. Lastly, four-­ dimensional (4D) MRI flow analyses are becoming increasingly available and have already exhibited abnormal vortices in repaired TOF that may contribute to vascular remodelling [68].

32.8.3 Treatment According to recent recommendations by the European Society of Cardiology (ESC) in 2010, root surgery following ASO is suggested when neo-aortic dimensions exceed 55  mm [69]. In the absence of guidance on the remaining lesions mentioned in this chapter, the same cutoff value of 55  mm has been previously proposed [63]. However, disputing evidence arises from a recent report of aortic dissection in a patient with repaired TOF and a maximal (transsinus) diameter of 53 mm on CT imaging [18]. Hence, the Canadian Cardiovascular Society (CCS) consensus guidance published in 2009

32  Congenital Aortopathy

may be more accurate in management of TOF aortopathy, as it supports intervention when aortic dimension reaches at least 55  mm [70]. Additional indications when the aortic diameter is below 50  mm include rapid progression in size (>1 cm/year), presence of at least moderate aortic regurgitation, and positive family history of aortic dissection [71]. The beneficial therapeutic effect of antihypertensives, such as angiotensin receptor blockers, has recently emerged in MFS aortopathy with lack of equivalent experimental or clinical data in other aortic pathologies [72]. However, arterial hypertension should be effectively controlled in CHD patients as it is a known risk factor for accelerated aortic dilatation [1]. Other lifestyle modifications include avoidance of obesity, tobacco, and stimulating recreational drugs [65]. Exercise prescription is challenging in this setting as sudden increases in mental or physical stress can trigger aortic dissection [65]. For patients with thoracic aortic aneurysm, the general recommendation is in favour of a regular routine of aerobic exercise and against strenuous activities that require a Valsalva manoeuvre [65].

32.9 Conclusion Aortic dilatation is frequently encountered in a wide range of CHD lesions. Although it appears to be less threatening compared to Marfan syndrome, it should not be overlooked as a significant number of adverse events including death may go unreported. Our depth of knowledge on congenital aortopathy will continue to increase as more patients with repaired CHD survive into adulthood. Until then, increased awareness of this complex entity is clearly warranted.

References 1. Dietz HC, et al. Marfan syndrome caused by a recurrent de novo missense mutation in the fibrillin gene. Nature. 1991;352:337–9. 2. El-Hamamsy I, Yacoub MH.  Cellular and molecular mechanisms of thoracic aortic aneurysms. Nat Rev Cardiol. 2009;6:771–86.

509 3. Zanjani KS, Niwa K. Aortic dilatation and aortopathy in congenital heart diseases. J Cardiol. 2013;61:16–21. 4. Bahnson HT, et al. Surgical treatment and follow-up of 147 cases of tetralogy of Fallot treated by correction. J Thorac Cardiovasc Surg. 1962;44:419–32. 5. Bull K, Somerville J, Ty E, Spiegelhalter D.  Presentation and attrition in complex pulmonary atresia. J Am Coll Cardiol. 1995;25:491–9. 6. Marelli AJ, Perloff JK, Child JS, Laks H. Pulmonary atresia with ventricular septal defect in adults. Circulation. 1994;89:243–51. 7. Niwa K, Siu SC, Webb GD, Gatzoulis MA. Progressive aortic root dilatation in adults late after repair of tetralogy of Fallot. Circulation. 2002;106:1374–8. 8. Jonsson H, Ivert T, Brodin LA.  Echocardiographic findings in 83 patients 13-26 years after intracardiac repair of tetralogy of Fallot. Eur Heart J. 1995;16:1255–63. 9. Bhat AH, Smith CJ, Hawker RE. Late aortic root dilatation in tetralogy of Fallot may be prevented by early repair in infancy. Pediatr Cardiol. 2004;25:654–9. 10. Tan JL, Davlouros PA, McCarthy KP, Gatzoulis MA, Ho SY.  Intrinsic histological abnormalities of aortic root and ascending aorta in tetralogy of Fallot: evidence of causative mechanism for aortic dilatation and aortopathy. Circulation. 2005;112:961–8. 11. Kim WH, et al. Aortic dissection late after repair of tetralogy of Fallot. Int J Cardiol. 2005;101:515–6. 12. Dodds GA, Warnes CA, Danielson GK. Aortic valve replacement after repair of pulmonary atresia and ventricular septal defect or tetralogy of Fallot. J Thorac Cardiovasc Surg. 1997;113:736–41. 13. Cheung YF, Hong WJ, Chan KW, Wong SJ. Modulating effects of matrix metalloproteinase-3 and -9 polymorphisms on aortic stiffness and aortic root dilation in patients after tetralogy of Fallot repair. Int J Cardiol. 2011;151:214–7. 14. Chowdhury UK, et  al. Role of fibrillin-1 genetic mutations and polymorphism in aortic dilatation in patients undergoing intracardiac repair of tetralogy of Fallot. J Thorac Cardiovasc Surg. 2008;136:757–66. 15. John AS, Rychik J, Khan M, Yang W, Goldmuntz E. 22q11.2 deletion syndrome as a risk factor for aortic root dilation in tetralogy of Fallot. Cardiol Young. 2014;24:303–10. 16. Senzaki H, et al. Arterial haemodynamics in patients after repair of tetralogy of Fallot: influence on left ventricular after load and aortic dilatation. Heart. 2008;94:70–4. 17. Rathi VK, et  al. Massive aortic aneurysm and dissection in repaired tetralogy of Fallot; diagnosis by cardiovascular magnetic resonance imaging. Int J Cardiol. 2005;101:169–70. 18. Wijesekera VA, et  al. Aortic dissection in a patient with a dilated aortic root following tetralogy of Fallot repair. Int J Cardiol. 2014;174:833–4. 19. Konstantinov IE, Fricke TA, d’Udekem Y, Robertson T.  Aortic dissection and rupture in adolescents after tetralogy of Fallot repair. J Thorac Cardiovasc Surg. 2010;140:71–3.

510 20. Guo T, et al. Genotype and cardiovascular phenotype correlations with TBX1  in 1,022 velo-cardio-facial/ DiGeorge/22q11.2 deletion syndrome patients. Hum Mutat. 2011;32:1278–89. 21. Niwa K, et al. Structural abnormalities of great arterial walls in congenital heart disease: light and electron microscopic analyses. Circulation. 2001;103:393–400. 22. John AS, McDonald-McGinn DM, Zackai EH, Goldmuntz E.  Aortic root dilation in patients with 22q11.2 deletion syndrome. Am J Med Genet A. 2009;149A:939–42. 23. Carlo WF, McKenzie ED, Slesnick TC. Root dilation in patients with truncus arteriosus. Congenit Heart Dis. 2011;6:228–33. 24. Frischhertz BP, Shamszad P, Pedroza C, Milewicz DM, Morris SA. Thoracic aortic dissection and rupture in conotruncal cardiac defects: a population-­ based study. Int J Cardiol. 2015;184:521–7. 25. Gutierrez PS, Binotto MA, Aiello VD, Mansur AJ.  Chest pain in an adult with truncus arteriosus communis. Am J Cardiol. 2004;93:272–3. 26. Gatzoulis MA.  Ross procedure: the treatment of choice for aortic valve disease? Int J Cardiol. 1999;71:205–6. 27. Charitos EI, et  al. Reoperations on the pulmonary autograft and pulmonary homograft after the Ross procedure: an update on the German Dutch Ross registry. J Thorac Cardiovasc Surg. 2012;144:813–21. 28. David TE, et al. Dilation of the pulmonary autograft after the Ross procedure. J Thorac Cardiovasc Surg. 2000;119:210–20. 29. David TE.  Ross procedure at the crossroads. Circulation. 2009;119:207–9. 30. Horer J, et  al. Neoaortic root diameters and aortic regurgitation in children after the Ross operation. Ann Thorac Surg. 2009;88:594–600. 31. Hokken RB, et  al. Does the pulmonary autograft in the aortic position in adults increase in diameter? An echocardiographic study. J Thorac Cardiovasc Surg. 1997;113:667–74. 32. Solymar L, Sudow G, Holmgren D. Increase in size of the pulmonary autograft after the Ross operation in children: growth or dilation? J Thorac Cardiovasc Surg. 2000;119:4–9. 33. Luciani GB, et  al. Fate of the aortic root late after Ross operation. Circulation. 2003;108:S61–7. 34. Simon-Kupilik N, et  al. Dilatation of the autograft root after the Ross operation. Eur J Cardiothorac Surg. 2002;21:470–3. 35. de Sa M, Moshkovitz Y, Butany J, David TE.  Histologic abnormalities of the ascending aorta and pulmonary trunk in patients with bicuspid aortic valve disease: clinical relevance to the ross procedure. J Thorac Cardiovasc Surg. 1999;118:588–94. 36. Takkenberg JJ, et  al. The Ross procedure: a sys tematic review and meta-analysis. Circulation. 2009;119:222–8. 37. Settepani F, et al. The Ross operation: an evaluation of a single institution’s experience. Ann Thorac Surg. 2005;79:499–504.

M. Prapa and M. A. Gatzoulis 38. Schoof PH, et  al. Degeneration of the pulmonary autograft: an explant study. J Thorac Cardiovasc Surg. 2006;132:1426–32. 39. Matsuki O, et al. Two decades’ experience with aortic valve replacement with pulmonary autograft. J Thorac Cardiovasc Surg. 1988;95:705–11. 40. Alsoufi B, et  al. Cardiac reoperations following the Ross procedure in children: spectrum of surgery and reoperation results. Eur J Cardiothorac Surg. 2012;42:25–30. 41. Kincaid EH, Maloney JD, Lavender SW, Kon ND. Dissection in a pulmonary autograft. Ann Thorac Surg. 2004;77:707–8. 42. Rabkin DG, Reid BB, Doty JR.  Acute on chronic pulmonary autograft dissection. Interact Cardiovasc Thorac Surg. 2015;20:563–4. 43. Venkataraman R, Vaidyanathan KR, Sankar MN, Cherian KM. Late dissection of pulmonary autograft treated by valve-sparing aortic root replacement. J Card Surg. 2009;24:443–5. 44. Rabkin-Aikawa E, et  al. Clinical pulmonary autograft valves: pathologic evidence of adaptive remodeling in the aortic site. J Thorac Cardiovasc Surg. 2004;128:552–61. 45. Rabkin-Aikawa E, Farber M, Aikawa M, Schoen FJ.  Dynamic and reversible changes of interstitial cell phenotype during remodeling of cardiac valves. J Heart Valve Dis. 2004;13:841–7. 46. Prapa M, Dimopoulos K.  Arterial switch repair to transposition of great arteries: so far so good. Int J Cardiol. 2012;160:1–3. 47. Losay J, et al. Late outcome after arterial switch operation for transposition of the great arteries. Circulation. 2001;104:121–6. 48. Vandekerckhove KD, et  al. Long-term follow-up of arterial switch operation with an emphasis on function and dimensions of left ventricle and aorta. Eur J Cardiothorac Surg. 2009;35:582–7. 49. Schwartz ML, Gauvreau K, del Nido P, Mayer JE, Colan SD. Long-term predictors of aortic root dilation and aortic regurgitation after arterial switch operation. Circulation. 2004;110:128–32. 50. Hutter PA, et  al. Fate of the aortic root after arterial switch operation. Eur J Cardiothorac Surg. 2001;20:82–8. 51. Bove T, et al. Midterm assessment of the reconstructed arteries after the arterial switch operation. Ann Thorac Surg. 2008;85:823–30. 52. Takeuchi T. Histological analysis of ascending aorta and proximal pulmonary artery in patients with one-­ stage Jatene procedure. Acta Pediatr Cardiol Jpn. 1996;12:506–12. 53. Agnoletti G, et al. Acute angulation of the aortic arch predisposes a patient to ascending aortic dilatation and aortic regurgitation late after the arterial switch operation for transposition of the great arteries. J Thorac Cardiovasc Surg. 2008;135:568–72. 54. Ono M, et al. Valve-sparing operation for aortic root aneurysm late after mustard procedure. Ann Thorac Surg. 2007;83:2224–6.

32  Congenital Aortopathy 55. Nowitz A. Acute ascending aortic dissection 41 years after mustard procedure. J Cardiothorac Vasc Anesth. 2013;27:735–9. 56. Cohen MS, et  al. Neo-aortic root dilation and valve regurgitation up to 21 years after staged reconstruction for hypoplastic left heart syndrome. J Am Coll Cardiol. 2003;42:533–40. 57. Fitzgerald KK, Bhat AM, Conard K, Hyland J, Pizarro C. Novel SMAD3 mutation in a patient with hypoplastic left heart syndrome with significant aortic aneurysm. Case Rep Genet. 2014;2014:591516. 58. van de Laar IM, et  al. Mutations in SMAD3 cause a syndromic form of aortic aneurysms and dissections with early-onset osteoarthritis. Nat Genet. 2011;43:121–6. 59. Erez E, et  al. Valve-sparing aortic root replacement for patients with a Fontan circulation. J Heart Valve Dis. 2012;21:175–80. 60. Pizarro C, Baffa JM, Derby CD, Krieger PA. Valve-­ sparing neo-aortic root replacement after Fontan completion for hypoplastic left heart syndrome. J Thorac Cardiovasc Surg. 2011;141:1083–4. 61. Shuhaiber JH, et al. Repair of symptomatic neoaortic aneurysm after third-stage palliation for hypoplastic left heart syndrome. J Thorac Cardiovasc Surg. 2006;131:478–9. 62. Egan M, Phillips A, Cook SC.  Aortic dissection in the adult Fontan with aortic root enlargement. Pediatr Cardiol. 2009;30:562–3. 63. Dearani JA, Burkhart HM, Stulak JM, Sundt TM, Schaff HV.  Management of the aortic root in adult patients with conotruncal anomalies. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu. 2009;12:122–9. 64. Chong WY, Wong WH, Chiu CS, Cheung YF. Aortic root dilation and aortic elastic properties in children after repair of tetralogy of Fallot. Am J Cardiol. 2006;97:905–9.

511 65. Hiratzka LF, et  al. 2010 ACCF/AHA/AATS/ACR/ ASA/SCA/SCAI/SIR/STS/SVM guidelines for the diagnosis and management of patients with Thoracic Aortic Disease: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, American Association for Thoracic Surgery, American College of Radiology, American Stroke Association, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society of Interventional Radiology, Society of Thoracic Surgeons, and Society for Vascular Medicine. Circulation. 2010;121:266–369. 66. Yetman AT, Graham T.  The dilated aorta in patients with congenital cardiac defects. J Am Coll Cardiol. 2009;53:461–7. 67. Francois K.  Aortopathy associated with congenital heart disease: a current literature review. Ann Pediatr Cardiol. 2015;8:25–36. 68. Geiger J, et  al. 4D-MR flow analysis in patients after repair for tetralogy of Fallot. Eur Radiol. 2011;21:1651–7. 6 9. Baumgartner H, et  al. ESC guidelines for the management of grown-up congenital heart disease (new version 2010). Eur Heart J. 2010;31: 2915–57. 70. Silversides CK, et  al. Canadian cardiovascular society 2009 consensus conference on the management of adults with congenital heart disease: outflow tract obstruction, coarctation of the aorta, tetralogy of Fallot, Ebstein anomaly and Marfan’s syndrome. Can J Cardiol. 2010;26:80–97. 71. Milewicz DM, Dietz HC, Miller DC.  Treatment of aortic disease in patients with Marfan syndrome. Circulation. 2005;111:150–7. 72. Brooke BS, et al. Angiotensin II blockade and aortic-­ root dilation in Marfan’s syndrome. N Engl J Med. 2008;358:2787–95.

Aortic Connective Tissue Histopathology

33

Mary N. Sheppard

33.1 Introduction It is usually medial disease which results in expansion of the wall of the aorta. Three disease processes can destroy the media. First, atheroma, although an intimal inflammatory disease, is associated with widespread medial atrophy; second, noninflammatory degeneration of the media (aortopathy); and, finally, there is inflammatory aortitis of the media and adventitia (aortitis). There are many comorbid abnormalities that are associated with one or more of these conditions, including hypertension, polycystic kidney disease, genetic mutations (such as Marfan syndrome (MFS), Ehlers-Danlos and Loeys-Dietz) and developmental defects such as coarctation, bicuspid aortic valve (BAV), connective tissue disorders (such as lupus, rheumatoid arthritis, scleroderma, ankylosing spondylitis), osteogenesis imperfecta and Turner syndrome, as well as injury.

M. N. Sheppard (*) CRY Department of Cardiovascular Pathology, Cardiovascular Sciences Research Centre, Level 1, Jenner Wing, St. George’s University of London, London, UK e-mail: [email protected]

33.2 Congenital Abnormalities of the Aorta Williams-Beuren syndrome (WBS) is an aortopathy which results in narrowing and thickening of the media which leads to supravalvular aortic stenosis and can also involve the pulmonary artery. It can also lead to hypertension and aortic dissection. The media shows disorganisation of the smooth muscle. WBS is a congenital disorder, which involves the heterozygous deletion of the elastin gene and other genes on chromosome 7. Truncus arteriosus communis is a rare congenital heart defect (CHD), accounting for only 1% of all congenital cardiac abnormalities. It has been associated with other malformations of the heart, mainly truncal valve (bicuspid/quadricuspid) and aortic arch abnormalities such as right, interrupted and hypoplastic aortic arch. Coarctation of the aorta (CoA), a local narrowing of the aortic arch, accounts for 7% of all CHDs. Because surgical repair for coarctation of the aorta has been performed since 1945, growing numbers of patients with repaired coarctation are reaching adulthood. Primary transcatheter intervention with stenting for coarctation emerged as an alternative to surgery since the 1980s. Although short-term outcomes are good after coarctation repair, alterations of vascular form and function persist. Hypertension mediates much of the late morbidity with increased rates of stroke, coronary artery disease and heart failure

© Springer-Verlag GmbH Austria, part of Springer Nature 2019 O. H. Stanger et al. (eds.), Surgical Management of Aortic Pathology, https://doi.org/10.1007/978-3-7091-4874-7_33

513

514

M. N. Sheppard

Fig. 33.1  Shows the arch of the aorta from above. Note narrowing beyond the subclavian with aneurysmal expansion due to a patch insertion which ruptured and killed the patient

after coarctation repair. Prevalence of hypertension in patients with coarctation increases over time, with a majority of patients being affected by middle age. Other late complications include recoarctation, which can usually be addressed with percutaneous balloon dilation and stenting with covered stents. Aneurysms at the coarctation repair site occur (Fig.  33.1). Intracranial aneurysms occur five times more commonly in patients with coarctation than in the general population. Finally, bicuspid aortic valve disease, which is present in at least half of these patients, requires surveillance and ultimately becomes the most common reason for reoperation. The incidence of bicuspid aortic valve (BAV) is highest in patients with coarctation of the aorta (36%) and interrupted aortic arch (36%).

33.3 Aortic Aneurysms Aneurysms can be defined as external bulges of a blood-containing structure, and all aneurysms carry a risk of external rupture. In true aortic aneurysms, the wall is made up of all the constituents of the aortic wall, i.e. intima, media and adventitia. True aneurysms are caused by atherosclerosis, medial degenerative disease or aortitis. They may be confined to a short segment of the aorta and bulge out to one side – these are saccular (Fig. 33.2). Diffuse aneurysms involve the

Fig. 33.2  Shows a saccular aneurysm with a thin-walled expansion in the arch of the aorta

whole circumference of the aorta and often extend over a long distance and are almost always associated with aortic regurgitation, due to dilatation of the aortic root at the level of the supra-­ aortic ridge. Thoracic aortic aneurysms (TAA) are usually due to medial degenerative disease or aortitis, while abdominal aortic aneurysms are associated with atheroma. The great majority of atherosclerotic aneurysms are in the abdomen ­ below the renal artery.

33.3.1 Genetic Causes of Aortic Aneurysms Aneurysms especially thoracic ones are familial in a minority of cases. A number of genetic defects in connective tissue synthesis are known. Inherited disorders are known to affect major arteries, including Marfan syndrome (MS), Ehlers-Danlos syndrome (EDS), Loeys-Dietz, bicuspid aortic valve (BAV) and nonsyndromic familial aortic dissection and osteogenesis imperfecta. Marfan disease is the best known and is the result of defects in the fibrillin gene which interferes with the adhesion of connective tissue and predisposes to aneurysm formation and also dissection. Abdominal aortic aneurysms at a young

33  Aortic Connective Tissue Histopathology

age are a feature of certain rare Marfan families. Marfan disease is an autosomal dominant genetic disorder of connective tissue. The phenotypic expressions of the fibrillin gene disorders are very wide with mixtures of cardiac, skeletal and ocular abnormalities. A specific genetic locus encoding fibrillin 1 is responsible for Marfan syndrome (MFS), with the development of ascending thoracic aneurysms. Fibrillin is a 350Kd glycoprotein that is the major component of the 12 mm extracellular microfibrils which act as a network for elastin deposition and is a constituent of the elastic fibre. In aortas from patients with MFS, elastin is deficient in desmosine cross-­ linking residues, and the content of elastin is decreased by almost 50% with structural alterations of elastin fibres being characterised by enlarged interlaminar spaces (between elastin laminae) and loss of interlaminar elastin fibrils. This loss of elastin content and decrease in cross-­ linking explain the higher prevalence of MFS patients having aortic aneurysm. The gene abnormalities produce a wide range of results ranging from virtually all the fibrillin being the mutant type to families in whom the amount of mutant fibrillin produced is less than 10%, the rest being the wild type, due to the normal gene inherited from the normal parent. The result is a huge phenotypic range of severity which includes fully developed severe skeletal and cardiac disease through the MAAS syndrome (mild aortic and mitral regurgitation with minimal skeletal abnormality) to familial aortic dissection without any skeletal manifestations. Collagen is the main matrix protein resisting expansile pressure in the aorta. Defects in the synthesis of type I collagen occur in osteogenesis imperfecta in which there is often generalised aortic dilatation with diffuse calcification. Type IV Ehlers-Danlos syndrome is due to a defect in type III collagen and also leads to aneurysm formation and spontaneous rupture in the aorta and other large arteries. A number of groups have now identified gene defects in type III collagen which are predominantly expressed as aortic aneurysm formation with minimal or no systemic abnormalities. These nonsyndromic familial thoracic aortic aneurysms and dissections (TAAD)

515

are inherited in families as an autosomal dominant disorder with a variable age of onset of the aortic disease. Studies have mapped genes causing nonsyndromic familial TAAD to 5q13-15 (TAAD1) and 11q23.2-q24 (FAA1). A third locus for nonsyndromic TAAD was mapped to 3p24-­ 25 and termed the TAAD2 locus. This locus overlaps a previously mapped second locus for Marfan syndrome, termed the MFS2 locus. Missense mutations in ACTA2 are responsible for 14% of inherited ascending thoracic aortic aneurysms and dissections. The mean age at presentation for patients with these syndromes is significantly younger than the mean age of presentation in sporadic cases and significantly older than that of patients with Marfan syndrome. Patients with a family history of aortic aneurysms had faster growth rates compared with patients with sporadic TAA and patients with Marfan syndrome. Classical Ehlers-Danlos syndrome (cEDS) is a rare connective tissue disorder primarily characterised by hyperextensible skin, defective wound healing, abnormal scars, easy bruising and generalised joint hypermobility; arterial dissections are rarely observed. Mutations in COL5A1 and COL5A2 encoding type V collagen account for more than 90% of the patients so far characterised. In addition, cEDS phenotype was reported in a small number of patients carrying the c.934C>T mutation in COL1A1 that results in an uncommon substitution of a non-glycine residue in one Gly-Xaa-Yaa repeat of the proalpha1(I)-chain p.(Arg312Cys), which leads to disturbed collagen fibrillogenesis due to delayed removal of the type I procollagen N-propeptide. This specific mutation has been associated with propensity to arterial rupture in early adulthood; indeed, in literature the individuals harbouring this mutation are also referred to as “(classic) vascular-like” EDS patients. Loeys-Dietz syndrome (LDS) is a disorder of connective tissue which shares overlapping features with Marfan syndrome (MFS) and the vascular type of Ehlers-Danlos syndrome, including aortic root dilatation and skin abnormalities. It is clinically classified into types 1 and 2. LDS type 1 can be recognised by craniofacial

516

c­ haracteristics, e.g. hypertelorism, bifid uvula or cleft palate, whereas these are absent in LDS type 2. It is important to recognise LDS because its vascular pathology is aggressive: characteristics found in the majority of these LDS patients are aortic root dilatation, cleft palate and/or a bifid/abnormal uvula. Because aortic dissection and rupture in LDS tend to occur at a young age or at aortic root diameters not considered at risk in MFS, and because the vascular pathology can be seen throughout the entire arterial tree, patients should be carefully followed up and monitored [1]. At the present time, however, the majority of aneurysms due to medial degenerative disease are of unknown cause. The consensus view is that these patients have an arterial wall that is predisposed to dilate in the presence of normal or moderately elevated haemodynamic stress  – this predisposition is even further unmasked as age-­ related changes occur in the media.

M. N. Sheppard

Fig. 33.3  Shows a jagged irregular transverse tear in the ascending aorta above the aortic valve and sinotubular junction. Note the dissection with an intramural haematoma in the ascending aorta

33.4 Dissecting Aneurysms Dissecting aneurysms of the aorta (better termed aortic dissections) are very different. This is the result of an intimal tear which allows blood to enter the media and form a track which extends both proximally and distally leading to enlargement of the aorta. All aortic dissections are characterised by a separation of the media with tracking of blood longitudinally. The dissection track is in the outer third of the media, in the plane of the capillary arcade. The majority of dissection tracks ultimately break outward and rupture through the adventitia. The intimal tear and the external exit point are often widely separated. An intimal tear is identified in the majority of cases in the ascending aorta (Fig. 33.3), usually about 2 cm above the aortic valve. The majority of tears are transverse, although they may be T-shaped or vertical. The intimal tear is located in the aortic arch in about 10% of cases, the descending aorta in 25% of cases and rarely in the abdominal aorta. Proximal extension of the dissection into the pericardium results in haemopericardium and death. Propagation of the

Fig. 33.4  This shows a transverse section of the ascending aorta with an intramural haematoma. Note the surrounding haemorrhage and expansion of the aortic root

intramural haematoma progresses distally as far as branching points and often involves the head and neck vessels. The development of this haematoma results in a false lumen full of blood with expansion and weakening of the aortic wall resulting in aneurysm formation (Fig.  33.4). It can also lead to blockage of the vessel due to the

33  Aortic Connective Tissue Histopathology

inward bulge of the haematoma with ischaemic damage to the brain, bowel, kidney and limbs. An intimal-medial tear with the formation of an intramural haematoma can have several consequences: 1. External rupture of the intramedial haematoma occurs. This is common in dissection of the ascending aorta because the external wall is very thin consisting mainly of adventitia with just a thin layer of outer media. This rupture results in massive haemorrhage into the mediastinum, pleural cavities or pericardium resulting in the sudden onset of chest pain and rapid death. 2. A re-entry tear may develop in the intima distal to the primary intimal tear leading to two aortic lumens. The frequency of external rupture in the ascending aorta means at this site re-entry is rare; about 10% of dissections on the abdominal aorta, however, develop a re-­ entry intimal tear. 3. The development of a chronic aneurysm with thrombus formation in the track which is still communicating with the lumen to some extent. 4. A small localised dissection may heal leaving a transverse or longitudinal U-shaped depression in the intima and media. 5. Dissecting aneurysm may produce stenosis of branches of the aorta due to extension of the haematoma into the media with occlusion of the vessel lumen and distal ischaemia. 6. Ten percent of all acute aortic dissections will progress to a chronic or healed phase. Many of these cases have a re-entry site in the abdominal aorta or distal dissections. The double-barrel aorta has a false channel that is often larger than the true lumen, so that the term aneurysm is more appropriate here than in acute dissections where there is often little dilatation. The lining of the false channel shows fibromuscular thickening. Mural thrombi in the channel may become organised, forming thrombo-atherosclerotic plaques. Calcification may become prominent in the wall of the false channel, and sometimes the lumen may be completely occluded

517

by thrombus. Healed dissections can be compatible with long survival and are occasionally discovered in patients as coincidental findings at autopsy, indicating that an acute dissection may be silent. Various types of classification of dissection of the aorta have been made, largely to allow comparison of different surgical repair tech­ niques and series. It should be emphasised that these classifications are purely for description relative to surgical treatment and do not imply any basic pathogenetic differences.

33.4.1 Aortic Medial Disease Dissecting aneurysms and ascending aortic aneurysms are usually caused by medial disease. Degenerative changes in the aortic media result in destruction and fragmentation of the elastic fibres and an increase in mucopolysaccharide (MPS) ground substance (Figs.  33.5 and 33.6).

Fig. 33.5  This histological section of the aorta shows destruction and collapse of the elastic fibres which are stained black with replacement by collagen which is stained red. Trichrome stain

518

M. N. Sheppard

aortic pulse p­ ressure. In the most recent pathological consensus document, overall degenerative alterations/damage to the aortic media results from eight individual histopathologic degenerative lesions affecting the lamellar unit (both cellular and extracellular) as observed on H&E and stains to highlight elastic fibres and extracellular matrix material (such as Verhoeffvan Gieson, Alcian blue, Movat’s pentachrome, Masson’s trichrome). Description and grading of these changes have been published recently by both European and American pathologists in order to come to a consensus on the histological changes [3].

Fig. 33.6  This histological section of the aorta shows accumulation of mucopolysaccharides which is stained blue. Alcian blue staining

Degenerative medial changes occur at variable sites, proximal more than distal in the aorta and in the outer curvature more than inner curvature, but marked differences may also occur randomly so that extensive sampling is required when examining these specimens from surgery [2]. The earliest terminology of degenerative aortic histopathology was the classic description of medionecrosis by Erdheim in 1930. Since then, different groups have used a variety of terms to describe the histopathologic changes in the ascending aortas affected by aneurysms and ageing. Unfortunately, no specific terms have gained full acceptance, and often a common term – such as “cystic medial degeneration/necrosis” – indicates different histopathological features in different studies. All aortic diseases are associated with microstructural changes, either to the content or architecture of the connective fibres elastin or collagen. It is alteration of the quantity and/or architecture of these fibres that leads to the mechanical, and hence functional, changes associated with aortic disease. Structural alterations in the walls of large arteries with progressing age cause a decrease in the total arterial compliance, which in turn leads to both a decreased distal blood flow and an increase in

33.4.2 Risk Factors for Aortic Dissection Epidemiological studies using case control methods and family studies clearly show that the risk of dissection is greatly increased in older hypertensive patients. Contrary to popular misconception, atheroma and inflammatory aortitis do not often cause dissection. The means by which these risk factors operate are only partially understood. Hypertension operates by increasing the mechanical forces which are imposed on the aorta but only as a potentiating factor on an aorta with some prior medial abnormality. The higher risk of aortic dissection in association with bicuspid aortic valves is well established. It has often been postulated that asymmetric high-velocity jets through bicuspid valves might alter stress distribution in the first part of the ascending aorta. There is probably a primary defect in the development of the aorta and cusps in BAV which is either at a structural or molecular level. The concept that there is a molecular defect responsible for aortic dissection is most strongly supported by Marfan disease, in which the risk is very high and 80% of subjects will ultimately die of cardiac disease. Taking dissecting aneurysms of the aorta overall, more than 80% occur over the age of 60  years, and hypertension is the only discernible risk factor. In patients under 40 years, dissection is strongly related to genetic defects of

33  Aortic Connective Tissue Histopathology

519

connective tissue synthesis such as Marfan disease. It is critically important to obtain a family history of thoracic aortic aneurysms and dissections, along with unexplained sudden death, when assessing an individual with possible aortic disease.

33.5 Inflammatory Medial Disease Aortitis is a rare but important cause of thoracic aortic disease [4]. This typically involves the ascending aorta and causes aneurysms. The archetypal cause of inflammatory medial disease used to be syphilis, but today that is rare. The aortic media has focal areas of medial destruction in which the elastic laminae are collapsed or absent (laminar medial necrosis), and surrounding these areas are aggregates of perivascular lymphocytes and plasma cells with or without giant cells. The inflammatory process extends from the adventitia along the vasa vasorum which shows marked intimal thickening. The areas of destruction in the media ultimately are converted to fibrous scars. The overlying intima develops irregular thickening due to smooth muscle proliferation as does the adventitia. In very acute syphilis, microgummata with small giant cells and central necrosis occurs, but it is far more usual to see the later more chronic stages of the aortitis in which they are absent and which have a lymphoplasmacytic picture only with transmural fibrosis. Very occasionally microorganisms can be found in the chronic stage.

33.5.1 Giant Cell Aortitis This is usually a disease of the elderly with an incidence of 15–30 cases per 100,000 persons over the age of 50 years. It is more common in women and black people. In temporal giant cell arteritis, extracranial arteritis and aortitis occur in 10–15% of patients and have been underestimated in the past. The aorta can look exactly similar to the tree bark effect seen in the intima in syphilis. The aorta may dilate leading to ­aneurysm

Fig. 33.7  This histological section of the aorta shows fibres being surrounded by epithelioid cells and giant cells in the media. Haematoxylin and eosin staining

formation and incompetence of the aortic valve. In the media, there is a particular pattern of inflammation in which a collapsed band of elastic is surrounded by inflammatory cells, which are mainly lymphocytes, epithelioid cells and multinucleate giant cells at the edge of this band (Fig.  33.7). Nonspecific chronic inflammatory cells including lymphocytes and plasma cells at the medial/adventitial junction may be seen in the chronic healing stage and make the appearance indistinguishable from other causes of aortitis. Aortitis is clinically categorised into groups that include Takayasu disease, giant cell aortitis and isolated aortitis. Aortitis can occur on a background of immune-mediated disease such as rheumatoid arthritis, lupus, ankylosing spondylitis and scleroderma and may progress despite aggressive immunosuppression regimens. The histological appearances are similar to the isolated form. Immunoglobulin 4-related disease (IgG4related disease) is a systemic inflammatory disease that presents with increases of serum IgG4. It may affect various systems, including the cardiovascular (CV) system. Assessment of serum IgG4 levels and involved organ biopsy is necessary for diagnosis. IgG4-related disease is characterised by fibrosclerosis, lymphocytic ­

520

infiltration and presence of IgG4-positive plasma cells. CV involvement may manifest as cardiac pseudotumors, inflammatory periaortitis, coronary arteritis and/or pericarditis. In case of active periarterial or coronary artery inflammation, 18FDG-PET will show FDG uptake at the area of the lesion [5]. Noninfectious aortitis is far more common today than infectious aortitis. Noninfectious aortitis is predominantly idiopathic/isolated in nature, occurring in elderly females. Lymphoplasmacytic aortitis can occur in association with ankylosing spondylitis, rheumatoid arthritis, systemic lupus, Behçet’s disease and in any of the systemic vasculitis/collagen diseases, or it can be idiopathic. Three histological patterns emerge  – necrotising aortitis with giant cells. In giant cell aortitis, there are plates of totally acellular medial tissue in which giant cells are present at the ends of the broken collapsed elastic laminae. There is the pattern of medial diffuse lymphoplasmacytic band-like aortitis and a mixed picture. Cases of necrotising aortitis with giant cells are usually idiopathic, while diffuse lymphoplasmacytic aortitis is associated with a systemic inflammatory disease. Medial degenerative change can be seen with all types of aortitis: knowledge of histopathological patterns may guide patient management and follow-up [4]. IgG4-related sclerosing disease can occur in the cardiovascular system, and both thoracic and inflammatory abdominal aortic aneurysms have been shown to belong to IgG4-related sclerosing disease. There are conspicuous fibrosclerotic changes, dense lymphoplasmacytic infiltration and occasional obliterative phlebitis in the adventitia. Immunohistochemistry shows numerous IgG4-positive plasma cell infiltrates. There is frequently a good response of IgG4related systemic disease to glucocorticoid treatment without additional therapy. Treatment of the aortitis may prevent progression of the IgG4related systemic disease to involvement of other organs such as the pancreas and liver and abdominal aorta leading to retroperitoneal fibrosis. Eosinophilic aortitis can occur in association with systemic Churg-Strauss syndrome but is very rare [6].

M. N. Sheppard

33.6 Atheroma/Atherosclerotic Aortic Aneurysms The aorta is a primary site for the development of atheroma/atherosclerosis. Examination of the intimal surface shows the full range of plaques from fatty streaks to ulcerated plaques covered by thrombus. The number of aortic plaques on a population basis is an indicator of the risk of ischaemic heart disease in that geographic population. Risk factors such as diabetes, hypertension and smoking are associated with a greater number of aortic plaques when compared to control subjects. The size of the lumen of the aorta is, however, such that thrombosis occurring over plaques is not a common cause of clinical symptoms. Plaque thrombosis in the aorta is initiated by loss of the cap over a lipid-rich plaque. This exposes the lipid core and a polyploid mass of lipid and thrombus forms over the plaque. Ultimately the thrombus is lysed or embolised downstream leaving shallow depressed areas in the intima which become recovered by endothelium. Rarely large occlusive thrombi can develop. Calcification can be extensive in these plaques. Pathological studies of the development of aortic atherosclerosis, looking at different age groups from infancy onward, show that atherosclerosis first develops in the abdominal aorta and is always present there to a greater degree than in the rest of the aorta. The lipid-filled plaques begin in the intima and spread into the media resulting in destruction and weakening of the aortic wall. Saccular and diffuse abdominal aortic aneurysms virtually always occur in the segment between the renal artery orifices and the aortic bifurcation. Indeed this is the commonest site of aortic aneurysms overall. In more peripheral vessels, rupture can give rise to emboli and stroke particularly in the cerebral circulation. Atherosclerotic abdominal aortic aneurysms are usually fusiform and involve the whole circumference of the aorta. The aneurysm may however be asymmetric and bulge more to one side than the other. The diameter can be anything from 7 to 30 cm. The wall consists of dense hyaline fibrous tissue with a lining of laminated old and recent thrombus. Abundant lipid-­ filled macrophages are mixed with the thrombus,

33  Aortic Connective Tissue Histopathology

and the adjacent aorta shows extensive atherosclerosis. It is usually very difficult to find any residual elastic tissue in the aneurysm wall. The natural history of abdominal aortic aneurysms is to expand, and the rate is accelerated by hypertension.

33.6.1 Pathogenesis of Atherosclerotic Aortic Aneurysms The risk factors for abdominal aortic aneurysms are somewhat different than for atherosclerosis in general. Atherosclerotic aneurysms are nine times more common in men than women and are strongly related to smoking and hypertension but less strongly to hyperlipidaemia than coronary artery disease. Abdominal aortic aneurysms in males are very clearly familial. For an index case, there is a 15% chance of an offspring developing an abdominal aortic aneurysm and up to a 30% chance in siblings set against a risk of 2–5% prevalence in the general population.

33.7 Aortic Periaortitis and Aneurysms The concept of chronic periaortitis is that there is a large adventitial component of inflammation in some cases of aortic atherosclerosis with intimal proliferation, medial thinning and adventitial fibrosis. The total aortic wall thickness may be markedly increased. The histological features are like those of conventional atherosclerotic aneurysms with the addition of large amounts of hyaline periaortic fibrous tissue within which are a large number of plasma cells and often lymphoid follicle with germinal centres. The fibrous tissue may be very cellular in areas with a whorled pattern. On computerised tomography there is a large abdominal periaortic fibrotic mass with or without aortic dilatation. The inflammation is presumed to be secondary to atherosclerosis which is often severe and advanced. Periaortitis has also been linked to retroperitoneal fibrosis in which there is a retroperitoneal mass with u­ reteric

521

obstruction. There is a view that the abdominal aortic aneurysm with this degree of periadventitial inflammation is primarily inflammatory, with atherosclerosis as a secondary phenomenon. Studies have shown shrinkage of the periaortic mass in response to steroid therapy in idiopathic retroperitoneal fibrosis and in aneurysms as well as spontaneous shrinkage with time. Routine autopsy histology suggests that 40% of the population over the age of 50  years will have some degree of periarterial and periaortic inflammation representing subclinical periaortitis and the clinical incidence may be in the order of 0.4. This form of thick-walled aneurysm may comprise up to 10% of lower abdominal aortic aneurysms and have a slightly lower risk of rupture but increased risk of ureteric obstruction coupled with a greater difficulty at surgery.

33.8 Traumatic Aneurysm of the Aorta Aortic false aneurysms are virtually always due to either penetrating trauma such as knife wounds or chest trauma or previous aortic surgery. The external bulge has a wall consisting of adventitia or fibrous tissue. Complete aortic transections are becoming more common due to high-velocity impact injuries and account for up to a fifth of the mortality in vehicle accidents. The great majority of these complete full-thickness transverse tears are just distal to the left subclavian artery at the isthmus of the aorta. About 20% are in the ascending aorta. Multiple tears are also common. A very small proportion of traumatic aortic tears are partial and lead to a subadventitial haematoma. This then passes into an aneurysm of the “false” type. Such aneurysms develop by about 3  months after the trauma and may persist for years. Most will ultimately rupture to cause death, unless treated surgically. These late post-­ traumatic aneurysms are often very discrete and saccular in type. The operative specimen should be examined histologically in blocks which pass from the normal aorta through the edge of the aneurysm sac. The characteristic appearance is of a normal media with a very abrupt change to the

522

fibrous wall of the aneurysm which contains no elastic tissue. The abrupt line of medial loss is the original site of the tear.

M. N. Sheppard

2. Stone JR, Basso C, Baandrup UT, Bruneval P, Butany J, Gallagher PJ, et al. Recommendations for processing cardiovascular surgical pathology specimens: a consensus statement from the standards and definitions Committee of the Society for Cardiovascular Pathology and the Association for European Cardiovascular Pathology. Cardiovasc Pathol. 33.9 Tumours of the Aorta 2012;21:2–16. 3. Halushka MK, Angelini A, Bartoloni G, Basso C, These are extremely rare and occur in elderly Batoroeva L, Bruneval P, et  al. Consensus statement on surgical pathology of the aorta from the males. Most are undifferentiated sarcomas. Society for Cardiovascular Pathology and the Angiographically the tumour may present as a Association For European Cardiovascular Pathology: thoracic or abdominal aortic aneurysm, or there II. Noninflammatory degenerative diseases – nomenmay be obstruction of a branch due to ingrowth clature and diagnostic criteria. Cardiovasc Pathol. 2016;25:247–57. of the tumour. Tumours arising in the media usually invade locally, while intimal lesions grow 4. Ryan C, Barbour A, Burke L, Sheppard MN.  Non-­ infectious aortitis of the ascending aorta: a histoalong the lumen and give rise to thromboemboli logical and clinical correlation of 71 cases including in peripheral vessels. overlap with medial degeneration and atheroma  – a challenge for the pathologist. J Clin Pathol. 2015;68:898–904. 5. Mavrogeni S, Markousis-Mavrogenis G, Kolovou References G.  IgG4-related cardiovascular disease. The emerging role of cardiovascular imaging. Eur J Radiol. 2017;86:169–75. 1. Aalberts JJ, Van den Berg MP, Bergman JE, du Marchie Sarvaas GJ, Post JG, van Unen H, et al. The 6. Segal OR, Gibbs JS, Sheppard MN. Eosinophilic aortitis and valvitis requiring aortic valve replacement. many faces of aggressive aortic pathology: Loeys-­ Heart. 2001;86:245. Dietz syndrome. Neth Heart J. 2008;16:299–304.

Clinical Aspects of Heritable Connective Tissue Disorders

34

Aline Verstraeten and Bart Loeys

34.1 Introduction Heritable connective tissue disorders were initially classified by Victor McKusick in his book Heritable Disorders of Connective Tissue, which was first published in 1956 [1]. Since then, this group of disorders has grown enormously. The organ systems that are most prominently involved are the integumentum (skin hyperextensibility, atrophic scarring, hernias), the eye (ectopia lentis, myopia), the skeleton (overgrowth, pectus deformities, scoliosis, clubfeet, joint hypermobility, or contractures), and the cardiovascular system (aortic ­aneurysm/dissection, valve disease). In this chapter we will focus on those connective tissue disorders with aortic involvement (Table 34.1).

34.2 Marfan Syndrome Probably the most common connective tissue disorder with prominent aortic involvement is Marfan syndrome (MFS). MFS has an estimate

A. Verstraeten (*) · B. Loeys Center for Medical Genetics, Antwerp University Hospital and University of Antwerp, Antwerp, Belgium e-mail: [email protected]; [email protected]

prevalence of 1/3000–1/5000. The disease is characterized by high penetrance but has important inter- and intrafamilial variability. This pleiotropic condition presents with manifestations in different organ systems, most importantly the cardiovascular, ocular, and musculoskeletal system. The cardiovascular manifestations can be life-­ threatening and include most importantly dilatation of the aortic root at the level of the sinuses of Valsalva, which if left untreated can lead to dissection and rupture of the aortic root. Overall, 70–80% of MFS patients will develop aortic root aneurysms throughout their lives [2, 3]. Later in life, and often after root surgery, MFS patients can also develop descending aortic aneurysms and dissections [4, 5]. Other common cardiovascular findings include dilatation of the pulmonary arteries, myxomatous valve changes causing valve insufficiency, and progressive left ventricular dysfunction. The ocular symptoms are lens dislocation, retinal detachment, progressive myopia, cataract, and glaucoma, leading to visual impairment. Ectopia lentis or lens dislocation is present in 30–60% of all MFS patients [6–9]. The most typical musculoskeletal findings are arachnodactyly, disproportional long limbs (with arm span to height ratios above 1.05), scoliosis, pectus deformities (both excavatum and carinatum), pes planus, and joint hypermobility. Other recurrent findings include a highly arched and narrow

© Springer-Verlag GmbH Austria, part of Springer Nature 2019 O. H. Stanger et al. (eds.), Surgical Management of Aortic Pathology, https://doi.org/10.1007/978-3-7091-4874-7_34

523

A. Verstraeten and B. Loeys

524 Table 34.1  Overview of connective tissue diseases with aortic involvement Syndrome Marfan syndrome (MFS)

Gene(s) FBN1 (MIM154700)

Loeys-Dietz syndrome (LDS)

TGFBR1 (MIM609192) TGFBR2 (MIM610168) SMAD3 (MIM613795)

Clinical features Cardiovascular Aortic root aneurysm, dilatation of pulmonary arteries, myxomatous valve changes, progressive cardiac dysfunction Thoracic aortic aneurysm, patent ductus arteriosus, widespread aortic and arterial aneurysms/tortuosity Arterial aneurysms/ dissections/tortuosity

SMAD2 (MIM601366)

Arterial aneurysms/ dissections/tortuosity

Filaminopathy

Meester-Loeys syndrome (MRLS)

Thoracic aortic aneurysm, arterial tortuosity, patent ductus arteriosus, bicuspid aortic valve Thoracic aortic aneurysm, bicuspid aortic valve, mitral valve disease BGN (MIM300989) Thoracic aortic aneurysm, mild valve insufficiency

TGFB2 (MIM614816) TGFB3 (MIM615582) FLNA (MIM300537)

Autosomal dominant ELN (MIM123700) Thoracic aortic aneurysm, cutis laxa (ADCL) pulmonary artery stenosis Autosomal recessive cutis laxa (ARCL) Arterial tortuosity syndrome (ATS) Congenital contractural arachnodactyly (CCA)

EFEMP2 (MIM614437) SLC2A10 (MIM208050) FBN2 (MIM121050)

Thoracic aortic aneurysm, arterial tortuosity Diffuse and severe arterial tortuosity/aneurysms Mitral valve prolapse, rare arterial aneurysms

palate, osteopenia, skeletal muscle and adipose tissue hypoplasia, dural ectasia, striae distensae, inguinal hernia, and pneumothorax or pulmonary emphysema. In 1991, Dietz and colleagues pinpointed FBN1 (MIM 154700) mutations (encoding for fibrillin-1) as the underlying molecular cause of MFS [10]. Fibrillin-1 is an important component of the microfibrils of the extracellular matrix. About 1850 different FBN1 mutations have meanwhile been described, and these encompass

Clinical features Non-cardiovascular Ocular: lens dislocation, retinal detachment, cataract, etc. Musculoskeletal: arachnodactyly, disproportional long limbs, scoliosis, pectus excavatum/carinatum Marfanoid features, hypertelorism, cleft palate, bifid uvula, craniosynostosis, clubfeet, dystrophic scarring, easy bruising Early osteoarthritis, hypertelorism, abnormal/bifid uvula/palate, arachnodactyly, scoliosis, striae, velvet skin, osteochondritis dissecans Osteoarthritis, palate abnormalities, pes planus, arachnodactyly, easy bruising, and striae Marfanoid features, hypertelorism, clubfeet, bifid uvula

Joint hypermobility, skin hyperextensibility, coagulopathy, seizures Joint hypermobility and contractures, striae, downslanting palpebral fissures, frontal bossing, hypertelorism, proptosis Loose redundant skin, genital prolapse, gastrointestinal diverticula/hernias, emphysema Loose redundant skin, gastrointestinal and urinary diverticula, emphysema Hypertelorism, bifid uvula/palate, arachnodactyly Joint contractures, kyphoscoliosis, marfanoid features

all types of mutations: insertions, large and small deletions, and point and splice mutations [11]. They mostly inherit in an autosomal dominant manner, but in about 25% of patients, the mutation arose de novo. For more details on the genetic basis of MFS, we refer to Chap. 38. In 2010 an international expert panel defined a set of criteria for making the diagnosis of MFS [12]. In absence of a positive family history, there are four possible situations which lead to an MFS diagnosis:

34  Clinical Aspects of Heritable Connective Tissue Disorders

1. Aortic root dilatation/aneurysm (Z-score > 2) and ectopia lentis 2. Aortic root dilatation/aneurysm (Z-score > 2) and a FBN1 mutation 3. Aortic root dilatation/aneurysm (Z-score > 2) and a systemic score ≥ 7 4. Ectopia lentis and a FBN1 mutation with known aortic root dilatation/aneurysm The systemic score involves a number of criteria that according to their specificity get 1, 2, or 3 points; and a maximum score of 20 can be obtained (for details see reference [12]). Management of MFS is based on two pillars. First, medical treatment with beta-blockers or angiotensin receptor blockers aims to slow down the rate of progression of aortic root enlargement. If the aortic root reaches diameters between 4.5 and 5 cm, a preventive root replacement surgery is performed. Over recent years, the surgical threshold has been lowered because of the availability of better aortic valve-sparing root replacement techniques.

34.3 Loeys-Dietz Syndrome Both clinically and molecularly, Loeys-Dietz syndrome (LDS) is closely related to MFS. In its most typical presentation, LDS is characterized by the triad of aortic aneurysms in the entire arterial tree accompanied by marked tortuosity, hypertelorism, and a bifid uvula/cleft palate [13]. Soon after the condition’s initial description, it was shown that there are also LDS forms displaying fewer outward features and findings reminiscent of Ehlers-Danlos syndrome [14]. As in MFS, LDS is characterized by widespread involvement of different organ systems. Typical craniofacial signs in addition to hypertelorism and bifid uvula/ cleft palate are malar flattening, retrognathia, and craniosynostosis in the more severely affected patients. Skeletal features demonstrate significant overlap with MFS (i.e., pectus excavatum/ carinatum, scoliosis, arachnodactyly, joint laxity, and contractures), but in general the overgrowth

525

is less pronounced. Both talipes equinovarus and cervical spine malformation and/or instability seem more prevalent in LDS.  Skin and integumentum abnormalities present as thin velvety translucent skin, dystrophic scars, hyperextensibility, easy bruising, and dural ectasia, while ocular involvement leads to strabismus, blue sclerae, and myopia. Symptoms distinguishing LDS from MFS are craniosynostosis, hypertelorism, cleft palate/bifid uvula, clubfeet, cervical spine instability, and, most predominantly, widespread arterial aneurysms with tortuosity and early rupture. Oppositely, one of the most important differentiating MFS features is lens dislocation. Surgical repair of LDS-related aortic root aneurysms should be “considered” with measurements around 4  cm, but factors such as family history, historical knowledge about the specific genotype (gene and/or mutation), severity of systemic findings, and the patient’s personal risk-benefit assessment should influence the decision [15]. Although aortic aneurysms in LDS were initially described to be highly aggressive and to rupture at young ages, even at smaller diameters than in MFS, we observed since a broader phenotypical spectrum. Patients with more outward features of LDS tend to have a more severe cardiovascular presentation. More research will be necessary to define the precise cardiovascular risk, but in the meantime we still recommend more complete imaging of the arterial tree in LDS patients. LDS follows an autosomal dominant inheritance pattern with variable clinical expression. Causal mutations have been identified in genes encoding key components of the TGFbeta signaling pathway: TGFBR1/2, SMAD2/3, and TGFB2/3 [13, 16–19]. About two thirds of the mutations are de novo [20]. They are responsible for the majority of cases with severe craniofacial and skeletal findings. The remaining one third of patients has a positive family history for the disease and generally presents with a milder phenotype with regard to systemic findings. For an extensive review of the genetic basis of LDS, we refer to Chap. 38.

526

Genetic testing should be considered in a context of: 1 . Presentation with the typical clinical triad 2. Early-onset aneurysms with a variable combination of clubfeet, easy bruising, joint hypermobility, arachnodactyly, camptodactyly, blue sclera, thin skin with atrophic scars, craniosynostosis, bicuspid aortic valve (BAV), patent ductus arteriosus (PDA), and atrial/ ventricular septum defects 3. MFS-like phenotype, especially in the absence of ectopia lentis, but with aortic and skeletal features that are not fulfilling the MFS diagnostic criteria 4. Families with autosomal dominant thoracic aortic aneurysms, especially with precocious aortic/arterial dissections or aortic disease beyond the aortic root 5. Phenotype of vascular Ehlers-Danlos syn drome and normal type III collagen biochemistry 6. Isolated young proband with aortic root dilatation/dissection

34.4 Ehlers-Danlos Syndrome Ehlers-Danlos syndrome (EDS) reconciles a group of connective tissue disorders that is clinically and genetically heterogeneous. Common denominators of all subtypes are abnormalities of the skin, ligaments and joints, blood vessels, and internal organs, with the most typical features being joint hypermobility, skin hyperextensibility, and fragility, as well as of the aortic/arterial walls and internal organs. About 25% of EDS patients show aortic aneurysmal disease [21]. The classic and hypermobile EDS forms account for over 90% of EDS cases. Vascular EDS represents less than 5%. All other subtypes are relatively to extremely rare. Different EDS subtypes were for the first time officially determined and numbered in the original international nosology of heritable connective tissue disorders of 1986 [22]. The Villefranche

A. Verstraeten and B. Loeys

nosology, dating from 1997, recognized six different subtypes and named the subtypes based on the typical clinical characteristics [23]. The latest nosology, proposed in 2017, delineates 13 subtypes and emphasizes the importance of molecular identification of causative variants (Table 34.2) [24]. EDS is caused by mutations in genes coding for collagens or proteins involved in collagen processing. Following the latest nosology, several conditions have been abandoned from the current EDS classification. These encompass the previously called filamin A-related EDS with periventricular nodular heterotopia, in which vascular rupture sporadically is observed, but also X-linked EDS with muscle hematoma, fibronectin-deficient EDS, and occipital horn syndrome (which is a copper deficiency syndrome). The 13 most commonly recognized EDS subtypes are summarized in Table  34.2. For each subtype major and minor criteria are proposed. The major criteria involve features that occur in the majority of patients and are fairly specific for the given subtype, while the minor criteria involve features that may overlap between EDS subtypes or with other heritable connective tissue diseases and might only be present in a small subset of patients. Notably, many clinical findings overlap between the different EDS subtypes, which makes clinically diagnosing EDS rather difficult and highlights the importance of molecular diagnostic confirmation. Subtypes with common cardiovascular findings are vascular EDS (vEDS), cardiac-­valvular EDS (cvEDS), and kyphoscoliotic EDS (kEDS). Patients with vEDS often die as a consequence of arterial rupture (which is not always preceded by an aneurysm), making vEDS the most life-threatening disease subtype. Most typically, rupture occurs in the midsized arteries, but vEDS patients can present with manifestations throughout the entire vascular tree. vEDS has an autosomal dominant inheritance pattern and is caused by mutations in the COL3A1 gene, which encodes the procollagen type 3 alpha 1 chain [25]. The latter protein homotrimerizes into type 3 c­ ollagen,

34  Clinical Aspects of Heritable Connective Tissue Disorders

527

Table 34.2  Ehlers-Danlos subtypes [1] Inheritance pattern Gene(s) AD COL5A1/ COL5A2/ COL1A1 Vascular EDS (vEDS) AD COL3A1/ COL1A1 Arthrochalasia EDS (aEDS) AD COL1A1/ COL1A2 Dermatosparaxis EDS AR ADAMTS2 (dEDS) Cardiac-valvular EDS AR COL1A2 (cvEDS) Kyphoscoliotic EDS (kEDS) AR PLOD1/ FKBP14 Classical-like EDS (clEDS) AR TNXB EDS subtype Classical EDS (cEDS)

Major clinical features Skin hyperextensibility, joint hypermobility

Thin/translucent skin, characteristic facial appearance, vascular fragility Severe joint hypermobility, congenital hip dislocation, skin hyperextensibility Extreme skin fragility, mild joint hypermobility, characteristic facial features Severe cardiac valvular defects, joint hypermobility, skin hyperextensibility Kyphoscoliosis, joint laxity, muscle hypotonia Skin hyperextensibility, joint hypermobility, easy bruising Muscle hypotonia/atrophy, proximal joint contractures, distal joint hypermobility Short stature, muscle hypotonia, bowing of limbs

Myopathic EDS (mEDS)

AD/AR

COL12A1

Spondylodysplastic EDS (spEDS)

AR

Musculocontractural EDS (mEDS) Periodontal EDS (PEDS)

AR

B4GALT7/ B3GALT6/ SLC39A13 CHST14/DSE Congenital contractures, characteristic craniofacial features, skin fragility/hyperextensibility C1R/C1S Severe periodontitis, lack of attached gingiva, pretibial plaques ZNF469/ Thin cornea, keratoconus, keratoglobus, blue sclerae PRDM5 ? Joint hypermobility, skin hyperextensibility, smooth velvety skin

Brittle cornea syndrome (BCS) Hypermobile EDS (hEDS)

AD AR AD

a major component of the extracellular matrix. Glycine missense mutations in COL3A1 cause the most severe syndromic vEDS presentations, whereas nonsense mutations may lead to lateronset manifestation of symptoms and isolated aortic disease [26]. In rare instances, a vEDS-like phenotype is caused by specific arginine substituting mutations in the COL1A1 gene [27]. Compound heterozygous or homozygous lossof-function mutations in COL1A2 lead to cvEDS, which is characterized by early-onset insufficiency of multiple cardiac valves [28]. Borderline aortic root measurements have also been documented [28]. Another EDS form in which aortic aneurysm and arterial rupture can occur is kEDS [29]. This autosomal recessive subtype of EDS is caused by mutations in PLOD1, a gene encoding

lysyl hydroxylase 1, which is an essential enzyme for collagen maturation and elastin polymerization [30].

34.5 Arterial Tortuosity Syndrome Arterial tortuosity syndrome (ATS) is an autosomal recessive disorder characterized by elongation and abnormal twisting and turning of the arteries (tortuosity) in addition to aortic aneurysms. Skeletal abnormalities, joint hypermobility, and skin findings are also recurrent manifestations. The gene responsible for ATS is SLC2A10, encoding the facilitative glucose transporter GLUT10 [31]. Similar to MFS and LDS, mutations in SLC2A10 cause an upregulation of

528

TGFbeta signaling [31, 32]. However, the exact mechanisms underlying the pathogenesis of ATS remain elusive [33].

34.6 Cutis Laxa Autosomal dominant cutis laxa (ADCL) was historically considered a strictly cutaneous disorder without systemic involvement, in contrast to autosomal recessive cutis laxa (ARCL), which is associated with high morbidity and mortality resulting from pulmonary emphysema and aortic aneurysms. However, recent reports also have highlighted the occurrence of aortic aneurysms in ADCL attributed to gain-of-function mutations in the ELN gene, coding for the extracellular matrix protein elastin [34–36]. As regards ARCL, predominantly mutations in FBLN4, encoding the elastin and collagen cross-linking protein fibulin-4, have been linked to arterial aneurysms [37, 38].

34.7 Congenital Contractural Arachnodactyly Beals syndrome or congenital contractural arachnodactyly (CCA) is characterized by an MFSlike appearance (tall, slender habitus in which arm span exceeds height) and long, slender fingers and toes (arachnodactyly). Progressive enlargement of the sinuses of Valsalva has been reported, but there is no evidence that the aortic dilatation progresses to dissection or rupture [39]. Infants have been observed with a severe/ lethal form characterized by multiple cardiovascular and gastrointestinal anomalies in addition to the typical skeletal findings. CCA is caused by mutations in FBN2 and is inherited in an autosomal dominant manner.

A. Verstraeten and B. Loeys

by mutations in the BGN gene, coding for biglycan [40]. Patients show significant clinical overlap with both MFS and LDS, including early-onset aortic root dilatation and dissection, hypertelorism, joint hypermobility and contractures, bifid uvula, and pectus deformities. In some families, females are also affected. The BGN mutational spectrum suggests loss-of-­ function as the mechanism of action, and an increase in TGFbeta signaling has been observed in patient fibroblasts [40].

34.9 Filamin A-Related Aortopathy Over time, mutations in FLNA, another X-linked gene, have been linked to a wide spectrum of disorders. Classification of these filaminopathies depends on the nature of the underlying mutation mechanism, namely, a gain-of-function or a loss-­ of-­function mechanism. Gain-of-function mutations cause otopalatodigital disorders (OPD) [41], while loss-of-function mutations result in periventricular nodular heterotopia (PVNH) with/without connective tissue findings [42] or X-linked cardiac valvular dystrophy [43]. FLNA mutations commonly result in embryonic lethality in males. Although far more frequent in females, live-born males with FLNA-related PVNH have been reported. Variable connective tissue findings in both sexes include joint hypermobility, skin hyperelasticity, translucent skin, and cardiovascular abnormalities, such as aortic dilatation and valve insufficiencies.

34.10 Shprintzen-Goldberg Syndrome

Shprintzen-Goldberg syndrome (SGS) is characterized by craniosynostosis, distinctive craniofacial features, skeletal changes, neurologic 34.8 Meester-Loeys Syndrome abnormalities, mild-to-moderate intellectual disability, and brain anomalies. Cardiovascular Meester-Loeys syndrome (MRLS) or BGN-­ anomalies (mitral valve prolapse, mitral regurgiassociated aortic aneurysm syndrome is an tation, and aortic regurgitation) occur, but aortic X-linked syndromic form of aortopathy caused root dilatation is less commonly observed than in

34  Clinical Aspects of Heritable Connective Tissue Disorders

MFS or LDS and can be mild. An important distinguishing feature of SGS (when compared to LDS and MFS) is the near-uniform incidence of developmental delay. In most SGS cases, disease is attributed to a heterozygous de novo missense mutation in SKI [44, 45]. The SKI protein is a known repressor of TGFbeta signaling, functionally linking SGS to LDS and MFS.

References 1. McKusick V. The cardiovascular aspects of Marfan’s syndrome: a heritable disorder of connective tissue. Circulation. 1955;11:321–42. 2. Aburawi EH, O’Sullivan J.  Relation of aortic root dilatation and age in Marfan’s syndrome. Eur Heart J. 2007;28:376–9. 3. Hwa J, Richards JG, Huang H, McKay D, Pressley L, Hughes CF, et  al. The natural history of aortic dilatation in Marfan syndrome. Med J Aust. 1993;158:558–62. 4. Mimoun L, Detaint D, Hamroun D, Arnoult F, Delorme G, Gautier M, et  al. Dissection in Marfan syndrome: the importance of the descending aorta. Eur Heart J. 2011;32:443–9. 5. den Hartog AW, Franken R, Zwinderman AH, Timmermans J, Scholte AJ, van den Berg MP, et al. The risk for type B aortic dissection in Marfan syndrome. J Am Coll Cardiol. 2015;65:246–54. 6. Kinori M, Wehrli S, Kassem IS, Azar NF, Maumenee IH, Mets MB. Biometry characteristics in adults and children with Marfan syndrome: from the Marfan eye consortium of Chicago. Am J Ophthalmol. 2017;177:144–9. 7. Maumenee IH.  The eye in the Marfan syndrome. Trans Am Ophthalmol Soc. 1981;79:684–733. 8. Chandra A, Ekwalla V, Child A, Charteris D.  Prevalence of ectopia lentis and retinal detachment in Marfan syndrome. Acta Ophthalmol. 2014;92:e82–3. 9. Konradsen TR, Zetterstrom C.  A descriptive study of ocular characteristics in Marfan syndrome. Acta Ophthalmol. 2013;91:751–5. 10. Dietz HC, Cutting GR, Pyeritz RE, Maslen CL, Sakai LY, Corson GM, et al. Marfan syndrome caused by a recurrent de novo missense mutation in the fibrillin gene. Nature. 1991;352:337–9. 11. Verstraeten A, Alaerts M, Van Laer L, Loeys B. Marfan syndrome and related disorders: 25 years of gene discovery. Hum Mutat. 2016;37:524–31. 12. Loeys BL, Dietz HC, Braverman AC, Callewaert BL, De Backer J, Devereux RB, et al. The revised Ghent nosology for the Marfan syndrome. J Med Genet. 2010;47:476–85. 13. Loeys BL, Chen J, Neptune ER, Judge DP, Podowski M, Holm T, et al. A syndrome of altered c­ ardiovascular,

529

craniofacial, neurocognitive and skeletal development caused by mutations in TGFBR1 or TGFBR2. Nat Genet. 2005;37:275–81. 14. Loeys BL, Schwarze U, Holm T, Callewaert BL, Thomas GH, Pannu H, et  al. Aneurysm syndromes caused by mutations in the TGF-beta receptor. N Engl J Med. 2006;355:788–98. 15. Maccarrick G, Black JH 3rd, Bowdin S, El-Hamamsy I, Frischmeyer-Guerrerio PA, Guerrerio AL, et  al. Loeys-Dietz syndrome: a primer for diagnosis and management. Genet Med. 2014;16:576–87. 16. Micha D, Guo DC, Hilhorst-Hofstee Y, van Kooten F, Atmaja D, Overwater E, et al. SMAD2 mutations are associated with arterial aneurysms and dissections. Hum Mutat. 2015;36:1145–9. 17. van de Laar IM, Oldenburg RA, Pals G, Roos-­ Hesselink JW, de Graaf BM, Verhagen JM, et  al. Mutations in SMAD3 cause a syndromic form of aortic aneurysms and dissections with early-onset osteoarthritis. Nat Genet. 2011;43:121–6. 18. Lindsay ME, Schepers D, Bolar NA, Doyle JJ, Gallo E, Fert-Bober J, et al. Loss-of-function mutations in TGFB2 cause a syndromic presentation of thoracic aortic aneurysm. Nat Genet. 2012;44:922–7. 19. Bertoli-Avella AM, Gillis E, Morisaki H, Verhagen JM, de Graaf BM, van de Beek G, et  al. Mutations in a TGF-beta ligand, TGFB3, cause syndromic aortic aneurysms and dissections. J Am Coll Cardiol. 2015;65:1324–36. 20. Loeys BL, Dietz HC.  In: Pagon RA, Adam MP, Ardinger HH, Wallace SE, Amemiya A, LJH B, et  al., editors. Loeys-Dietz syndrome. Seattle, WA: GeneReviews(R); 2008. 21. Wenstrup RJ, Meyer RA, Lyle JS, Hoechstetter L, Rose PS, Levy HP, et  al. Prevalence of aortic root dilation in the Ehlers-Danlos syndrome. Genet Med. 2002;4:112–7. 22. Beighton P, de Paepe A, Danks D, Finidori G, Gedde-­ Dahl T, Goodman R, et al. International nosology of heritable disorders of connective tissue, berlin, 1986. Am J Med Genet. 1988;29:581–94. 23. Beighton P, De Paepe A, Steinmann B, Tsipouras P, Wenstrup RJ.  Ehlers-Danlos syndromes: revised nosology, Villefranche, 1997. Ehlers-Danlos National Foundation (USA) and Ehlers-Danlos Support Group (UK). Am J Med Genet. 1998;77:31–7. 24. Malfait F, Francomano C, Byers P, Belmont J, Berglund B, Black J, et  al. The 2017 international classification of the Ehlers-Danlos syndromes. Am J Med Genet C Semin Med Genet. 2017;175:8–26. 25. Superti-Furga A, Gugler E, Gitzelmann R, Steinmann B.  Ehlers-Danlos syndrome type IV: a multi-exon deletion in one of the two COL3A1 alleles affecting structure, stability, and processing of type III procollagen. J Biol Chem. 1988;263:6226–32. 26. Pepin MG, Schwarze U, Rice KM, Liu M, Leistritz D, Byers PH.  Survival is affected by mutation type and molecular mechanism in vascular EhlersDanlos syndrome (EDS type IV). Genet Med. 2014;16:881–8.

530 27. Malfait F, Symoens S, De Backer J, Hermanns-Le T, Sakalihasan N, Lapiere CM, et al. Three arginine to cysteine substitutions in the pro-alpha (I)-collagen chain cause Ehlers-Danlos syndrome with a propensity to arterial rupture in early adulthood. Hum Mutat. 2007;28:387–95. 28. Schwarze U, Hata R, McKusick VA, Shinkai H, Hoyme HE, Pyeritz RE, et al. Rare autosomal recessive cardiac valvular form of Ehlers-Danlos syndrome results from mutations in the COL1A2 gene that activate the nonsense-mediated RNA decay pathway. Am J Hum Genet. 2004;74:917–30. 29. Rohrbach M, Vandersteen A, Yis U, Serdaroglu G, Ataman E, Chopra M, et al. Phenotypic variability of the kyphoscoliotic type of Ehlers-Danlos syndrome (EDS VIA): clinical, molecular and biochemical delineation. Orphanet J Rare Dis. 2011;6:46. 30. Ha VT, Marshall MK, Elsas LJ, Pinnell SR, Yeowell HN. A patient with Ehlers-Danlos syndrome type VI is a compound heterozygote for mutations in the lysyl hydroxylase gene. J Clin Invest. 1994;93:1716–21. 31. Coucke PJ, Willaert A, Wessels MW, Callewaert B, Zoppi N, De Backer J, et al. Mutations in the facilitative glucose transporter GLUT10 alter angiogenesis and cause arterial tortuosity syndrome. Nat Genet. 2006;38:452–7. 32. Willaert A, Khatri S, Callewaert BL, Coucke PJ, Crosby SD, Lee JG, et  al. GLUT10 is required for the development of the cardiovascular system and the notochord and connects mitochondrial function to TGFbeta signaling. Hum Mol Genet. 2012;21:1248–59. 33. Zoppi N, Chiarelli N, Cinquina V, Ritelli M, Colombi M.  GLUT10 deficiency leads to oxidative stress and non-canonical alphavbeta3 integrin-mediated TGFbeta signalling associated with extracellular matrix disarray in arterial tortuosity syndrome skin fibroblasts. Hum Mol Genet. 2015;24:6769–87. 34. Szabo Z, Crepeau MW, Mitchell AL, Stephan MJ, Puntel RA, Yin Loke K, et al. Aortic aneurysmal disease and cutis laxa caused by defects in the elastin gene. J Med Genet. 2006;43:255–8. 35. Guemann AS, Andrieux J, Petit F, Halimi E, Bouquillon S, Manouvrier-Hanu S, et  al. ELN gene

A. Verstraeten and B. Loeys triplication responsible for familial supravalvular aortic aneurysm. Cardiol Young. 2015;25:712–7. 36. Zhang MC, He L, Giro M, Yong SL, Tiller GE, Davidson JM.  Cutis laxa arising from frameshift mutations in exon 30 of the elastin gene (ELN). J Biol Chem. 1999;274:981–6. 37. Hucthagowder V, Sausgruber N, Kim KH, Angle B, Marmorstein LY, Urban Z.  Fibulin-4: a novel gene for an autosomal recessive cutis laxa syndrome. Am J Hum Genet. 2006;78:1075–80. 38. Hebson C, Coleman K, Clabby M, Sallee D, Shankar S, Loeys B, et al. Severe aortopathy due to fibulin-4 deficiency: molecular insights, surgical strategy, and a review of the literature. Eur J Pediatr. 2014;173:671–5. 39. Gupta PA, Putnam EA, Carmical SG, Kaitila I, Steinmann B, Child A, et al. Ten novel FBN2 mutations in congenital contractural arachnodactyly: delineation of the molecular pathogenesis and clinical phenotype. Hum Mutat. 2002;19:39–48. 40. Meester JA, Vandeweyer G, Pintelon I, Lammens M, Van Hoorick L, De Belder S, et al. Loss-of-function mutations in the X-linked biglycan gene cause a severe syndromic form of thoracic aortic aneurysms and dissections. Genet Med. 2017;19:386–95. 41. Moutton S, Fergelot P, Naudion S, Cordier MP, Sole G, Guerineau E, et al. Otopalatodigital spectrum disorders: refinement of the phenotypic and mutational spectrum. J Hum Genet. 2016;61:693–9. 42. Sheen VL, Jansen A, Chen MH, Parrini E, Morgan T, Ravenscroft R, et al. Filamin A mutations cause periventricular heterotopia with Ehlers-Danlos syndrome. Neurology. 2005;64:254–62. 43. Kyndt F, Gueffet JP, Probst V, Jaafar P, Legendre A, Le Bouffant F, et al. Mutations in the gene encoding filamin A as a cause for familial cardiac valvular dystrophy. Circulation. 2007;115:40–9. 44. Doyle AJ, Doyle JJ, Bessling SL, Maragh S, Lindsay ME, Schepers D, et  al. Mutations in the TGF-beta repressor SKI cause Shprintzen-Goldberg syndrome with aortic aneurysm. Nat Genet. 2012;44:1249–54. 45. Carmignac V, Thevenon J, Ades L, Callewaert B, Julia S, Thauvin-Robinet C, et  al. In-frame mutations in exon 1 of SKI cause dominant Shprintzen-Goldberg syndrome. Am J Hum Genet. 2012;91:950–7.

Bicuspid Aortic Valve: Timing of Surgery

35

Elizabeth H. Stephens and Michael A. Borger

35.1 Introduction Bicuspid aortic valve (BAV) disease is the most common congenital cardiovascular anomaly, being present in 1–2% of the general population [1]. The majority of BAV patients will require some sort of surgical intervention during their lifetime, and it is estimated that BAV disease is responsible for more morbidity and mortality than all other congenital cardiac malformations combined [2]. Several different classification systems exist that describe the various morphologic patterns of BAV. The most commonly used system, and the one referred to throughout this chapter, is that described by Sievers et  al. (Fig. 35.1) [3]. The association between BAV disease and thoracic aortic aneurysms has been long recognized, and the management of BAV-associated aortopathy has received increased attention in recent years. Uncertainty on when to intervene in these patients stems in part from conflicting or ambiguous guidelines on aortic management [4– 6], the inconsistent use of various indexed aortic size cutoffs [7, 8], as well as continued debate in E. H. Stephens (*) Department of Cardiac, Thoracic, and Vascular Surgery, Columbia University, New York, NY, USA M. A. Borger Department for Cardiac Surgery, Leipzig Heart Center, Leipzig, Germany e-mail: [email protected]; [email protected]

the literature and at national meetings as to the optimal timing of surgical intervention [9–11]. As a consequence, a large range of clinical practices exists with several practicing surgeons making decisions for aortic intervention using cutoff values that are not recommended by consensus guidelines [12]. Making the situation even more complex, numerous recent publications have revealed that different BAV phenotypes have a different effect on patient prognosis. In light of these recently published studies, we herein attempt to describe the best current recommendations for timing of aortic intervention in patients with BAV-associated aortopathy.

35.2 H  istory of Managing BAV Patients Differently While BAV-associated aortopathy was initially assumed to be solely attributable to hemodynamic alterations within the aorta of patients with aortic stenosis (i.e., post-stenotic dilatation), numerous studies have demonstrated dilated ascending aortas in BAV patients with hemodynamically normally functioning valves [13–17]. Histopathologic studies have also demonstrated that aortas in BAV patients have altered composition that may even share some similarities to connective tissue disease patients [18–20]. Furthermore, autopsy studies demonstrate overrepresentation of BAV in specimens of patients who die of aortic dissection (i.e., 10%, compared

© Springer-Verlag GmbH Austria, part of Springer Nature 2019 O. H. Stanger et al. (eds.), Surgical Management of Aortic Pathology, https://doi.org/10.1007/978-3-7091-4874-7_35

531

532

E. H. Stephens and M. A. Borger

Fig. 35.1  Sievers’ classification of bicuspid aortic valves. Reprinted with permission from [4]

to a BAV incidence of 1–2% [21]. Other studies have demonstrated that the aortas of BAV patients dilate faster than patients with normal, tricuspid aortic valves (TAV) [22, 23]. Such observations may lead one to conclude that an aggressive approach to aortic intervention in BAV patients – with thresholds that are lower than TAV patients – is justified in order to prevent the dreaded complications of aortic dissection and rupture. In light of more recent observations (discussed below), the situation is more complex than first thought). Furthermore, BAV patients are unique in the spectrum of aortopathies because the surgeon must simultaneously consider possible indications for intervention on the aortic valve. Indeed, some studies have shown a high need for aortic repair in patients who underwent previous aortic valve replacement (AVR) surgery for valve dysfunction (i.e., 43% over 15 years for aortas over 4.5 cm at time of AVR) [24]. Subsequent guidelines therefore recommended a more aggressive approach to the aorta (i.e., replacement at a diam-

eter of 4.5  cm) in BAV patients requiring AVR surgery.

35.3 Confounding Variables Despite the abovementioned evidence demonstrating that BAV patients’ aortas may be at higher risk for aortic complications than their TAV counterparts, additional information should be taken into account prior to adopting an aggressive approach to all BAV patients. For example, the abovementioned observation that patients with normal functioning BAV valves have dilated ascending aortas does not take into account that “normal” hemodynamic valvular function is traditionally determined by echocardiography. More sensitive 4D magnetic resonance imaging studies have revealed that flow patterns and shear stresses within the ascending aorta are not normal in BAV patients, even when the BAV is hemodynamically “normal” [25, 26]. In addition, the above studies of aortic complications recorded

35  Bicuspid Aortic Valve: Timing of Surgery

in BAV patients usually have no denominator [27]. To calculate the risk of dissection at a given diameter and weigh it against the risk of morbidity/mortality of potential surgical intervention, a patient population denominator is necessary and unfortunately lacking from such studies. While BAV aortas clearly are overrepresented in aortic dissection specimens and have been demonstrated to dissect at relatively small diameters (i.e., one study demonstrated that 35% dissect at ≤5.5  cm [28]), data from IRAD show that the majority of patients (i.e., 60%) experience type A aortic dissection at aortic diameters less than 5.5 cm [29]. Apparent discrepancies in the literature regarding the natural history of the BAV aorta have further clouded the matter. While some studies have shown increasing aortic dilation in BAV compared to TAV patients [22, 23], other studies have documented no differences in aortic growth rate [30]. Further studies of BAV patients have demonstrated variability of progression of dilation [26, 31] and heterogeneity in BAV aortopathy [32, 33] that appears to partly depend on cusp morphology [26, 31, 33, 34]. Additionally, valve function (stenosis vs. insufficiency) also appears to be a strong marker of differences in BAV aortopathy, with patients presenting with aortic insufficiency displaying a more malignant course [34–37]. Therefore, the management of BAV aortopathy becomes much more complex with many aspects that require individual consideration. With regard to the heterogeneity of BAV aortopathy, aortic valve cusp morphology has been associated with distinct aortopathy patterns. The majority of patients with fusion of the left and right cusps (i.e., Sievers 1 L/R) have dilation of the ascending aorta [26], but those with root dilation, who are more commonly male, tend to present earlier and experience faster aortic dilation [26, 35]. Sievers 1  L/R patients generally have sinuses that are larger than BAV patients with fusion of the right and non-coronary cusps (Sievers 1 R/N) [33, 38–40] and are more likely to present with aortic insufficiency [34, 40]. In contrast, Sievers 1 R/N patients are more likely

533

to have dilation of the mid-ascending aorta that can extend to the aortic arch [26, 33, 34, 38, 40, 41]. Sievers 1 R/N patients are also more likely to be female [26], have mitral valve prolapse [26, 33], have aortic stenosis as opposed to insufficiency [26, 34], and to be older at time of operation [38]. The distinct BAV cusp morphologies also have distinct blood flow patterns, jet areas, and areas of increased wall shear stress that likely contribute to the observed distinct patterns of aortopathy. Numerous studies have used 4D flow patterns to examine the potential role of hemodynamics in the development and progression of BAV aortopathy, as well as complications like aortic dissection. Comprehensive discussion of the advances in this area is outside the scope of this chapter, but degree of jet eccentricity [42] and degree of restricted cusp motion [43] have correlated with progression of aortic dilation in BAV patients (Fig. 35.2). Furthermore, altered aortic wall histology has been noted at the site of jet impact [44]. More recent studies have suggested that the angle between the left ventricular outflow tract and the proximal aorta may also contribute to the hemodynamic changes and aortic histologic changes seen in BAV patients [45]. Although several different classification systems have been described regarding observed heterogeneity of BAV-associated aortopathy, we do not find such classification systems to be particularly helpful when making decisions regarding patient management. We prefer to divide patients into two simple groups based on their BAV phenotype: 1. BAV stenosis—patients with predominant bicuspid aortic stenosis and associated dilation of the tubular (i.e., supracoronary) ascending aorta 2. BAV root—patients with predominant aortic insufficiency and dilation of the proximal aortic root (Fig. 35.3) While the extent of aortopathy within these two phenotype groups can vary markedly

E. H. Stephens and M. A. Borger

534

a

b

[m/s] 1.4 0.7 0.0

BAV

c

TAV

d

[m/s] 1.4 0.7 0.0

TAV

BAV

BAV

STJ

Asc Ao

Fig. 35.2  Eccentric blood flow in BAV. Reprinted with permission [43]

between patients at the time of presentation (i.e., some patients may present with valvular abnormality only and a normal-sized aorta, while others may present with diffuse dilation of the proximal root, ascending aorta, and arch), there is an ever-growing body of evidence suggesting that the BAV root phenotype has a more malignant clinical course than BAV stenosis. We therefore suggest that the surgeon takes the BAV phenotype into consideration when making decisions regarding aortic intervention. The root phenotype is a relatively small subgroup of BAV patients (~10%), with an apparent stronger genetic component, most commonly presenting in young males [46]. While examined as a distinct subgroup by some, most studies of BAV patients have not distinguished between BAV root and BAV stenosis phenotypes in their ­analyses nor considered dilation/growth of different portions of the aorta, likely clouding the picture.

With regard to BAV phenotype-associated prognosis, evidence continues to accumulate that phenotype plays an important role. In a study by Girdauskas et al. of BAV patients having undergone isolated AVR, freedom from adverse events (including need for proximal surgery, dissection/rupture, or death) after a mean follow-up of 9.4  years was lower for patients with aortic regurgitation compared with patients with predominant aortic stenosis (78% vs. 93%) [37]. A study by Detaint et al. found significant heterogeneity in BAV aortic dilation progression with 43% not progressing over a mean 3.6  year follow-up and no relationship overall between aortic growth and valve function (regurgitation vs. stenosis) [47]. However, the subgroup of BAV patients with root dilation demonstrated faster aortic dilation if aortic regurgitation was present [47]. A number of other studies have demonstrated faster aortic dilatation in BAV with regurgitation [26, 35]

35  Bicuspid Aortic Valve: Timing of Surgery

a

535

b

Fig. 35.3 Different types of BAV pathology. (a). Example of BAV with insufficiency and aortic root aneurysm. At the top shown is the root aneurysm (patient’s head is at the base of the photo, left ventricle is at top of photo) and below is the valve. (b). BAV with stenosis and

ascending aortic aneurysm. The top image shows the ascending aortic aneurysm (patient’s head again is at the base of the photo, left ventricle is at top of photo) and below is the stenotic valve

and with the “root” phenotype [26]. A recently published meta-analysis demonstrated a tenfold increase for aortic dissection in patients with BAV insufficiency when compared to BAV stenosis [48], although earlier studies had suggested aortic stenosis as a risk factor for complications such as dissection [23]. Aortic regurgitation in BAV patients is also associated

with differences in collagen composition of the aorta compared to BAV patients with stenosis, suggestive of an underlying phenotypic difference [49, 50], as well as increased elastic fiber loss in the BAV patients with insufficiency [51], although conceivably some of these differences could be attributable to differences in flow patterns.

536

35.4 Turning Tide While previous societal guidelines had compared BAV patients to those with connective tissue disease with regard to their aortopathy (i.e., consideration should be given to aortic intervention at 4.0–5.0  cm, Class I, level of evidence C) [52], more recent studies have tempered this aggressive recommendation. In part this is due to the abovementioned acknowledgment that much of the data we previously had did not contain an accurate denominator of patients, thus inhibiting our ability to adequately assess risk [27]. The turning tide has also been aided by further studies documenting that BAV patients with modestly dilated aortas fare well after isolated AVR surgery with relatively low rates of aortic dissection and other adverse events and a lack of correlation between preoperative aortic diameter and aortic dissection in those patients in whom this complication occurred [53, 54]. Using a cutoff of 5.0 cm for intervention on the aorta, Girdauskas et  al. documented a 3% incidence of aneurysmal progression requiring proximal aortic repair over a mean follow-up of 11  years in BAV stenosis patients with moderately dilated aortas (average size prior to AVR 4.6  cm) and no incidence of aortic dissection during follow-up [54]. Their overall 15-year freedom from adverse event rate of 93% suggests that aggressive intervention on the aorta is not warranted, at least in patients presenting with BAV stenosis. Other studies have also shown that the aortas in such patients do not dilate [55, 56], nor do the unreplaced sinuses dilate [57], with a low risk of late complications in patients with an aortic diameter of 4.0–5.0 cm undergoing AVR surgery [36, 53]. Furthermore, a meta-analysis has shown that the natural history of BAV aortopathy may not be as dangerous as previously thought, especially in patients with aortic stenosis [48, 58]. More careful inspection of the study by Svensson et al. reveals that the mean diameter of the ascending aorta in BAV patients progressing to aortic dissection was 6.0, which is comparable with TAV patients [28]. As mentioned previously, the majority of aortic dissections occur at a

E. H. Stephens and M. A. Borger

d­ iameter less than 5.5 cm, and BAV patients are not overrepresented in the smaller diameter categories [29]. A recent study of BAV patients undergoing AVR without any intervening aortic repair found no significant increase in incidence of aortic dissection compared to control (i.e., tricuspid AV) patients after a median follow-up of 6.6  years [59]. This study also found a much higher rate of aortic dissection in Marfan patients post-AVR, underscoring the large difference between BAV and connective tissue patients [58]. Increasing discussion regarding the heterogeneity observed in BAV aortopathy, particularly in light of the more malignant phenotype associated with BAV insufficiency [36, 37], has led to an emphasis on an individualized approach to this heterogeneous population. Indeed some surgeons have advocated taking into account intraoperative assessment such as perceived fragility and thin aortic wall, as well as shape of aortic dilation (i.e., tubular shape prompting intervention more so than eccentric dilation) [60]. However, we do not feel that such recommendations can be currently supported by the literature.

35.5 Calculated Risk The decision to intervene on the aorta in a BAV patient should be based on calculated risk and benefit (i.e., future risk of aortic dissection or rupture versus risk of surgery). Some centers have shown the ability to perform aortic valve and ascending aorta replacement with no increase in morbidity or mortality compared to isolated AVR [61], and a recent study of NSQIP data also found no increased mortality [62], but this is not the case for most centers. Consistent with other studies demonstrating the influence of center volume on outcomes of particular cardiac operations, a study examining root and valve plus ascending replacements found that operative mortality was 58% less in the high-volume centers compared to the low-volume centers [63]. The risk of an isolated ascending aorta replacement similarly has a finite surgical risk (3.4% risk of in-hospital or 30-day mortality and 3.2%

35  Bicuspid Aortic Valve: Timing of Surgery

537

stroke according to a recent analysis of STS data) [64], which increases when aortic arch surgery is included (5.1% risk of in-hospital or 30-day mortality and 5.3% risk of stroke) [64].

35.7 Current Recommendation: Aortic Intervention in BAV Patients with Indications for Aortic Valve Surgery

35.6 Current Recommendation: Aortic Intervention in BAV Patients Regardless of Aortic Valve Status

For those BAV patients who meet indications for aortic valve surgery, the recommended cutoff for concomitant ascending aortic replacement is 4.5  cm (Class IIa, level of evidence C in both AHA/ACC 2014 guidelines [65] and ESC 2014 guidelines) [4].

Based on the available data, societies have recommended the following cutoffs for intervention on BAV patients with aortic dilatation without indications for intervention on the valve: 5.0 cm if the patient has risk factors including aortic coarctation, hypertension, a family history of dissection, or rapid growth (>3 mm/year). This is a Class IIa, level of evidence C recommendation in the AHA/ACC 2014 guidelines [65] and a Class I, level C recommendation in the ESC 2014 guidelines [4]. For those without the above risk factors, the recommended cutoff for intervention is 5.5  cm (Class I, level B in AHA/ACC 2014 guidelines [65] and Class I, level C in ESC 2014 guidelines) [4]. Lower thresholds can be considered in patients with small BSA or stature, rapid progression of aortic dilation, BAV aortic regurgitation (associated with a more malignant phenotype), females planning pregnancy, or in patients with a strong preference for early surgery. Valve-sparing aortic root replacement (i.e., David operation) can be considered in BAV patients with dilation of the aortic root and normal valve function or aortic regurgitation with good cusp mobility. Studies have shown no increased risk of reoperation or recurrent aortic regurgitation in carefully selected BAV patients compared to tricuspid AV patients [66–68]. However, these patients may have a long-term risk of developing aortic stenosis [69]. Because valve-sparing aortic root replacement is more technically challenging in BAV patients, such procedures should be performed in referral centers with a large experience with these procedures.

References 1. Ward C.  Clinical significance of the bicuspid aortic valve. Heart. 2000;83:81–5. 2. Borger M, David TE.  Management of the valve and ascending aorta in adults with bicuspid aortic valve disease. Semin Thorac Cardiovasc Surg. 2005;17:143–7. 3. Sievers HH, Schmidtke C.  A classification system for the bicuspid aortic valve from 304 surgical specimens. J Thorac Cardiovasc Surg. 2007;133:1226–33. 4. Erbel R, Aboyans V, Boileau C, Bossone E, Bartolomeo RD, Eggebrecht H, et  al. Guidelines ESCCfP 2014 ESC guidelines on the diagnosis and treatment of aortic diseases: document ­ covering acute and chronic aortic diseases of the thoracic and abdominal aorta of the adult. The task force for the diagnosis and treatment of aortic diseases of the European Society of Cardiology (ESC). Eur Heart J. 2014;35:2873–926. https://doi.org/10.1093/eurheartj/ ehu281. 5. Nishimura RA, Otto CM, Bonow RO, Carabello BA, Erwin JP 3rd, Guyton RA, et al. Members AATF. 2014 AHA/ACC guideline for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation. 2014;129:e521–643. 6. Svensson LG, Adams DH, Bonow RO, Kouchoukos NT, Miller DC, O’Gara PT, et  al. Aortic valve and ascending aorta guidelines for management and quality measures. Ann Thorac Surg. 2013;95:S1–66. 7. McDonald ML, Smedira NG, Blackstone EH, Grimm RA, Lytle BW, Cosgrove DM.  Reduced survival in women after valve surgery for aortic regurgitation: effect of aortic enlargement and late aortic rupture. J Thorac Cardiovasc Surg. 2000;119:1205–12. 8. Svensson LG, Blackstone EH, Cosgrove DM. Surgical options in young adults with aortic valve disease. Curr Probl Cardiol. 2003;28:417–80. 9. Kallenbach K, Sundt TM, Marwick TH.  Aortic surgery for ascending aortic aneurysms under 5.0  cm

538 in diameter in the presence of bicuspid aortic valve. JACC Cardiovasc Imaging. 2013;6:1321–6. 10. Elefteriades JA. Editorial comment: should aortas in patients with bicuspid aortic valve really be resected at an earlier stage than those in patients with tricuspid valve? Cardiol Clin. 2010;28:315–6. 11. Wald O, Korach A, Shapira OM.  Should aortas in patients with bicuspid aortic valve really be resected at an earlier stage than tricuspid? PRO. Cardiol Clin. 2010;28:289–98. 12. Verma S, Yanagawa B, Kalra S, Ruel M, Peterson MD, Yamashita MH, et al. Knowledge, attitudes, and practice patterns in surgical management of bicuspid aortopathy: a survey of 100 cardiac surgeons. J Thorac Cardiovasc Surg. 2013;146:1033–40. 13. Hahn RT, Roman MJ, Mogtader AH, Devereux RB.  Association of aortic dilation with regurgitant, stenotic and functionally normal bicuspid aortic valves. J Am Coll Cardiol. 1992;19:283–8. 14. Basso C, Boschello M, Perrone C, Mecenero A, Cera A, Bicego D, et  al. An echocardiographic survey of primary school children for bicuspid aortic valve. Am J Cardiol. 2004;93:661–3. 15. Nistri S, Sorbo MD, Marin M, Palisi M, Scognamiglio R, Thiene G. Aortic root dilatation in young men with normally functioning bicuspid aortic valves. Heart. 1999;82:19–22. 16. Pachulski RT, Weinberg AL, Chan KL. Aortic aneurysm in patients with functionally normal or minimally stenotic bicuspid aortic valve. Am J Cardiol. 1991;67:781–2. 17. Cecconi M, Manfrin M, Moraca A, Zanoli R, Colonna PL, Bettuzzi MG, et al. Aortic dimensions in patients with bicuspid aortic valve without significant valve dysfunction. Am J Cardiol. 2005;95:292–4. 18. Boyum J, Fellinger EK, Schmoker JD, Trombley L, McPartland K, Ittleman FP, Howard AB.  Matrix metalloproteinase activity in thoracic aortic aneurysms associated with bicuspid and tricuspid aortic valves. J Thorac Cardiovasc Surg. 2004;127:686–91. 19. Ikonomidis JS, Jones JA, Barbour JR, Stroud RE, Clark LL, Kaplan BS.  Expression of matrix metalloproteinases and endogenous inhibitors within ascending aortic aneurysms of patients with bicuspid or tricuspid aortic valves. J Thorac Cardiovasc Surg. 2007;133:1028–36. 20. Fedak PW, de Sa MP, Verma S, Nili N, Kazemian P, Butany J, et al. Vascular matrix remodeling in patients with bicuspid aortic valve malformations: implications for aortic dilatation. J Thorac Cardiovasc Surg. 2003;126:797–806. 21. Edwards WD, Leaf DS, Edwards JE. Dissecting aortic aneurysm associated with congenital bicuspid aortic valve. Circulation. 1978;57:1022–5. 22. Etz CD, Zoli S, Brenner R, Roder F, Bischoff M, Bodian CA, et  al. When to operate on the bicuspid valve patient with a modestly dilated ascending aorta. Ann Thorac Surg. 2010;90:1884–90. 23. Davies RR, Kaple RK, Mandapati D, Gallo A, Botta DM Jr, Elefteriades JA, Coady MA.  Natural history

E. H. Stephens and M. A. Borger of ascending aortic aneurysms in the setting of an unreplaced bicuspid aortic valve. Ann Thorac Surg. 2007;83:1338–44. 24. Borger MA, Preston M, Ivanov J, Fedak PW, Davierwala P, Armstrong S, David TE.  Should the ascending aorta be replaced more frequently in patients with bicuspid aortic valve disease? J Thorac Cardiovasc Surg. 2004;128:677–83. 25. Hope MD, Hope TA, Meadows AK, Ordovas KG, Urbania TH, Alley MT, Higgins CB.  Bicuspid aortic valve: four-dimensional MR evaluation of ascending aortic systolic flow patterns. Radiology. 2010;255:53–61. 26. Della Corte A, Bancone C, Buonocore M, Dialetto G, Covino FE, Manduca S, et  al. Pattern of ascending aortic dimensions predicts the growth rate of the aorta in patients with bicuspid aortic valve. JACC Cardiovasc Imaging. 2013;6:1301–10. 27. Sundt TM.  Replacement of the ascending aorta in bicuspid aortic valve disease: where do we draw the line? J Thorac Cardiovasc Surg. 2010;140:S41–4. 28. Svensson LG, Kim KH, Lytle BW, Cosgrove DM.  Relationship of aortic cross-sectional area to height ratio and the risk of aortic dissection in patients with bicuspid aortic valves. J Thorac Cardiovasc Surg. 2003;126:892–3. 29. Pape LA, Tsai TT, Isselbacher EM, Oh JK, O’Gara PT, Evangelista A, et  al. Aortic diameter >or = 5.5  cm is not a good predictor of type a aortic dissection: observations from the international registry of acute aortic dissection (IRAD). Circulation. 2007;116:1120–7. 30. La Canna G, Ficarra E, Tsagalau E, Nardi M, Morandini A, Chieffo A, Maisano F, Alfieri O.  Progression rate of ascending aortic dilation in patients with normally functioning bicuspid and tricuspid aortic valves. Am J Cardiol. 2006;98:249–53. 31. Thanassoulis G, Yip JW, Filion K, Jamorski M, Webb G, Siu SC, Therrien J. Retrospective study to identify predictors of the presence and rapid progression of aortic dilatation in patients with bicuspid aortic valves. Nat Clin Pract Cardiovasc Med. 2008;5:821–8. 32. Fazel SS, Mallidi HR, Lee RS, Sheehan MP, Liang D, Fleischman D, et  al. The aortopathy of bicuspid aortic valve disease has distinctive patterns and usually involves the transverse aortic arch. J Thorac Cardiovasc Surg. 2008;135:901–7. 33. Schaefer B, Lewin M, Stout K, Gill E, Prueitt A, Byers P, Otto C.  The bicuspid aortic valve: an integrated phenotypic classification of leaflet morphology and aortic root shape. Heart. 2008;94:1634–8. 34. Kang JW, Song HG, Yang DH, Baek S, Kim DH, Song JM, et al. Association between bicuspid aortic valve phenotype and patterns of valvular dysfunction and bicuspid aortopathy: comprehensive evaluation using MDCT and echocardiography. JACC Cardiovasc Imaging. 2013;6:150–61. 35. Della Corte A, Bancone C, Quarto C, Dialetto G, Covino FE, Scardone M, et al. Predictors of ascending aortic dilatation with bicuspid aortic valve: a wide

35  Bicuspid Aortic Valve: Timing of Surgery spectrum of disease expression. Eur J Cardiothorac Surg. 2007;31:397–404. 36. Girdauskas E, Disha K, Raisin HH, Secknus MA, Borger MA, Kuntze T. Risk of late aortic events after an isolated aortic valve replacement for bicuspid aortic valve stenosis with concomitant ascending aortic dilation. Eur J Cardiothorac Surg. 2012;42:83237. 37. Girdauskas E, Disha K, Secknus M, Borger M, Kuntze T.  Increased risk of late aortic events after isolated aortic valve replacement in patients with bicuspid aortic valve insufficiency versus stenosis. J Cardiovasc Surg. 2013;54:653–9. 38. Russo CF, Cannata A, Lanfranconi M, Vitali E, Garatti A, Bonacina E.  Is aortic wall degeneration related to bicuspid aortic valve anatomy in patients with valvular disease? J Thorac Cardiovasc Surg. 2008;136:937–42. 39. Schaefer BM, Lewin MB, Stout KK, Byers PH, Otto CM. Usefulness of bicuspid aortic valve phenotype to predict elastic properties of the ascending aorta. Am J Cardiol. 2007;99:686–90. 40. Della Corte A, Bancone C, Dialetto G, Covino FE, Manduca S, D’Oria V, et  al. Towards an individualized approach to bicuspid aortopathy: different valve types have unique determinants of aortic dilatation. Eur J Cardiothorac Surg. 2014;45:e118–24. 41. American College of Cardiology/American Heart Association Task Force on Practice Guidelines, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography, Interventions, Society of Thoracic Surgeons, Bonow RO, Carabello BA, Kanu C, de Leon AC Jr, Faxon DP, Freed MD, et  al. ACC/AHA 2006 guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (writing committee to revise the 1998 Guidelines for the Management of Patients with Valvular Heart Disease): developed in collaboration with the Society of Cardiovascular Anesthesiologists: endorsed by the Society for Cardiovascular Angiography and Interventions and the Society of Thoracic Surgeons. Circulation. 2006;114:e84–231. 42. Hope MD, Wrenn J, Sigovan M, Foster E, Tseng EE, Saloner D.  Imaging biomarkers of aortic disease: increased growth rates with eccentric systolic flow. J Am Coll Cardiol. 2012;60:356–7. 43. Della Corte A, Bancone C, Conti CA, Votta E, Redaelli A, Del Viscovo L, Cotrufo M. Restricted cusp motion in right-left type of bicuspid aortic valves: a new risk marker for aortopathy. J Thorac Cardiovasc Surg. 2012;144:360–9. 44. Girdauskas E, Rouman M, Disha K, Scholle T, Fey B, Theis B, Petersen I, Borger MA, Kuntze T. Correlation between systolic transvalvular flow and proximal aortic wall changes in bicuspid aortic valve stenosis. Eur J Cardiothorac Surg. 2014;46:234–9. 45. Girdauskas E, Rouman M, Disha K, Espinoza A, Dubslaff G, Fey B, et  al. Predicting aortopathy in patients with bicuspid aortic valve stenosis: focus

539 on the aortic root functional parameters. Eur J Cardiothorac Surg. 2016;49:635–43. 46. Girdauskas E, Borger MA.  Bicuspid aortic valve and associated aortopathy: an update. Semin Thorac Cardiovasc Surg. 2013;25:310–6. 47. Detaint D, Michelena HI, Nkomo VT, Vahanian A, Jondeau G, Sarano ME. Aortic dilatation patterns and rates in adults with bicuspid aortic valves: a comparative study with Marfan syndrome and degenerative aortopathy. Heart. 2014;100:126–34. 48. Girdauskas E, Rouman M, Disha K, Espinoza K, Misfeld M, Borger MA, Kuntze T.  Meta-analysis of aortic dissection after previous aortic valve replacement for bicuspid aortic valve disease. J Am Coll Cardiol. 2015;66:1409–11. 49. Cotrufo M, Della Corte A. The association of bicuspid aortic valve disease with asymmetric dilatation of the tubular ascending aorta: identification of a definite syndrome. J Cardiovasc Med. 2009;10:291–7. 50. Cotrufo M, Della Corte A, De Santo LS, Quarto C, De Feo M, Romano G, et al. Different patterns of extracellular matrix protein expression in the convexity and the concavity of the dilated aorta with bicuspid aortic valve: preliminary results. J Thorac Cardiovasc Surg. 2012;130:504–11. 51. Girdauskas E, Rouman M, Borger MA, Kuntze T.  Comparison of aortic media changes in patients with bicuspid aortic valve stenosis versus bicuspid valve insufficiency and proximal aortic aneurysm. Interact Cardiovasc Thorac Surg. 2013;17:931–6. 52. Hiratzka LF, Bakris GL, Beckman JA, Bersin RM, Carr VF, Casey DE Jr, American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, American Association for Thoracic Surgeons, American College of Radiology, American Stroke Association, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography, Interventions, Society of Interventional Radiology, Society of Thoracic Surgeons, Society for Vascular Medicine, et al. 2010 ACCF/AHA/AATS/ACR/ASA/SCA/SCAI/SIR/STS/ SVM guidelines for the diagnosis and management of patients with thoracic aortic disease: a report of the American College of Cardiology Foundation/ American Heart Association Task Force on Practice Guidelines, American Association for Thoracic Surgery, American College of Radiology, American Stroke Association, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society of Interventional Radiology, Society of Thoracic Surgeons, and Society for Vascular Medicine. Circulation. 2010;121:e266–369. 53. Girdauskas E, Disha K, Borger MA, Kuntze T. Long-­ term prognosis of ascending aortic aneurysm after aortic valve replacement for bicuspid versus tricuspid aortic valve stenosis. J Thorac Cardiovasc Surg. 2014;147:276–82. 54. McKellar SH, Michelena HI, Li Z, Schaff HV, Sundt TM. Long-term risk of aortic events following ­aortic

540 valve replacement in patients with bicuspid aortic valves. Am J Cardiol. 2010;106:1626–33. 55. Abdulkareem N, Soppa G, Jones S, Valencia O, Smelt J, Jahangiri M. Dilatation of the remaining aorta after aortic valve or aortic root replacement in patients with bicuspid aortic valve: a 5-year follow-up. Ann Thorac Surg. 2013;96:43–9. 56. Lee SH, Kim JB, Kim DH, Jung SH, Choo SJ, Chung CH, Lee JW. Management of dilated ascending aorta during aortic valve replacement: valve replacement alone versus aorta wrapping versus aorta replacement. J Thorac Cardiovasc Surg. 2013;146:802–9. 57. Park CB, Greason KL, Suri RM, Michelena HI, Schaff HV, Sundt TM.  Fate of nonreplaced sinuses of Valsalva in bicuspid aortic valve disease. J Thorac Cardiovasc Surg. 2011;142:278–84. 58. Hardikar AA, Marwick TH.  Surgical thresholds for bicuspid aortic valve associated aortopathy. JACC Cardiovasc Imaging. 2013;6:1311–20. 59. Itagaki S, Chikwe JP, Chiang YP, Egorova NN, Adams DH. Long-term risk for aortic complications after aortic valve replacement in patients with bicuspid aortic valve versus Marfan syndrome. J Am Coll Cardiol. 2015;65:2363–9. 60. Sievers HH, Stierle U, Mohamed SA, Hanke T, Richardt D, Schmidtke C, Charitos EI. Toward individualized management of the ascending aorta in bicuspid aortic valve surgery: the role of valve phenotype in 1362 patients. J Thorac Cardiovasc Surg. 2014;148:2072–80. 61. Rinewalt D, McCarthy PM, Malaisrie SC, Fedak PW, Andrei AC, Puthumana JJ, Bonow RO.  Effect of aortic aneurysm replacement on outcomes after bicuspid aortic valve surgery: validation of contemporary guidelines. J Thorac Cardiovasc Surg. 2014;148:2060–9. 62. Opotowsky AR, Perlstein T, Landzberg MJ, Colan SD, O’Gara PT, Body SC, Ryan LF, Aranki S, Singh MN.  A shifting approach to management of the thoracic aorta in bicuspid aortic valve. J Thorac Cardiovasc Surg. 2013;146:339–46.

E. H. Stephens and M. A. Borger 63. Hughes GC, Zhao Y, Rankin JS, Scarborough JE, O’Brien S, Bavaria JE, et  al. Effects of institutional volumes on operative outcomes for aortic root replacement in North America. J Thorac Cardiovasc Surg. 2013;145:166–70. 64. Williams JB, Peterson ED, Zhao Y, O’Brien SM, Andersen ND, Miller DC, Chen EP, Hughes GC.  Contemporary results for proximal aortic replacement in North America. J Am Coll Cardiol. 2012;60:1156–62. 65. Nishimura RA, Otto CM, Bonow RO, Carabello BA, Erwin JP 3rd, Guyton RA, American College of C, American College of Cardiology/American Heart A, American Heart A, et al. 2014 AHA/ACC guideline for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association task force on practice guidelines. J Thorac Cardiovasc Surg. 2014;148:e1–e132. 66. Kvitting JP, Kari FA, Fischbein MP, Liang DH, Beraud AS, Stephens EH, Mitchell RS, Miller DC.  David valve-sparing aortic root replacement: equivalent mid-term outcome for different valve types with or without connective tissue disorder. J Thorac Cardiovasc Surg. 2012;145:117–26. 67. Kunihara T, Aicher D, Rodionycheva S, Groesdonk HV, Langer F, Sata F, Schafers HJ. Preoperative aortic root geometry and postoperative cusp configuration primarily determine long-term outcome after valve-­ preserving aortic root repair. J Thorac Cardiovasc Surg. 2012;143:1389–95. 68. David TE, Feindel CM, David CM, Manlhiot C.  A quarter of a century of experience with aortic valve-sparing operations. J Thorac Cardiovasc Surg. 2014;148:872–9. 69. Miller DC.  Rationale and results of the Stanford modification of the David V reimplantation technique for valve-sparing aortic root replacement. J Thorac Cardiovasc Surg. 2015;149:112–4.

Biscuspid Aortic Valve Repair

36

Nhue Do, Joel Price, and Lars G. Svensson

36.1 Introduction Bicuspid aortic valve (BAV) disease is the most common congenital defect with a prevalence of 1–2% in the population [1–3]. BAV follows an autosomal dominant inheritance pattern with low penetrance, afflicting 10–36% of first-degree relatives of patients with BAV [4–6]. Patients with BAV are at risk to develop a number of conditions including aortic complications, aortic valve stenosis, and aortic valve insufficiency [7–10]. Patients with severe, symptomatic aortic insufficiency (AI) experienced increased risk of heart failure and death [11]. The standard treatment for aortic valve disease today continues to be aortic valve replacement (AVR). AVR results in improved symptoms and survival and can be performed with excellent contemporary results [12– 14]. However, AVR does necessitate the placement of a prosthesis which is subject to valve-related events over the long term [15]. N. Do Cardiovascular Surgery, Johns Hopkins All Children’s Heart Institute, St Petersburg, FL, USA e-mail: [email protected] J. Price (*) Cardiac, Aortic and Endovascular Surgery, University of British Columbia, Vancouver, BC, Canada e-mail: [email protected] L. G. Svensson Cleveland Clinic, Cleveland, OH, USA e-mail: [email protected]

Aortic bioprostheses experience high rates of structural valve deterioration, particularly in younger patients. For this reason, mechanical prostheses are often selected for patients of younger ages. Patients with bicuspid aortic valves and AI are frequently young. The necessity for lifelong anticoagulation makes a mechanical prosthesis an imperfect solution [15]. In response to these concerns, several centers have developed techniques to repair regurgitant aortic valves. The development of effective and durable techniques, for leaflet repair and aortic annulus reconstruction, has enabled the repair of the regurgitant aortic valve in selected patients. The primary goal of AVR surgery is to restore a durable surface of coaptation to the regurgitant valve. Successful valve repair is predicated on a thorough understanding of particularly the mechanism of dysfunction and annular size. A full evaluation involves working through the CLASS Schema (Commissures, Leaflets, Annulus, Sinuses, and ST junction) [15, 16]. Once the mechanism has been identified, the appropriate technique can be selected to address the pathology.

36.2 Anatomy and Classification The functional aortic annulus (FAA) consists of the sinotubular junction (STJ), the aortoventricular junction (AVJ), and the anatomic

© Springer-Verlag GmbH Austria, part of Springer Nature 2019 O. H. Stanger et al. (eds.), Surgical Management of Aortic Pathology, https://doi.org/10.1007/978-3-7091-4874-7_36

541

N. Do et al.

542

c­ rown-­shaped insertion of the aortic valve leaflets [17]. A normal, trileaflet, aortic valve (AV) has three leaflets of nearly equal size, each occupying approximately 120° of annular circumference. The leaflets insert in a crown-like formation in the FAA, with the nadir at the level of the AVJ and the apex at the level of the STJ. Leaflet coaptation occurs at the center of the AV orifice, with a coaptation height that is approximately at the mid-level between the AVJ and the STJ [17]. The congenitally bicuspid aortic valve consists of only two leaflets. Practically, the BAV exists as a spectrum of phenotypes ranging from a valve consisting of two equally sized leaflets and commissures and with two opposing sinuses to two underdeveloped cusps joined by one well-­ developed raphe and a fully developed single leaflet [4, 18]. Sometimes, there can be two raphes yielding the so-called unicuspid valve. The raphe usually occurs between two underdeveloped cusps, each occupying less than 120° of the annular circumference. A systematic classification of BAV phenotypes was developed by Sievers et al. [18] from pathological specimens from operative patients and is based on the number of raphes, the spatial position of leaflets and raphes, and the functional status of the entire valve. Patients with a BAV frequently have associated abnormalities of the aorta including dilation, aneurysm, and dissection. Researches have described histologic abnormalities in the wall of the proximal aorta of patients with BAV.  This condition has been dubbed the bicuspid aortopathy and resembles some connective tissue disorders. Suggested abnormalities include a deficient fibrillin-1 content leading to cystic medial necrosis and disruption of the extracellular matrix [2, 7, 10, 19–24]. Other authors have suggested that hemodynamic abnormalities related to the BAV contribute to aneurysm formation in these patients [25]. The risk of aortic rupture and dissection may be elevated in BAV patients. Reports of the increased risk of aortic dissection in patients with a BAV aortopathy have ranged from a 5-fold to as high as a 12-fold increase [7, 22, 23]. As such, patients with bicuspid valve disease and aneurysm of the aortic root and ascending

aorta may need to be more aggressively managed, with lower thresholds for surgical intervention, compared to the patient without a BAV.

36.3 Mechanisms of Aortic Insufficiency Similar to and based on the Carpentier classification system for mitral valve regurgitation, El Khoury et  al. developed a functional classification for aortic valve insufficiency (AI) based on leaflet motion [26, 27]. According to this classification, AI is divided into three types. Type I is associated with normal leaflet motion and dilation of one or more components of the FAA. Type II AI is related to excessive leaflet motion which is effectively prolapse, and Type III AI is characterized by restricted leaflet motion. Aortic insufficiency in BAV disease can be caused by any of the above mechanisms and frequently a combination of Type I with either Type II or III [28]. AI in Sievers type 0 bicuspid valve is frequently due to prolapse of one or both leaflets (Type II). The mechanism of AI in the Sievers type 1 bicuspid valve can be due to leaflet prolapse (Type II), most commonly the conjoint cusp, or a restrictive raphe (Type III), which retracts the free margin of the conjoint cusp resulting in a triangular coaptation defect. As is the case for mitral valve repair, aortic valve repair requires both leaflet repair and may require annuloplasty to yield durable results [29, 30]. A number of FAA annuloplasty techniques have been described for the aortic valve. The selection of the appropriate technique depends on the type and extend of FAA dilation and the quality of the aortic root tissue. These considerations will be discussed below.

36.4 Patient Selection While patients with BAV frequently develop aortic stenosis [2, 3, 18], these patients are generally not candidates for aortic valve repair to inadequate quality and quantity of leaflet tissue. The indications for valve repair for AI are the same as

36  Biscuspid Aortic Valve Repair

for aortic valve replacement, according to published guidelines [11]. These include symptomatic, severe AI or asymptomatic severe AI with LV dysfunction or dilation. The decision to repair a bicuspid aortic valve must include consideration of the risk-benefit profile for each individual patient, surgeon experience and outcomes, and an assessment of the feasibility of repair based on evaluation of leaflet quality and quantity. Aortic repair of the BAV is favored in the young patient in order to avoid the risks associated with long-term anticoagulation for a mechanical valve prosthesis and the limited durability of biological prostheses in this patient cohort [15, 31]. In general, patients with aortic regurgitation, in the absence of excessive leaflet calcification or fibrosis, are candidates for aortic valve repair. However, in the contemporary era of durable aortic prostheses, the feasibility of a durable repair must be carefully evaluated. An intraoperative assessment of leaflet tissue quality and quantity is essential. The presence of significant calcification or fibrosis or a large, restrictive raphe may require extensive excision and patch techniques. The presence of multiple large fenestrations or very thin, fragile leaflets may indicate poor tissue quality. Repair of a diminutive leaflet may result in restrictive leaflet motion postoperatively. In any of these scenarios, a durable repair may not be feasible, and a replacement should be seriously considered.

36.5 Techniques for BAV Repair 36.5.1 Incision, Setup, and Exposure The sternum is opened through a standard median sternotomy or a minimal invasive J incision [32]. Cannulation of the distal ascending aorta and two-stage venous cannulation of right atrium are performed in a standard fashion. A transverse aortotomy is performed 1  cm above the sinotubular junction. The incision is made starting above the noncoronary sinus and extended circumferentially so that only the posterior 1–2 cm of aortic wall, directly above the

543

left coronary ostium, is left intact. In the presence of ascending aortic aneurysm, the aorta is incised higher, in the middle of the portion to be excised. This is done in order to avoid inadvertently cutting the commissures or coronary ostia which can be elevated in these patients. Once the valve is visualized, the aorta is incised vertically caudad, stopping 1 cm above the commissures, and a transverse incision is performed as described above. A traction suture is placed in the ascending aorta to retract it upward, and a traction suture is placed at the apex of each of the two commissures. A third can be placed above the raphe to assist in visualization. Axial traction applied to the commissural traction sutures allows demonstration of the physiological aortic valve closure position and the height of coaptation. At this point, the assessment of tissue quality and quantity must be performed; position and degree of restriction of the raphe are noted. The presence of fenestrations, calcifications, or fibrosis should be noted. Identification of leaflet prolapse or restriction is facilitated by examining leaflet motion, tissue quality, and the position of the free margin of each leaflet. A prolapsing leaflet can be identified by the position of the free margin below the reference coaptation point. The reference point corresponds to the height of the free margin of the non-prolapsing leaflets and the level to which the prolapsing leaflet must be elevated. This will usually occur in the midheight of the sinuses of Valsalva. In the rare situation that there is no non-­prolapsing reference leaflet, the mid-sinus height may be used as an external reference.

36.5.2 Leaflet Repair Techniques For leaflet prolapse, valve assessment will reveal the free margin of one or more leaflets below the level of coaptation with varying levels of tissue quality. Cusp prolapse is usually associated with excess length of the free margin, which can be corrected using central free margin plication, triangular resection, free margin resuspension, or figure-of-8 suspensory sutures.

N. Do et al.

544

36.5.3 Trusler and Figure-of-8 Suspensory Suture The shortening of one leaflet at a commissure by a concertina stitch of the prolapsing leaflet was describe by Trusler et al. [33]. The “Trusler” suspensory suture is a modification of the original stitch by Frater et  al. [34]. The figure-of-8 suspensory suture described by Svensson and colleagues further involves the placement of a 5-0 PTFE suture through the free margin of the leaflet at the commissure with plication of a portion of the elongated cusps of the prolapsed leaflet to the aortic wall and placing the sutures at about 5  mm above the commissures and tieing the sutures on the outside of the aorta [35] to help resuspend the prolapse cusp to a normal level where an adequate margin of coaptation can occur (Fig. 36.1a).

36.5.4 Free Margin Central Plication Free margin plication is a simple, fast, and versatile technique. The conjoint leaflet is frequently the prolapsing leaflet, and as such the plication is often performed in proximity to the raphe. The

Fig. 36.1  Illustration of bicuspid aortic valve repair with (A) figure-of-8 suspensory suture and (B) Cabrol subcommissural annuloplasty suture. Reprinted with permission [59]

prolapsing leaflet is pulled parallel to the reference point and to one side with forceps. A 7-0 or 6-0 polypropylene or polyester suture is passed through the prolapsing leaflet, from the aortic to ventricular side, at the point at which it meets the central reference point. The prolapsing leaflet is then pulled in the opposite direction, and the same suture is passed from the ventricular to the aortic side of the cusp where it meets the reference point. The suture can then be tied which will create a fold of excess tissue on the aortic side of the leaflet. If this fold is not excessively large, the plication is extended onto the body of the aortic leaflet by running a suture from the base of the fold to the free margin (Figs. 36.2 and 36.3).

36.5.5 Triangular Resection A triangular resection of significant excess leaflet tissue begins with plication sutures in identical fashion to free the free margin technique described previously. The plication sutures are used to mark the margin of resection at the free leaflet edge, and the excess tissue is excised with scissors toward the base of the fold created by tension on the plication sutures. Excessive resec-

B B

A

A

36  Biscuspid Aortic Valve Repair

545

Fig. 36.2  Free margin central plication with extension onto the body of the leaflet. Reprinted with permission [63]

36.5.6 Free Margin Resuspension

Fig. 36.3  Central plication of a nonrestrictive raphe, with extension onto the body of the leaflet

tion of leaflet tissue can result in leaflet restriction and excessive tension resulting in a failed repair; thus the general recommendation is a conservative resection with an approximately 1 mm ridge of tissue on either side to facilitate closure. The defect is closed primarily from the base to free margin. A locking suture can be performed to avoid purse stringing of the leaflet. Long-term data, however, have shown a higher failure rate for resection versus plication.

There are a few situations in which resuspension is particularly useful. This technique is indicated in the setting of a fragile free margin with multiple fenestrations and when there is a paucity of tissue such that a plication would restrict motion. This technique can also be used to reinforce the free margin after pericardial patch leaflet augmentation. A 7-0 or 6-0 polytetrafluoroethylene (PTFE) suture is passed twice through the aortic wall, in a locking fashion, at the apex of one commissure of the prolapsing leaflet. The suture is then passed in a running fashion over the length of the free margin. Two locked stitches are performed at the apex of the opposite commissure. The identical procedure is repeated with a second 7-0 PTFE.  The length of the free margin is reduced by grasping the leaflet at the middle of the free margin with forceps and applying gentle traction on each arm of the PTFE sutures. Both sides of the sutures are tightened to plicate the free margin until it reaches the same height as the reference point. The two suture ends are subsequently tied at each commissure (Fig. 36.4).

N. Do et al.

546

because of clot formation on the patches, and should be avoided as much as possible.

36.5.8 Annuloplasty Techniques

Fig. 36.4  Technique for free margin resuspension with polytetrafluoroethylene suture passed along the length of the free margin. Reprinted with permission [63]

36.5.7 Repair Techniques for Restrictive Raphe If the raphe is fibrotic and not overly restrictive, simple shaving of the raphe, to reduce bulk and restriction, will yield a supple and mobile leaflet that can be repaired with the above techniques for prolapse. Frequently, a suspensory ligament will be present, tethering the raphe to the aortic wall. Cutting of this fibrous band of tissue will significantly increase leaflet mobility and allow for a prolapse repair. When the raphe is found to be calcified and restrictive, a triangular resection can be performed. When resecting the raphe, an attempt is made to remove all diseased tissue while preserving as much normal tissue as possible. When the resulting defect is small, a primary repair can be performed. This is accomplished with a running suture from the base to the free margin. If the defect is large, a triangular pericardial patch can be used to close the defect. Patch material can be autologous or bovine pericardium [36–38]; the use of core matrix has also been described [39]. The patch material is tailored to match the defect and attached to the leaflet edges with a running locked 6-0 or 5-0 polypropylene suture. Both these techniques “bicuspidize” the valve. Experienced centers however have found that the use of a patch repair is associated with a greater risk of both stroke and repair failure, partly

As previously mentioned, comprehensive BAV repair includes not only intervention on the leaflets but stabilization annulus to render the repair durable. Three primary techniques have been described for aortic annuloplasty during bicuspid aortic valve repair: (1) aortic valve preservation using the remodeling technique [15, 16, 28, 30, 35, 40–52] including an inclusion technique [53], (2) aortic valve-sparing root replacement (AVSRR) using the reimplantation technique [15, 16, 28, 42–45], and (3) subcommissural annuloplasty with Cabrol sutures [15, 16, 35, 43, 45] sometimes in combination with downsizing replacement of the ascending aorta.

36.5.9 Subcommissural Annuloplasty (Cabrol Sutures) The purpose of a subcommissural annuloplasty is to reduce the diameter of the AVJ, increase the surface of leaflet coaptation, narrow the intercommissural angle, and reduce tension on the leaflet repair. This is accomplished by reduction of the width of the interleaflet triangle. A pledgeted 2-0 braided suture is first passed from the aortic to ventricular side, in the interleaflet triangle, and then out to the aortic side at the same level on the adjacent leaflet. The suture is then tied on a free pledget. The height of the suture can be adjusted according to the degree of desired annular reduction and size of the annulus. The technique is typically performed at midcommissural height (Fig. 36.1b).

36.5.10  Root Preservation The techniques for AVSRR, both remodeling and reimplantation varieties, have been extensively described [41–48]. Both techniques can be modified for the bicuspid aortic valve [28, 30]. An

36  Biscuspid Aortic Valve Repair

547

important modification of the reimplantation technique for the BAV involves making the type 1 valve symmetrical within the prosthesis. This involves forcing both the conjoint and non-­ conjoint leaflets to occupy 50% of the circumference of the graft. This maneuver provides additional cusp tissue, increases leaflet mobility and coaptation, and helps avoid the need for cusp augmentation. There are some data to suggest that this may be useful for durability of the repair [54–56]. It has been demonstrated that BAV orientation closest to 180° portends favorable results after bicuspid repair [48]. Data have also demonstrated the reimplantation technique is associated with greater degrees of reduction in annular dimension and improved freedom from AI and reoperation [29, 45, 57–59]. An important caveat, however, is that a larger tube graft is required for a bicuspid valve reimplantation versus for a trileaflet valve. While an annuloplasty is usually advisable, the choice of technique is multifactorial, taking into account AVJ, sinus, and STJ diameters and tissue quality of the aortic root. The current recommended size threshold for concomitant replacement of the aortic root or ascending aorta is 4.5 cm when BAV surgery is to be performed [60–63]. In the setting of bicuspid aortic valve repair, replacement of the aortic root consists of an AVSRR utilizing either the reimplantation or remodeling techniques. In addition to size, AVSRR should also be considered if the quality of the aortic root is deemed to be poor with very thin, fragile tissue [28]. When performed, AVSRR serves not only as prophylaxis against future a­ ortic events but

as an annuloplasty for the valve repair. When a root replacement is unnecessary, the FAA can be stabilized using a subcommissural annuloplasty with Cabrol sutures, with or without downsizing replacement of the ascending aorta.

36.6 Outcomes In the largest published series, Svensson et  al. [35] evaluated long-term outcomes of bicuspid valve repairs in 728 patients, with a median follow-­up time of 8.3 years. Techniques of repair in this analysis included Cabrol sutures, plication of incompletely fused cusps, suspensory sutures, and mobilization of the commissures within a tube graft. Hospital mortality was 0.41%. Longterm survival was 97%, 94%, and 88% at 5, 10, and 15 years, respectively. For the entire series, freedom from reoperation was 87%, 78%, and 64% at 5, 10, and 15  years, respectively, corresponding to a linearized risk of reoperation of 2.6%/year. In the more recently operated patients with more aggressive and modern repair techniques, the reoperation rate appeared to be lower. The most common reasons for reoperation were cusp prolapse (38%), combined aortic stenosis and regurgitation (17%), and root aneurysm causing AI (15%) (Fig.  36.5). Risk factors for reoperation during the late hazard phase were greater LV end-systolic volume and greater LV wall thickness. This study also provided interesting information regarding the progression of transvalvular gradients and LV mass following AV repair based on echocardiographic follow-up. Aortic valve gradients initially fell after repair

Reoperation Reasons AS/AR (17%) Prolapse (38%) Root Aneurysm (15%) Other (3%) AS/Aneurysm (15%)

Fig. 36.5  Reasons for reoperation following bicuspid repair (AR aortic regurgitation, AS aortic stenosis, AVR aortic valve replacement, CABG coronary artery bypass

AS Only (12%)

grafting, MVR mitral valve replacement). Reprinted with permission [35]

N. Do et al.

548

but then began to rise steadily after approximately 3 years (Fig. 36.6). Postoperative change in LV mass index followed a similar pattern (Fig.  36.7). Of note, none of the patients who required a reoperation died from the procedure; hence patients did not pay a price in survival from bicuspid valve repair. Boodhwani et al. [28] evaluated early and midterm results of a systematic approach to bicuspid aortic valve repairs in 122 consecutive patients. No early mortality occurred, and late survival was 97% at 8 years. Freedom from late AV reoperation

AV Mean Gradient (mmHg)

a

25

20

15

10

Preop 0

3

6

9

12

15

9

12

15

Years

b

40

AV Peak Gradient (mmHg)

Fig. 36.6  Aortic valve (AV) gradient after repair of bicuspid aortic valve disease. Circles represent grouped data without regard to repeated measurements. Solid lines are model-­ based estimates. Preoperative (Preop) values shown as a bold dot. (a) Mean gradient. (b) Peak gradient. Reprinted with permission [35]

was 98% and 87% at 5 and 8 years, respectively, and freedom from recurrent AI was 94% at 5 years. Further comparison of repairs with a root replacement procedure versus subcommissural annuloplasty or supracoronary aortic replacement showed statistically significant greater freedom from recurrent AI in the root replacement group (95% vs 80% at 5 years, p = 0.03). The use of a pericardial patch repair for cusp augmentation was associated with lower freedom from recurrent AI.  In a later paper from the same group, Price et al. compared 163 BAV repairs with 307 tricus-

35

30

25

20 Preop

0

3

6 Years

36  Biscuspid Aortic Valve Repair

549

150 145

LV Mass Index (g •m-2)

140 135 130 125 120 115 110

Preop 0

3

6 Years

9

12

15

Fig. 36.7  Left ventricular (LV) mass index after aortic valve repair for bicuspid disease. Circles represent grouped data without regard to repeated measurements.

Solid line is model-based estimate. Preoperative (Preop) value shown as a bold dot. Reprinted with permission [35]

pid repairs [57]. The study demonstrated that BAV repair has similar outcomes compared to tricuspid aortic valve repair in terms of freedom from AI, freedom from aortic valve reoperation, and freedom from aortic valve replacement. In this series, BAV repair patients who underwent AVSRR had lower risk of recurrent AI. Schafers et al. [47] reported their early experience with bicuspid aortic valve repair in 173 patients. The authors found that freedom from reoperation was 97% at 5 years in those undergoing BAV repair with concomitant remodeling AVSRR.  In contrast, freedom from reoperation was only 53% in those not undergoing root replacement. A later update of their experience [48] in 316 patients showed a 10-year survival of 92% and a 10-year freedom from reoperation of 81%. Independent predictors of the need for reoperation included age (HR 0.96), AVJ diameter ≥29 mm (HR 1.30), effective height ≥9 mm (HR 0.74), commissural orientation 1  cm/year or an overall aortic diameter of >5.5  cm [23]. Continued patency, rather than 82.3.1 Role of the Aortic Arch complete thrombosis of the false lumen, is associated with aortic dilatation during follow-up and Treatment of the acutely dissected aortic arch poorer outcomes [24]. More than 60% of deaths remains unresolved. At present there is growing associated with acute type B aortic dissection consensus that any dissected arch should be result from focal rupture of the false lumen. explored during hypothermic circulatory arrest. Approximately 30% of type B dissections are In the absence of an arch tear, an open distal classified as complicated at initial presentation anastomosis of the graft and the conjoined aortic with malperfusion or impending rupture [25]. wall layers at the junction of the ascending and Endovascular repair of the thoracic aorta (thoarch portions are justified. In dissecting and non-­ racic endovascular aneurysm repair—TEVAR) dissecting aneurysms extending to the down- has changed the management algorithm for

82  Endovascular Treatment of Aortic Diseases

1175

Fig. 82.3  TEVAR in type A aortic dissection: localised type A dissection shown a simple large entry which was successfully sealed by a suitable stent graft followed by successful remodelling

1176

A. Mitsis and C. A. Nienaber

Fig. 82.4  TEVAR in type B aortic dissection: subacute type B dissection case subjected to elective stent grafting leads to successful remodelling over time. Note the reaction of the false lumen from left to right

patients with acute complicated Stanford type B aortic dissection [26]. Endovascular treatment has the potential advantages of excluding and obliterating the false lumen to correct malperfusion and protecting against long-term aneurysmal dilatation of the aorta by remodelling the aorta with expansion of the true lumen at the expense of the false lumen (Figs. 82.4 and 82.5). The relative safety of endovascular treatment has replaced open surgery as the first-line treatment for complicated type B dissection. Open repair is now rarely indicated for the management of type B aortic dissection due to the excellent results achieved with an endovascular intervention. The core principle is to place a covered stent graft over the entry tear in the descending thoracic aorta, resulting in depressurisation of the false lumen and rapid expansion of the true lumen. TEVAR is the treatment of choice in these complicated acute type B AD cases [23].

82.4.1.1 Technique Planning for TEVAR includes clinical examination, laboratory tests and imaging to classify the type of dissection, its duration and its possible complications. The most important aspect is to

identify the entry tear; the next step is to identify the extension of the dissection and the possible involvement of side branches resulting to any malperfusion. TEVAR is usually performed via femoral artery access and retrograde transarterial advancement of a large bore device (20–26  F) housing the collapsed self-expandable stent graft. Arterial access is obtained by a surgical cut-down or percutaneous puncture (cut-down). The stent graft is delivered over a stiff wire. TOE or intravascular ­ultrasonography is helpful in identifying the narrow true lumen and the correct position of the guide wire. When the target position is reached, it is crucial to temporarily reduce blood pressure (either pharmacologically or by rapid pacing) to avoid displacement of the device. To prevent an iatrogenic (treatment-related) retrograde Stanford type A dissection, endografts should be sized to the diameter of the non-dissected proximal landing zone in the healthy region of the aorta (usually just proximal to the left subclavian artery) to avoid traumatic oversizing; balloon dilatation of the stent graft should be avoided. The length of the aorta to be covered by endograft remains controversial. In the setting of a contained rupture in aortic dissection, the

82  Endovascular Treatment of Aortic Diseases 18/08/2016

1177 05/09/2016

06/01/2017

Fig. 82.5  TEVAR in type B aortic dissection: (a) Computed tomography (cross sectinal, saggital oblique and 3D reconstracted images) of a patient presenting with a localised dissection of the descending aorta (b) CT images at the same level immediate post- TEVAR (c) CT images at 4 month follow up illustrating the ideal remodelling of the aorta

descending thoracic aorta should be stented from the left subclavian artery to just above the coeliac trunk to minimise retrograde perfusion of the false lumen. In patients with dynamic (intermittent) malperfusion, coverage of the proximal entry tear using an endograft will usually expand the true lumen sufficiently to reperfuse ischaemic viscera and legs. Cases in which it is not clear whether reperfusion has taken place, the true lumen may be expanded further distally by implantation of a bare stent [26].

82.4.2 Uncomplicated Type B Dissection of the descending aorta is less frequently lethal, and medical management of uncomplicated type B dissection results in approximately 90% survival to hospital discharge

[27]. Traditionally, in the absence of malperfusion or signs of early progression, medical therapy alone to control pain and blood pressure used to be the treatment of choice. So, in an uncomplicated type B dissection, a primary conservative approach with close surveillance was justified until complications arose. However, an early TEVAR with the aim to stabilise the tear and prevent late complications by inducing aortic remodelling processes is not only feasible today but supported by recent studies in favour of this approach. The ADSORB trial investigated the use of stent-graft treatment in acute uncomplicated patients and showed that stent-graft treatment can result in aortic remodelling, but longer-term results are required to determine whether this treatment has an effect on survival [28]. The INSTEAD trial investigated the use of endovas-

1178

cular treatment in uncomplicated patients, and INSTEAD-XL showed that stent-graft treatment can improve aortic-specific survival at 5 years; thus pre-emptive TEVAR should be considered in these patients to improve late outcomes [29]. This trial also showed that endovascular repair is still associated with complications, such as stroke and paraplegia, and that aortic events continue to occur even after endovascular treatment. The concept of TEVAR already embraced to replace open surgery for managing complications of type B dissection (even without any randomised data) may now be extended to manage stable (initially uncomplicated) type B dissection because the potential to remodel dissected aorta and prevent late expansion and malperfusion has been confirmed [30].

82.5 Intramural Haematoma (IMH) An important early stage of aortic dissection is intramural haematoma (IMH), which usually presents with the same clinical picture and risk profile as overt aortic dissection. With a pathogenesis that explains the high rate of progression to overt aortic dissection and a prognosis and survival that are similar to those in aortic dissection, urgent diagnosis of IMH is very important. Intramural haematoma is diagnosed in the presence of a circular or crescent-shaped thickening of >5 mm of the aortic wall in the absence of detectable tear [31]; 10% of IMHs are seen in the ascending aorta or aortic arch (type A), respectively, and 60–70% in the descending thoracic aorta (type B) [32, 33]. Given the lack of available evidence, the current consensus suggests that the general approach to IMH is to monitor and treat it like aortic dissection [34]. In type A IMH, swift surgery is indicated (Class I, Level of evidence C); in cases of type B IMH, initial medical therapy under careful surveillance is recommended (Class I, Level of evidence C). In complicated type B IMH, TEVAR should be considered (Class IIa, Level of evidence C). In patients with IMH, intimal lesions/laceration

A. Mitsis and C. A. Nienaber Table 82.2  Predictors of adverse evolution in the setting of intramural haematoma [1] Persistent and recurrent pain Difficult blood pressure control Ascending aorta involvement Maximum aortic diameter ≥50 mm Progressive maximum aortic wall thickness >11 mm Enlarging aortic diameter Recurrent pleural effusion Penetrating ulcer or ulcer-like projection in the involved segment Organ ischaemia

can often be found in the inner curvature of the aortic arch by careful CTA analysis. This may be a target for stent-graft implantation in patients with a progressive/complicated IMH (Table 82.2) [1].

82.6 P  enetrating Aortic Ulcer (PAU) Penetrating aortic ulcer (PAU) is defined as ulceration of an aortic atherosclerotic plaque penetrating the internal elastic lamina into the media. They progressively enlarge and develop into either a saccular or fusiform aneurysm (type A PAU) [35, 36]. The most common location of PAU is the middle and lower descending thoracic aorta (type B PAU) followed less frequently by lesions in the aortic arch or abdominal aorta and rarely in the ascending aorta. Persistent pain, haemodynamic instability and expansion >20 mm width or depth represent a higher risk for disease progression and should trigger surgical or endovascular treatment as transmural rupture ranges from 8 to 42% [37–39]. Lacerations and IMH usually occur at points of greatest hydraulic stress (right lateral ascending aorta or adjacent to the ligamentum arteriosum), whereas penetrating ulcers are typically found in the descending or abdominal aorta. In cases of penetrating aortic ulcer, treatment may be recommended, when patients are symptomatic or the ulcer demonstrates expansion. TEVAR is increasingly becoming popular intervention since the patients are often poor candidates for

82  Endovascular Treatment of Aortic Diseases

conventional surgery due to advanced age and comorbidities.

82.7 Aortic Trauma Aortic trauma is usually caused by rapid deceleration in serious motor vehicle accidents, pedestrian injuries and fall from height [40]. The aortic segment that is subject to the greatest strain is located just beyond the aortic isthmus, with aortic ruptures occurring at this location in about 90% of cases [41]. Recent reports indicate that intimal haemorrhage (with and without partial intimal laceration) tends to heal spontaneously; when the lesion involves intimal and medial layers, false aneurysm formation occurs [42]. The aneurysm is fusiform in the case of a circumferential lesion, while in a partial laceration, it appears as localised diverticulum. Aneurysm progression in the acute and subacute phases is a concern, especially with circumferential lesions, and close monitoring is warranted to anticipate surgical repair [43]. Recently, the development of endovascular technique has provided additional opportunities for the treatment of acute onset and chronic aortic trauma [44–46]. Immediate endovascular treatment is indicated in patients with complete transection of the aortic wall and free bleeding into the mediastinum or pseudocoarctation syndrome, whereas delayed treatment is advised when limited disruption of the aorta is present with media and adventitia still intact (Fig. 82.6).

82.8 Aneurysms of the Thoracic Aorta Second to atherosclerosis, aneurysm is the most frequent disease affecting the aorta; 20–27% of patients with abdominal aortic aneurysm (AAA) have both synchronous and consecutive thoracic aortic aneurysms (TAA), which are more common in women and in the elderly [47, 48]. Patients with aortic aneurysm are at increased risk of cardiovascular events with a 10-year risk of mortality from another cardiovascular cause as high as 15 times the risk of an aorta-related death

1179

[49]. Asymptomatic aneurysms are initially managed medically, while repair is indicated for symptomatic and expanding aneurysms and those beyond 55 mm in diameter in the ascending aorta or 60  mm in the descending aorta regardless of the site or symptoms. A novel predictor for rupture of thoracic aortic aneurysm, the aortic size index, may be useful to predict increasing rates of rupture, dissection or death [50]. Individual body surface area information is utilised for the aortic size index (aortic diameter/m2) enabling improved and individualised selection for surgical repair. An aortic size index stratification ≥2.75  cm/m2 represents a low risk (approximately 4%/year), 2.75–4.24 cm/m2 a moderate risk (approximately 8% per year) and >4.25  cm/m2 a high risk (approximately 20% per year), underlining the importance of relative aortic size for predicting complications [51]. Aortic dissection and rupture are most severe complications leading to high operative risk in urgent or emergency situations. Operative mortality has been reported between 1.5 and 2.6% [52, 53]. In addition to medical treatment, surgical and/or endovascular management may become necessary. Regarding endovascular treatment, the diameter (20  mm) of the healthy proximal and distal landing zones are evaluated to assess the feasibility of TEVAR. If landing zones are shorter or significantly angulated, prior transposition or bypass surgery/rerouting of the involved aortic branch may have to be considered. In TAA, the stent-­graft diameter should exceed the reference aortic diameter at the landing zones by at least 10–15%. Evaluation of access vessels (sizing, calcification and tortuosity) is also of major importance. Alternative access sites are the iliac arteries, the infrarenal aorta or even directly the ascending aorta. Anatomic constraints in the infrarenal aorta or the thoracoabdominal transition with excessive tortuosity may lead to a loss of pushability and may preclude advancement of the prosthesis into the area of interest. These difficult situations may be overcome using, in addition to a superstiff guidewire, a second protected stiff buddy wire or a pull-through procedure via the brachial artery.

1180

A. Mitsis and C. A. Nienaber

a

b

Fig. 82.6 (a) Traditional surgical repair of a traumatic lesion just beyond the aortic isthmus with an interposition graft. (b) Novel endovascular approach with placing a

suitable comfortable stent graft across a traumatic aortic lesion to prevent rupture

82  Endovascular Treatment of Aortic Diseases

No randomised trials exist to guide the choice between open surgery and TEVAR in the treatment of descending aortic aneurysms. Non-­ randomised comparisons and meta-analyses have shown lower early mortality after TEVAR than open surgery even in sicker patients. Nevertheless overall mid-term survival does not differ between TEVAR and surgery [54–57].

82.8.1 Coarctation Coarctation is a common congenital anomaly, with an incidence of 20–60 per 100,000 live births, and represents 5–8% of all congenital cardiovascular disorders [58]. It is typically located opposite the ligamentum arteriosum just distal to the left subclavian artery. It has male predominance and is often associated with other conditions (such as bicuspid aortic valve, patent ductus arteriosus and aneurysms of the circle of Willis) or syndromes (such as Shone’s complex or Turner’s syndrome) [58]. There is usually a dilation of the descending aorta distal to the coarctation, with a possible dual pathology of reduced integrity of the vessel wall and abnormal flow patterns leading to local shear stress and resulting dilatation [59]. As a result of the obstruction caused by the coarctation, collateral vessels develop to increase flow into the descending aorta. Increased flow through intercostal arteries results in their dilation and notching along the inferior aspect of the ribs, which usually takes 8–10  years to become significant enough to be observed on a chest radiograph. The presence of collateral flow indicates a haemodynamically significant coarctation. Furthermore, a pressure gradient greater than 20 mm Hg is considered an indication for intervention. Cardiac catheterisation with manometry (a peak-to-peak gradient of 20 mmHg indicates a haemodynamically significant coarctation of the aorta in the absence of well-developed collaterals) and angiography are still the ‘gold standard’ for evaluation of this condition at many centres before and after operative or interventional treatment.

1181

Several therapeutic strategies are available for the treatment of aortic coarctation depending on the morphology of the affected aorta as well as the age and clinical condition of the patient [60, 61]. Surgery for aortic coarctation in particular an end-to-end anastomosis is recommended at an early age because long-term results seem to be sustainable. Recently, interventional procedures and balloon angioplasty have come into widen use and provide good results [62, 63]. An accurate selection of favourable anatomy by high-­ resolution imaging modalities is particularly important for interventional procedures to ensure a low rate of complications [64]. The question of whether to use covered or non-covered stents remains unresolved. Notably, despite intervention, antihypertensive drugs may still be necessary. Aneurysm formation, restenosis, aortic dissection and pseudoaneurysms have been reported after surgery or balloon angioplasty in up to 42% of patients [65]. Therefore, routine follow-up is recommended for patients who underwent repair of an aortic coarctation, independently of both surgical technique and timing of repair. New interventional techniques, such as endovascular stent grafting, have currently been applied to the treatment of postsurgical patch aneurysms with excellent results, avoiding the need for further surgical intervention [66, 67].

82.9 Special Pathologies 82.9.1 Marfan Syndrome Among hereditary diseases affecting the aorta, Marfan syndrome is the most prevalent connective tissue disorder, with an estimated incidence of 1 in 7000 and autosomal dominant inheritance with variable penetrance. More than 150 mutations on the fibrillin-1 (FBN-1) gene have been identified encoding for a defective fibrillin in the extracellular matrix, which may affect the ocular, cardiovascular, skeletal and pulmonary systems, as well as the skin and dura mater. The diagnosis of Marfan syndrome is currently based on the revised clinical criteria of the ‘Ghent nosology’

1182

[68]. Marfan syndrome and the delineation of criteria for differentiating other inherited conditions (genotypes) from the Marfan phenotype are attracting interest [69, 70]. Prognosis is mainly determined by progressive dilation of the aorta, leading to aortic dissection or rupture, which is the major cause of death with untreated patients reaching on average an age of 40 years. Dilation of the aortic root is found in 60–80% of patients. The rate of dilation is heterogeneous and unpredictable. The risk of type A dissection clearly increases with increasing aortic root diameter, but dissection may occasionally occur even in patients with only mild aortic dilation. Composite replacement of the aortic valve and ascending aorta has become a low risk and a very durable operation in experienced centres. In patients with anatomically normal valves, in whom the insufficiency is due to the dilated annulus or dissection, valve-sparing operations with root replacement by a Dacron prosthesis and with reimplantation of the coronary arteries into the prosthesis (David’s procedure) or remodelling of the aortic root (Yacoub’s procedure) have now become the preferred surgical procedures [71]. The placement of a personalised external aortic root support (PEARS), computer designed and manufactured to match the aortic root morphology of the individual patient, was introduced in 2004 as a conservative approach for Marfan patients; PEARS is an emerging technology and seems that it reduces the longitudinal stresses in the ascending aorta by up to 52% [72, 73].

82.9.2 Ehlers–Danlos Syndrome Ehlers–Danlos syndrome is a heterogeneous group of hereditable connective tissue disorders characterised by articular hypermobility, skin hyperextensibility and tissue fragility. Eleven types of Ehlers–Danlos syndrome have been characterised: the true prevalence is unknown. An aggregate incidence of 1  in 5000 births is often cited, with no racial or ethnic predisposition. Aortic involvement is seen primarily in autosomal dominant Ehlers–Danlos syndrome type IV. There are no data to support a thresh-

A. Mitsis and C. A. Nienaber

old diameter for intervention in cases of thoracic aneurysms following an individualised approach [74].

82.9.3 Loeys–Dietz Syndrome Loeys–Dietz syndrome is an autosomal dominant aortic aneurysm syndrome combining the triad of arterial tortuosity and aneurysms in the whole arterial tree, hypertelorism and bifid uvula. Observation in both children and adults led to the recommendation of early surgical intervention at ascending aortic diameters of >42  mm [75]. In general TEVAR is not recommended in patients with connective tissue disease except as a bailout procedure or bridge to definitive open surgical therapy or as a procedure following prior aortic repair when both landing zones lie between prosthetic grafts (e.g. intercostal patch aneurysm after Marfan’s TAA repair) [76, 77].

82.10 Summary and Outlook Endovascular keyhole interventions to the aorta, without the trauma and complexity of open surgery, especially in the light of our ageing comorbid population, will be in clear demand for both distal and even proximal aortic pathologies. TEVAR offers a valid treatment option for the elderly patients deemed at excessive risk for open surgery but also for fit patients with suitable anatomies. The role of a multidisciplinary Aortic Centre is crucial for both management and treatment of acute and chronic aortic pathologies. The ideal Aortic Team consists of a team of cardiac and vascular surgeons, interventional cardiologists, anaesthesiologists and imaging specialists who work together to improve diagnoses, screening, treatment and aftercare of such patients. In particular, acute aortic dissection and other types of acute aortic syndrome are best managed in aortic centres by experienced teams; regionalised care and referral networks should become standard of care for this evolving field of interdisciplinary medicine.

82  Endovascular Treatment of Aortic Diseases

References

1183

thoracic aortic stent graft placement: insight from the European registry on endovascular aortic repair complications. Circulation. 2009;120:S276–81. 1. Erbel R, Aboyans V, Boileau C, Bossone E, Di 16. Midulla M, Renaud A, Martinelli T, Koussa M, Bartolomeo R, Eggebrecht H, et  al. Authors/task Mounier-Vehier C, Prat A, Beregi JP.  Endovascular force members. 2014 ESC guidelines on the diagfenestration in aortic dissection with acute malpernosis and treatment of aortic diseases. Eur Heart J. fusion syndrome: immediate and late follow-up. J 2015;36:2779. Thorac Cardiovasc Surg. 2011;142:66–72. 2. Meszaros I, Morocz J, Szlavi J, Schmidt J, Tornoci L, 17. Nienaber CA, Sakalihasan N, Clough RE, Aboukoura Nagy L, et al. Epidemiology and clinicopathology of M, Mancuso E, Yeh JS, et al. Thoracic endovascular aortic dissection. Chest. 2000;117:1271–8. aortic repair (TEVAR) in proximal (type A) aortic 3. Clouse WD, Hallett JW Jr, Schaff HV, Spittell PC, dissection: ready for a broader application? J Thorac Rowland CM, Ilstrup DM, et  al. Acute aortic disCardiovasc Surg. 2017;153:S3–11. section: population-based incidence compared with 18. Shrestha M, Martens A, Kruger H, Maeding I, Ius F, degenerative aortic aneurysm rupture. Mayo Clin Fleissner F, Haverich A. Total aortic arch replacement Proc. 2004;79:176–80. with the elephant trunk technique: single-­Centre 30-year 4. Mehta RH, et al. Chronobiological patterns of acute results. Eur J Cardiothorac Surg. 2014;45:289–96. aortic dissection. Circulation. 2002;106:1110–5. 19. Roselli EE, Rafael A, Soltesz EG, Canale L, Lytle 5. Landenhed M, et  al. Risk profiles for aortic dissecBW. Simplified frozen elephant trunk repair for acute tion and ruptured or surgically treated aneurysms: DeBakey type I dissection. J Thorac Cardiovasc Surg. a prospective cohort study. J Am Heart Assoc. 2013;145:S197–201. 2015;4:e001513. 20. Jakob H, Dohle DS, Piotrowski J, Benedik J, 6. Nienaber CA, Clough RE, Sakalihasan N, Suzuki T, Thielmann M, Marggraf G, Erbel R, Tsagakis K. Six-­ Gibbs R, Mussa F, et  al. Aortic dissection. Nat Rev year experience with a hybrid stent graft prosthesis for Dis Primers. 2016;2:16071. extensive thoracic aortic disease: an interim balance. 7. VIRTUE Registry Investigators. Mid-term outcomes Eur J Cardiothorac Surg. 2012;42:1018–25. and aortic remodelling after thoracic endovascular 21. Cao P, De Rango P, Czerny M, Evangelista A, Fattori repair for acute, subacute, and chronic aortic disR, Nienaber C, Rousseau H, Schepens M. Systematic section: the VIRTUE Registry. Eur J Vasc Endovasc review of clinical outcomes in hybrid procedures Surg. 2014;48:363–71. for aortic arch dissections and other arch diseases. J 8. Trimarchi S, Eagle KA, Nienaber CA, et  al. Role Thorac Cardiovasc Surg. 2012;144:1286–300. of age in acute type A aortic dissection outcome: 22. Eggebrecht H, Nienaber CA, Neuhäuser M, et  al. report from the International Registry of Acute Endovascular stent-graft placement in aortic dissecAortic Dissection (IRAD). J Thorac Cardiovasc Surg. tion: a meta-analysis. Eur Heart J. 2006;27:489–98. 2010;140:784–9. 23. Fattori R, Cao P, De Rango P, Czerny M, Evangelista 9. Nienaber CA, Fattori R, Lund G, et  al. Nonsurgical A, Nienaber C, et al. Interdisciplinary expert consenreconstruction of thoracic aortic dissection by stent-­ sus document on management of type B aortic dissecgraft placement. N Engl J Med. 1999;340:1539–45. tion. J Am Coll Cardiol. 2013;61:1661–78. 10. Dake MD, Kato N, Mitchell RS, et al. Endovascular 24. Onitsuka S, Akashi H, Tayama K, Okazaki T, Ishihara stent-graft placement for the treatment of acute aortic K, Hiromatsu S, et al. Long-term outcome and progdissection. N Engl J Med. 1999;340:1546–52. nostic predictors of medically treated acute type B aor 11. Chiappini B, et al. Early and late outcomes of acute tic dissections. Ann Thorac Surg. 2004;78:1268–73. type A aortic dissection: analysis of risk factors in 487 25. Fattori R, Tsai TT, Myrmel T, Evangelista A, Cooper consecutive patients. Eur Heart J. 2005;26:180–6. JV, Trimarchi S, et al. Complicated acute type B dis 12. Trimarchi S, et  al. Contemporary results of surgery section: is surgery still the best option? A report from in acute type A aortic dissection: the International the International Registry of Acute Aortic Dissection. Registry of Acute Aortic Dissection experience. J JACC Cardiovasc Interv. 2008;1:395–402. Thorac Cardiovasc Surg. 2005;129:112–22. 26. Grabenwöger M, Alfonso F, Bachet J, Bonser R, 13. Tang GH, et al. Surgery for acute type A aortic dissecCzerny M, Eggebrecht H, European Association tion in octogenarians is justified. J Thorac Cardiovasc for Cardio-Thoracic Surgery (EACTS); European Surg. 2013;145:S186–90. Society of Cardiology (ESC); European Association 14. Weigang E, Parker JA, Czerny M, Lonn L, Bonser of Percutaneous Cardiovascular Interventions RS, Carrel T, et  al. Should intentional endovascular (EAPCI), et al. Thoracic Endovascular Aortic Repair stent-graft coverage of the left subclavian artery be (TEVAR) for the treatment of aortic diseases: a posipreceded by prophylactic revascularisation? Eur J tion statement from the European Association for Cardiothorac Surg. 2011;40:858–68. Cardio-Thoracic Surgery (EACTS) and the European 15. Eggebrecht H, Thompson M, Rousseau H, Czerny Society of Cardiology (ESC), in collaboration with the M, Lönn L, Mehta RH, Erbel R, European Registry European Association of Percutaneous Cardiovascular on Endovascular Aortic Repair Complications. Interventions (EAPCI). Eur J Cardiothorac Surg. Retrograde ascending aortic dissection during or after 2012;42:17–24.

1184 27. Acosta S, Blomstrand D, Gottsater A. Epidemiology and long-term prognostic factors in acute type B aortic dissection. Ann Vasc Surg. 2007;21:415–22. 28. Brunkwall J, Kasprzak P, Verhoeven E, et  al. Endovascular repair of acute uncomplicated aortic type B dissection promotes aortic remodelling: 1 year results of the ADSORB trial. Eur J Vasc Endovasc Surg. 2014;48:285–91. 29. NienaberCA RH, Eggebrecht H, Kische S, Fattori R, Rehders TC, et al. Randomized comparison of strategies for type B aortic dissection: the INvestigation of STEnt Grafts in Aortic Dissection (INSTEAD) trial. Circulation. 2009;120:2519–28. 30. Nienaber CA, Kische S, Rousseau H, Eggebrecht H, Rehders TC, Kundt G, et al. INSTEAD-XL trial. Endovascular repair of type B aortic dissection: long-­ term results of the randomized investigation of stent grafts in aortic dissection trial. Circ Cardiovasc Interv. 2013;6:407–16. 31. Wolff KA, Herold CJ, Tempany CM, et  al. Aortic dissection: atypical patterns seen at MR imaging. Radiology. 1991;181:489–95. 32. Nienaber CA, Eagle KA.  Aortic dissection: new frontiers in diagnosis and management: part II: therapeutic management and follow-up. Circulation. 2003;108:772–8. 33. Evangelista A, et  al. Acute intramural hematoma of the aorta: a mystery in evolution. Circulation. 2005;111:1063–70. 34. Evangelista A, Czerny M, Nienaber C, Schepens M, Rousseau H, Cao P, Moral S, Fattori R. Interdisciplinary expert consensus on management of type B intramural haematoma and penetrating aortic ulcer. Eur J Cardiothorac Surg. 2015;47:209–17. 35. Nathan DP, et  al. Presentation, complications, and natural history of penetrating atherosclerotic ulcer disease. J Vasc Surg. 2012;55:10–5. 36. Robertson MM, et al. A double-blind controlled comparison of fluoxetine and lofepramine in major depressive illness. J Psychopharmacol. 1994;8:98–103. 37. Ganaha F, et al. Prognosis of aortic intramural hematoma with and without penetrating atherosclerotic ulcer: a clinical and radiological analysis. Circulation. 2002;106:342–8. 38. Coady MA, Rizzo JA, Elefteriades JA.  Pathologic variants of thoracic aortic dissections: penetrating atherosclerotic ulcers and intramural hematomas. Cardiol Clin. 1999;17:637–57. 39. Movsowitz HD, Lampert C, Jacobs LE, et  al. Penetrating atherosclerotic aortic ulcers. Am Heart J. 1994;128:1210–7. 40. Richens D, Kotidis K, Neale M, et  al. Rupture of the aorta following road traffic accidents in the United Kingdom 1992–1999: the results of the co-­ operative crash injury study. Eur J Cardiothorac Surg. 2003;23:143–8. 41. Kodali S, Jamieson WR, Leia-Stephens M, Miyagishima RT, Janusz MT, Tyers GF.  Traumatic rupture of the thoracic aorta. A 20-year review: 1969– 1989. Circulation. 1991;84:S40–6.

A. Mitsis and C. A. Nienaber 42. Hunt JP, Baker CC, Lentz CW, Rutledge RR, Oller DW, Flowe KM, et al. Thoracic aorta injuries: management and outcome of 144 patients. J Trauma. 1996;40:547–55. 43. Rousseau H, Soula P, Perreault P, et  al. Delayed treatment of traumatic rupture of the thoracic aorta with endoluminal covered stent. Circulation. 1999;99:498–504. 44. Scheinert D, Krankenberg H, Schmidt A, et  al. Endoluminal stent-graft placement for acute rupture of the descending thoracic aorta. Eur Heart J. 2004;25:694–700. 45. Demers P, Miller C, Scott Mitchell R, et al. Chronic traumatic aneurysms of the descending thoracic aorta: mid-term results of endovascular repair using first and second-generation stent-grafts. Eur J Cardiothorac Surg. 2004;25:394–400. 46. Hultgren R, Larsson E, Wahlgren CM, Swedenborg J.  Female and elderly abdominal aortic aneurysm patients more commonly have concurrent thoracic aortic aneurysm. Ann Vasc Surg. 2012;26:918–23. 47. Chaer RA, et  al. Synchronous and metachronous thoracic aneurysms in patients with abdominal aortic aneurysms. J Vasc Surg. 2012;56:1261–5. 48. Karthikesalingam A, et al. The shortfall in long-term survival of patients with repaired thoracic or abdominal aortic aneurysms: retrospective case-control analysis of hospital episode statistics. Eur J Vasc Endovasc Surg. 2013;46:533–41. 49. Davies RR, Gallo A, Coady MA, Tellides G, Botta DM, Burke B, Coe MP, Kopf GS, Elefteriades JA. Novel measurement of relative aortic size predicts rupture of thoracic aortic aneurysms. Ann Thorac Surg. 2006;81:169–77. 50. Muhs BE, et  al. Dynamic cine-CT angiography for the evaluation of the thoracic aorta; insight in dynamic changes with implications for thoracic endograft treatment. Eur J Vasc Endovasc Surg. 2006;32:532–6. 51. Elefteriades J. Natural history of thoracic aortic aneurysms: indications for surgery, and surgical versus nonsurgical risks. Ann Thorac Surg. 2002;74:S1877–80. 52. Perreas K, Samanidis G, Dimitriou S, Kalogris P, Balanika M, Antzaka C, Khoury M, Michalis A.  Outcomes after ascending aorta and proximal aortic arch repair using deep hypothermic circulatory arrest with retrograde cerebral perfusion: analysis of 207 patients. Interact Cardiovasc Thorac Surg. 2012;15:456–61. 53. Walsh SR, et  al. Endovascular stenting versus open surgery for thoracic aortic disease: systematic review and meta-analysis of perioperative results. J Vasc Surg. 2008;47:1094–8. 54. Cheng D, et al. Endovascular aortic repair versus open surgical repair for descending thoracic aortic disease. A systematic review and meta-analysis of comparative studies. J Am Coll Cardiol. 2010;55:986–1001. 55. Bavaria JE, et  al. Endovascular stent grafting versus open surgical repair of descending thoracic aortic aneurysms in low-risk patients: a multi-

82  Endovascular Treatment of Aortic Diseases center comparative trial. J Thorac Cardiovasc Surg. 2007;133:369–77. 56. Makaroun MS, Dillavou ED, Wheatley GH, Cambria RP.  Five-year results of endovascular treatment with the Gore TAG device compared with open repair of thoracic aortic aneurysms. J Vasc Surg. 2008;47:912–8. 57. Baumgartner H, Bonhoeffer P, De Groot NM, de Haan F, Deanfield JE, Galie N, et  al. ESC guidelines for the management of grown-up congenital heart disease (new version 2010). Eur Heart J. 2010;31:2915–57. 58. Bissell MM, Hess AT, Biasiolli L, Glaze SJ, Loudon M, Pitcher A, et  al. Aortic dilation in bicuspid aortic valve disease: flow pattern is a major contributor and differs with valve fusion type. Circ Cardiovasc Imaging. 2013;6:499–507. 59. Carr JA.  The results of catheter-based therapy compared with surgical repair of adult aortic coarctation. J Am Coll Cardiol. 2006;47:1101–7. 60. Gibbs JL.  Treatment options for coarctation of the aorta. Heart. 2000;84:11–3. 61. Mullen MJ. Coarctation of the aorta in adults: do we need surgeons? Heart. 2003;89:3–5. 62. Ovaert C, McCrindle BW, Nykanen D, et al. Balloon angioplasty of native coarctation: clinical outcomes and predictors of success. J Am Coll Cardiol. 2000;35:988–96. 63. Paddon AJ, Nicholson AA, Ettles DF, et al. Long-term follow-up of percutaneous balloon angioplasty in adult aortic coarctation. Cardiovasc Intervent Radiol. 2000;23:364–7. 64. Varma C, Benson LN, Butany J, et al. Aortic dissection after stent dilatation for coarctation of the aorta: a case report and literature review. Catheter Cardiovasc Interv. 2003;59:528–35. 65. Bell RE, Taylor PR, Aukett M, et  al. Endoluminal repair of aneurysms associated with coarctation. Ann Thorac Surg. 2003;75:530–3. 66. Ince H, Petzsch M, Rehders T, et  al. Percutaneous endovascular repair of aneurysm after previous coarctation surgery. Circulation. 2003;108:2967–70.

1185 67. De Paepe A, Devereux RB, Dietz HC, Hennekam RC, Pyeritz RE. Revised diagnostic criteria for the Marfan syndrome. Am J Med Genet. 1996;62:417–26. 68. Collod G, et  al. A second locus for Marfan syn drome maps to chromosome 3p24.2-p25. Nat Genet. 1994;8:264–8. 69. Milewicz DM, Pyeritz RE, Crawford S, Byers PH. Marfan syndrome: defective synthesis, secretion, and extracellular matrix formation of fibrillin by cultured dermal fibroblasts. J Clin Invest. 1992;89:79–86. 70. Kallenbach K, Baraki H, Khaladj N, Kamiya H, Hagl C, Haverich A, Karck M. Aortic valve-sparing operation in Marfan syndrome: what do we know after a decade? Ann Thorac Surg. 2007;83:S764–8. 71. Treasure T, Takkenberg JJ, Golesworthy T, Rega F, Petrou M, Rosendahl U, et  al. Personalised external aortic root support (PEARS) in Marfan syndrome: analysis of 1-9 year outcomes by intention-to-treat in a cohort of the first 30 consecutive patients to receive a novel tissue and valve-conserving procedure, compared with the published results of aortic root replacement. Heart. 2014;100:969–75. 72. Singh SD, Xu XY, Pepper J, Izgi C, Treasure T, Mohiaddin RH.  Effects of aortic root motion on wall stress in the Marfan aorta before and after personalised aortic root support (PEARS) surgery. J Biomech. 2016;49:2076–84. 73. Germain DP.  Ehlers-Danlos syndrome type IV. Orphanet J Rare Dis. 2007;2:32. 74. Loeys BL, Schwarze U, Holm T, Callewaert BL, Thomas GH, Pannu H, et  al. Aneurysm syndromes caused by mutations in the TGF-beta receptor. N Engl J Med. 2006;355:788–98. 75. Cooper DG, Walsh SR, Sadat U, Hayes PD, Boyle JR. Treating the thoracic aorta in Marfan syndrome: surgery or TEVAR? J Endovasc Ther. 2009;16:60–70. 76. Ince H, Rehders TC, Petzsch M, Kische S, Nienaber CA. Stent-grafts in patients with Marfan syndrome. J Endovasc Ther. 2005;12:82–8. 77. Adams JD, Kern JA. Blunt thoracic aortic injury: current issues and endovascular treatment paradigms. Endovasc Today. 2014;13:38–42.

Secondary Procedures After Primary Thoracic Endovascular Aortic Repair (TEVAR): Pathologies and Management

83

Michal Nozdrzykowski, Friedrich-Wilhelm Mohr, and Jens Garbade

83.1 Introduction The introduction in the early 1990s and development of new stent grafts have expanded the therapeutic spectrum onto aortic disease [1]. Initially, TEVAR (thoracic endovascular aortic repair) was reserved for the acute treatment of high-risk patients with unclear overall life expectancy. The procedural success has led to a progressive use, and now TEVAR has evolved and is proposed as an alternative for most thoracic aortic pathologies affecting the descending or thoracoabdominal aorta. In compare to open aortic surgery, TEVAR shows lower perioperative mortality and morbidity rates. Although there are no direct comparative studies between open and endovascular repair for chronic dissection, retrospective reports indicate similar long-term outcome despite being early [2]. However, the long-term results have remained limited, and the durability of TEVAR continues to be a concern. Furthermore, the application of endograft technology has expanded to include “offlabel” indications, such as aneurysmal disease with borderline anatomy and landing zones, arch hybrid procedures, traumatic transection and infection. With the wide use of TEVAR, the number of serious complications increases [3–6]. It seems likely M. Nozdrzykowski (*) · F.-W. Mohr · J. Garbade Department of Cardiac Surgery, Heart Center, University of Leipzig, Leipzig, Germany e-mail: [email protected]; [email protected]

that the incidence of severe stent graft-related complications or even TEVAR failure is currently under-reported in the medical literature. The reintervention can occur up to 30% at 3 years in cases of aortic dissection with aneurysmal degeneration [7]. Although the secondary interventions are often endovascular, patients with major complications require conversion to open repair. In large studies the incidence ranges between 2.2 and 7.9% [5, 8, 9]. The surgical approach is challenging and often not well-defined.

83.2 T  ype of Complication After Thoracic Endovascular Aortic Repair (TEVAR) 83.2.1 Endoleakage The most frequent complication after TEVAR is the occurrence of endoleaks, which are usually treated endovascular (EVAR) according to the source of blood flow causing the leak. Type I indicates a leak at the proximal (Ia) or distal (Ib) attachment sites to the vessel wall and can be seen during insertion of the initial stent graft or during a follow-up surveillance imaging exam. Type II indicates retrograde flow from a branch artery. Type III indicates a lack at a graft junction when multiple grafts are used. Type IV is endotension from graft porosity and identified during implantation of the device, when the patient is

© Springer-Verlag GmbH Austria, part of Springer Nature 2019 O. H. Stanger et al. (eds.), Surgical Management of Aortic Pathology, https://doi.org/10.1007/978-3-7091-4874-7_83

1187

1188

fully anticoagulated. These endoleaks are rarely seen in the devices that are used nowadays and will seal spontaneously when the coagulation profile is normalized. Endoleaks have an incidence ranging from 1 to 29% [10–12], and proximal type Ia endoleak is the most prevalent type. Risk factors for the development of an endoleak are the coverage of the left subclavian artery, a larger diameter aneurysm aorta, an extensive aortic coverage with multiple stent grafts, shorter-­than-­recommended landing zones and a male sex [12]. Parmer and colleagues suggested that using devices that are longer instead of multiple short devices may provide superior results by decreasing the incidence of postoperative endoleak [12]. The occurrence of late endoleak represents one of the major limitations of endovascular treatment and illustrates the need for mandatory lifelong imaging follow-up. The imaging algorithm in case of detectable endoleaks cannot be generalized and remains according to the individual discretion of the treating physician. For the moment multi-detector CT angiography (MDCTA) is the most widely used technique for detection of endoleaks, and classification is done by digital subtraction angiography (DSA).

83.2.1.1 Management of Endoleakage Conventional endoleak management consists of aggressive repair of type I and III endoleaks and observation of type II leaks [12, 13]. However, currently exists any consensus concerning interventional/endovascular or open surgical reintervention in cases of type I or III endoleaks. Endovascular/interventional repair includes various catheter techniques such as re-­TEVAR, reballooning of the implanted stent graft, endovascular coiling of aortic branches and endostapling. Regarding type I endoleaks, if an adequate landing zone for additional stent graft placement is available (at least 20 mm in the zone 3 or 4), second-time TEVAR (re-TEVAR) is the method of first choice. If re-TEVAR is not feasible because of steep angulation of the aortic arch, too short proximal or distal landing zone or progression of the aortic lesion to the aortic arch,

M. Nozdrzykowski et al.

conversion to the open repair is required. Moreover in young patients, open repair will be used more frequently because of unclear durability of the stent grafts. In our study, the type Ia endoleak was the most common indication (32%) for open conversion [5]. In a study by Roselli et al. [8] of 50 patients who underwent secondary open repair after TEVAR, 14 (28%) of those were type Ia endoleak. The causes of those endoleaks were important morphological details such as too short as recommended landing zone, massive aortic calcification and large aneurysm during primary stent graft implantation. Open surgical repair in most cases of type Ia endoleak required replacement of the aortic arch and even ascending aorta by the reverse frozen elephant trunk technique. The reverse frozen elephant trunk is the reversed modification of the single-stage frozen elephant trunk technique. This procedure is performed with the use of deep hypothermic circulatory arrest with right axillary artery cannulation with or without selective brain perfusion based on the surgeon’s choice and the complexity of the reconstruction needed. The ascending and total arch is replaced with separate reimplantation of the arch vessels. The distal end of the surgical graft will be then sutured directly to the stent graft and native descending aorta. This procedure can be combined with additional stent grafts deployment antegradely at reverse frozen elephant trunk repair to extend the previous TEVAR.  Another option represents arch debranching with antegrade stent graft delivery. Chronic type Ib endoleak required thoracoabdominal repair using the left heart bypass technique. Surgical access will be achieved via left-sided posterolateral thoracotomy through the fifth or sixth intercostal space and, if required, retroperitoneal approach to gain access to the abdominal aorta. For cardiopulmonary bypass the cannulation of the femoral vessels will be performed. After clamping of the aorta, complete removal of the stent graft is recommended. If the stent graft is proximally placed in the aortic arch upon the brachiocephalic trunk, the clamp of the aorta is not feasible, and the use of hypothermic circulatory arrest is necessary.

83  Secondary Procedures After Primary Thoracic Endovascular Aortic Repair (TEVAR): Pathologies…

1189

Table 83.1  Treatment strategy of endoleak after primary thoracic endovascular aortic repair (TEVAR) (summary) Management Watch and wait Endovascular Open surgery

Types of endoleak Type 1a No Primarya Yese

Type 1b No Primarya Yese

Type 2 Yes Secondaryb,c Yese

Type 3 No Primaryd Yese

Treatment options: extension stent graft, balloon angioplasty, endostapling, embolization If the type II endoleak persists beyond 6 months, because of sac expansion and risk of rupture c Treatment option: embolization d Treatment option: stent graft (junctional) placement e If the primary endovascularly treatment unsuitable or failed a

b

If the safe complete removal of the stent graft in proximal segment (area of the aortic arch) is not possible, partial explantation of the endograft could be performed. In these cases trimming of the stent graft and subsequent anastomosis with the aortic graft (e.g. Dacron graft), respectively with residual aorta, is possible. Re-ballooning is used to reattach the implanted stent graft to the aortic wall in cases of type I endoleak. If re-ballooning is not successful, extension of the seal zone should be performed. It can be reached using additional stent graft placement, usually with a balloon-expandable stent because of the need for large stent sizes with strong radial force. Another alternative procedure is the use of EndoAnchors (Aptus Heli-FX® EndoAnchor® System) to provide fixation and sealing between endovascular aortic grafts and the native artery. The Aptus Heli-FX EndoAnchor System is indicated for use in patients whose endovascular grafts have exhibited endoleak or migration or are at risk of such complications, in whom augmented radial fixation or sealing is required to regain or maintain adequate aneurysm exclusion. The EndoAnchor may be implanted at the time of the initial endograft placement or during a secondary (i.e. repair) procedure. This method is not recommended by excessive calcification (porcelain aorta) and excessive thrombus in aorta. Type III endoleak will be treated by additional stent graft (junctional) placement. Type II endoleaks are rarely actively treated and will undergo watchful waiting approach because of its tendency to spontaneously thrombose (about 40%). If the type II endoleak persists

beyond 6  months, it should be treated (unless shrinkage of the aneurysm sac is documented) because of sac expansion and risk of rupture. The first treatment of choice for type II endoleaks is embolization. Goal of the treatment is to disrupt the communications between the inflow and outflow vessels, thus occluding the vessels and the communicating channels. The approach for embolization can be either transarterial or direct via translumbar or transabdominal puncture. The embolic agent used varies between the studies: most commonly used are coils and liquid embolic agents such as n-butyl cyanoacrylate (NBCA) and ethylene-vinyl-alcohol copolymer (EVOH, Onyx®), and often a combination of these materials is needed. Gelfoam slurry or thrombins are less frequently used in addition to coils. Table 83.1 summarized treatment strategy of different types of endoleaks after primary thoracic endovascular aortic repair (TEVAR).

83.2.2 Aortoesophageal (AEF) and Aortobronchial Fistulae (ABF) Secondary aortoesophageal (AEF) and aortobronchial (ABF) fistulae after primary TEVAR are uncommon and highly fatal conditions with very poor outcome (30-day mortality 33–100%) [3, 14, 15]. Recent large reports suggest that the incidence of AEF or ABF post TEVAR ranges between 1.5 and 1.9% [5, 14, 15]. Data from the European Registry of Endovascular Aortic Repair Complications (EuREC) showed that the incidence of ABF post-TEVAR is 0.6% [16] and AEF

M. Nozdrzykowski et al.

1190

is 1.5% [17]. However, the incidence of secondary AEF and ABF is most likely under-reported due to patients that are lost during follow-up and might even increase in the future, since TEVAR is nowadays not exclusively used in aortic emergencies or elderly high-risk patients. Regarding the mechanisms underlying AEF formation, it has been speculated that the fistula may arise secondary to the development of pseudoaneurysm, chronic endoleaks leading to erosion into the adjacent oesophagus and penetration of the stent graft through the aortic wall into the oesophagus [14, 15]. Czerny et al. [17] hypothesized that AEF development, a most likely stepwise process, may be associated with the need for an emergency procedure and the presence of mediastinal haematoma prior to TEVAR. In addition, potential ischemic necrosis of the oesophageal wall due to stent coverage of the arteries in the mid-oesophageal segment may be involved [6, 14]. The clinical presentation of AEF/ABF is generally referred to as Chiari’s triad, consisting of mid-thoracic pain, sentinel arterial haemorrhage and ultimately fatal exsanguination after an unpredictable symptom-free interval. Pipinos and Reddy [18] reviewed 500 cases of AEF where 59% of patients had mid-thoracic pain, 45% experienced dysphagia, 65% had herald bleeding and 45% showed Chiari’s triad. In cases of ABF the first, and often the only, symptom is haemoptysis.

83.2.2.1 Management of AEF/ABF Except in patients presenting massive exsanguinating haemorrhage who require immediate emergency surgery, a number of diagnostic tools (CT, oesophagoscopy, oesophageal contrast studies, bronchoscopy, FDG-PET, blood culture, white blood cell scan) can be used to confirm the fistula and plan treatment. CT scan is generally the primary diagnostic step. In the CT scan are present indirect signs of fistulisation, such as air bubbles into the thrombus, periaortic fluid and air collection, oesophageal or bronchial wall thickening and lung consolidation [3, 19]. Endoscopy is the most sensitive and specific method for the diagnosis of AEF/ABF but often requires seda-

tion and entails the risk of dislocation of the clots during the endoscopy, which can cause fatal bleeding (Fig. 83.1). Conservative medical treatment results in no late survival [20]. Control or prevention of fatal bleeding is mandatory, and definitive aortic treatment should be accompanied by methods to stop continuous contamination through the fistula to avoid aortic prosthesis infection. Surgery for AEF/ABF requires individual planning and should be performed promptly and in a radical fashion. Three main issues must be fulfilled: (1) aortic repair to prevent exsanguination, (2) radical debridement of the aortic bed and aortic reconstruction and (3) oesophageal or bronchial reconstruction. Besides radical treatment modalities, such as in situ reconstructions or extraanatomic bypasses, temporary solutions such as drainage or endovascular procedures will be considered.

a

b

Fig. 83.1  Diagnosis of aortoesophageal fistulae (black arrows) by endoscopy (a). Computed tomography (b)

83  Secondary Procedures After Primary Thoracic Endovascular Aortic Repair (TEVAR): Pathologies…

83.2.2.2 In Situ Reconstruction The in situ reconstruction within graft replacement and oesophagectomy is the standard treatment of choice for AEF. Options for using graft are Dacron grafts [21] or cryopreserved aortic allografts [22, 23]. The use of cryopreserved aortic homografts relies on the fact that they are more resistant against infection [24]. To prevent reinfection rifampicin-soaked grafts and omental flaps to cover the replaced grafts are used unusually as effective surgical options. The omental tissue due to its high vascularity and neovascularization potential is as well more resistant against infection.

1191

erative changes. Before and after decontamination by a low concentrated antibiotic cocktail, microbiologic tests are performed of various tissue samples. The homografts are then transferred to an ice-cold cryoprotective solution, frozen in liquid nitrogen vapour to −100  °C and finally stored in the vapour phase of liquid nitrogen at −180  °C.  Only great vessels with a warm ischaemia time less than 6 h are collected. The grafts are thawed and washed immediately before implantation. The use of cryopreserved arterial homografts in the abdominal aortic surgery shows promising outcomes with lower mortalities of 5–9%, 100% limb salvage and 2% pseudoaneurysmal degeneration [25, 26]. Surgical Technique Also in the thoracic aortic surgery, clinical Usually surgical access is achieved via a left-­ studies [24, 27] and studies in animal models sited posterolateral thoracotomy. In cases with have shown that cryopreserved allografts are proximal location of the stent graft (in the aortic resistant against infection; therefore, better cliniarch), a median sternotomy with left anterolateral cal results are expected compared with regular extension into the intercostal space must be cho- fabric grafts. Cryopreserved arterial homografts sen. Arterial cannulation for left heart bypass is allow safe in situ reconstruction, decrease early performed via the femoral artery. Retrograde dis- and midterm mortality and reduce antibiotic tal aortic perfusion (about 3  L/min) should be requirements [28]. utilized for adequate visceral and spinal cord proThe use of cryopreserved grafts is more often tection during the entire procedure. Full cardio- preferred in cases with virulent organisms or frank pulmonary bypass is necessary in cases requiring purulence, such as abscess. The use of allografts surgery of the aortic arch. Then additionally can- entails some limitations, in particular due to the nulation of the axillary artery has to be under- risk of early rupture and long-term durability in taken to allow for selective cerebral perfusion terms of aneurysmal formation. Additionally the (SCP). Moderate hypothermia (30–33 °C) is rec- aortic homografts are not always immediately ommended because surgical excision of the available, especially in emergency cases. Czerny endograft and reconstruction can be extremely et al. [29] reported the results of using self-made time-consuming. Also perioperative use of cere- xenopericardial (bovine pericardium) to treat prosbrospinal fluid (CSF) drainage as an additional thetic graft infection or endovascular graft infecmeasure to minimize the risk of paraplegia is rec- tion in 15 patients. They concluded that treating ommended. Reimplantation of the intercostal postoperative graft infections or performing endoarteries is rarely performed [3–6]. vascular treatment of thoracic, thoracoabdominal and abdominal aortic diseases by complete Aortic Homografts removal of the infected prosthetic material and The ascending aorta, the aortic arch and the entire extensive debridement followed by orthotopic vasdescending aorta are removed in a sterile fashion cular reconstruction with intraoperatively prepared from heart-beating organ donors who fulfil heart xenopericardial tube grafts as neoaortic segments valve selection criteria for cryopreserved heart provides excellent results with regard to durability valve homografts. After harvesting, the great ves- and freedom from reinfection and reoperation. sels are immediately stored at 4 °C in an ice-cold They also mentioned that this new concept, bovine solution and are further prepared, measured and pericardial tube grafts, may be superior to cryoprethoroughly examined by angioscopy for degen- served homografts because the likelihood of calci-

M. Nozdrzykowski et al.

1192

fication seems to be less important and that another advantage of customized xenopericardial tissue is its availability, which turns out to be a problem with Fig. 83.2. Using of antibiotic-impregnated prosthetic grafts is another alternative to homograft. The efficacy of rifampicin-impregnated polyester prostheses for in situ replacement of infected aortic stent grafts has been demonstrated both experimentally and in clinical trials. However, their efficacy depends on the bacterium responsible for the infection. Most recently, Okita et  al. [30] reported their results for open surgery of primary and secondary AEF with a remarkable low hospital mortality rate of 26.7%. Their surgical strategy comprised simultaneous resection of the aorta and the oesophagus followed by in situ reconstruction of the descending aorta using a rifampicin-soaked Dacron graft with additional coverage of an omental (or intercostal muscle) flap [30].

a

Also using of “silver acetate-coated” polyester Dacron grafts seems to be beneficial in cases of AEF or ABF with concomitant infection. Its antimicrobial efficacy is based on the action of silver at the different levels in the bacterial defence mechanism. Silver is located in the collagen layer of the graft as well as in the fabric of the graft itself. It allows substantial release immediately after implantation and sustained release up to 30  days after implantation.

83.2.2.3 Extraanatomic Bypass The extraanatomic bypass, which has been reported primary by Yonago etc all. in 1969, can be considered as an alternative surgical strategy to manage primary or secondary AEF/ABF [31], in high-risk patients, especially in cases of infected endovascular stent grafts. For more information, see Chap. 78.

b

c

Fig. 83.2 (a) “Self-made” pericardial tube grafts from bovine pericardial patches, (b) lateral view of “selfmade” pericardial tube grafts and (c) replacement of the

descending aorta with “self-made” neoaorta of two bovine pericardial patches

83  Secondary Procedures After Primary Thoracic Endovascular Aortic Repair (TEVAR): Pathologies…

83.2.2.4 Endovascular Treatment The widely accepted open surgery for AEF/ABF is associated with a high mortality and morbidity rate due to the poor clinical condition of patients at the time of surgery. Therefore, less invasive concepts to reduce perioperative mortality have been evaluated with special attention on thoracic endovascular aortic repair (TEVAR). Re-TEVAR as a treatment option for secondary AEF/ABF seems very questionable since the infected prosthesis remains in place and debridement of infected tissue cannot be performed. Moreover, life-long antibiotic therapy would be necessary in these high-risk patients. Many case reports have proposed a variety of combination of TEVAR with surgical aortic repair, oesophageal stent grafting, tracheobronchial or oesophageal reconstruction, mediastinal drainage or even endoscopic use of fibrin glue at the level of the fistula [32, 33]. However long-term efficacy of these treatment options is undefined. Recent clinical data indicate that re-TEVAR may be performed in cases of ABF with acceptable early outcome [14, 34]. However, only stent graft removal with aortic replacement and bronchial repair, e.g. flap coverage or lobectomy, represents the only curative treatment in these patients and presented favourable 1-year survival (62.5%) in this cohort [3]. On the other hand, re-TEVAR for AEF has been shown to result in an extremely poor overall outcome and therefore should not be considered as a viable treatment option [34, 35]. Emergency TEVAR for AEF should therefore only be used as a ‘bridge-to-surgery’ in haemodynamically unstable patients to obtain immediate control of aortic bleeding and allow for transportation to an aortic centre [4, 36]. Definitive treatment should be done as early as the patient’s physical condition permits after stent graft insertion. Infection control is a key issue for long-term survival. As for the time interval from TEVAR to open repair, it will be proposed that the optimal timing is 90 mmHg). Additionally, highly normal serum haemoglobin as well adequate oxygenation is important to prevent spinal cord injury.

83.3 Conclusions TEVAR is a promising option for different aortic pathologies and offers superior outcome when compared to open surgical repair, relating to hospital mortality and major neurological events. Despite potential benefits of TEVAR, the very liberal use of this approach should be avoided, considering potential early and late complications, including retrograde type A aortic dissection, endoleaks, aortoesophageal/aortobronchial fistulae (AEF/ABF), stent graft infection or malperfusion syndrome. Many of these complications could be avoided by consequent exclusion of patients with unsuitable anatomies. In cases of endoleak and malperfusion syndrome, endovascular approach is preferred. For other complications, such as including retrograde type A aortic dissection, AEF/ABF and stent graft infection, only the conversion to open aortic surgery is the curative treatment option. If serious complications occur, the multidisciplinary team approach in specialized aortic centres providing the full range of diagnostic and treatment options is strongly recommended.

1197

Table 83.2  Treatment strategy of complications after primary thoracic endovascular aortic repair (TEVAR) (summary) Type of complication AEF ABF Stent graft infection rAAD Stent graft collapse Malperfusion

Management strategy Endovascular Open surgery Not indicated Primary indicateda–d Infrequently Primary indicatede indicateda,c,d Not indicated Primary indicateda,b,d Not indicated Primary indicatedf Primary Secondary indicatedg indicatedh Indicatedi Primary indicatedj

AEF aortoesophageal fistulae, ABF aortobronchial fistulae, rAAD retrograde type A aortic dissection a In situ reconstruction: Dacron grafts; cryopreserved aortic allografts, rifampicin-soaked grafts, “silver acetate-­ coated” polyester Dacron grafts, “self-made” xenopericardial tube grafts b Extraanatomic bypass c Oesophageal or bronchial repair d Concomitant procedures: use of pedicled omental flaps, perigraft catheters for antibiotic administration e No signs of stent graft infection f Reverse frozen elephant trunk (infrequently ascending hemiarch or total arch replacement) g Implantation of bare stent; re-ballooning not indicated h If the primary endovascularly treatment unsuitable or failed i Alternative to surgery: use of fenestrated and branched grafts; use of “chimney technique” j Revascularization of supra-aortic branches

Table 83.2 summarized treatment strategy of different complications after primary thoracic endovascular aortic repair (TEVAR)

1198

References

M. Nozdrzykowski et al.

15. Eggebrecht H, Mehta RH, Dechene A, et  al. Aortoesophageal fistula after thoracic aortic stent-­ graft placement: a rare but catastrophic complication 1. Cambria RP, Crawford RS, Cho JS, et  al. A multiof a novel emerging technique. JACC Cardiovasc center clinical trial of endovascular stent graft repair Interv. 2009;2:570–6. of acute catastrophes of the descending thoracic aorta. 16. Czerny M, Reser D, Eggebrecht H, et al. Aorto-­bronchial J Vasc Surg. 2009;50:1255–64. and aorto-pulmonary fistulation after thoracic endo 2. Leshnower BG, Szeto WY, Pochettino A, et  al. vascular aortic repair: an analysis from the European Thoracic endografting reduces morbidity and remodRegistry of Endovascular Aortic Repair Complications. els the thoracic aorta in DeBakey III aneurysms. Ann Eur J Cardiothorac Surg. 2015;48:252–7. Thorac Surg. 2013;95:914–21. 17. Czerny M, Eggebrecht H, Sodeck G, et al. New insights 3. Luehr M, Etz CD, Nozdrzykowski M, et al. Emergency regarding the incidence, presentation and treatment open surgery for aorto-oesophageal and aorto-bronoptions of aorto-oesophageal fistulation after thoracic chial fistulae after thoracic endovascular aortic repair: endovascular aortic repair: the European Registry of a single-centre experiencedagger. Eur J Cardiothorac Endovascular Aortic Repair Complications. Eur J Surg. 2015;47:374–82. Cardiothorac Surg. 2014;45:452–7. 4. Nozdrzykowski M, Etz CD, Luehr M, et al. Optimal 18. Pipinos II, Reddy DJ. Secondary aortoesophageal fistreatment for patients with chronic Stanford type B tulae. Ann Vasc Surg. 1999;13:649–52. aortic dissection: endovascularly, surgically or both? 19. Bergqvist D, Bjorck M. Secondary arterioenteric fisEur J Cardiothorac Surg. 2013;44:165–74. tulation—a systematic literature analysis. Eur J Vasc 5. Nozdrzykowski M, Luehr M, Garbade J, et  al. Endovasc Surg. 2009;37:31–42. Outcomes of secondary procedures after primary 20. Hollander JE, Quick G.  Aortoesophageal fistula: a thoracic endovascular aortic repairdagger. Eur J comprehensive review of the literature. Am J Med. Cardiothorac Surg. 2016;49:770–7. 1991;91:279–87. 6. Girdauskas E, Falk V, Kuntze T, et al. Secondary sur 21. Reardon MJ, Brewer RJ, LeMaire SA, et al. Surgical gical procedures after endovascular stent grafting of management of primary aortoesophageal fistula the thoracic aorta: successful approaches to a chalsecondary to thoracic aneurysm. Ann Thorac Surg. lenging clinical problem. J Thorac Cardiovasc Surg. 2000;69:967–70. 2008;136:1289–94. 22. Vogt PR, Brunner-LaRocca HP, Lachat M, et  al. 7. Scali ST, Feezor RJ, Chang CK, et al. Efficacy of thoTechnical details with the use of cryopreserved racic endovascular stent repair for chronic type B aorarterial allografts for aortic infection: influence tic dissection with aneurysmal degeneration. J Vasc on early and midterm mortality. J Vasc Surg. Surg. 2013;58:10–7. 2002;35:80–6. 8. Roselli EE, Abdel-Halim M, Johnston DR, et al. Open 23. Pirard L, Creemers E, Van Damme H, et  al. In situ aortic repair after prior thoracic endovascular aortic aortic allograft insertion to repair a primary aortorepair. Ann Thorac Surg. 2014;97:750–6. esophageal fistula due to thoracic aortic aneurysm. J 9. Patterson B, Holt P, Nienaber C, et al. Aortic patholVasc Surg. 2005;42:1213–7. ogy determines midterm outcome after endovascular 24. Bisdas T, Pichlmaier MA, Wilhlemi M, et al. Use of repair of the thoracic aorta: report from the Medtronic cryopreserved arterial homografts for the treatment Thoracic Endovascular Registry (MOTHER) dataof stent infections and pseudoaneurysms: regarding base. Circulation. 2013;127:24–32. “renal artery stent infection and pseudoaneurysm 10. Fattori R, Nienaber CA, Rousseau H, et  al. Results management”. Ann Vasc Surg. 2010;24:300. of endovascular repair of the thoracic aorta with 25. Setacci C, De Donato G, Setacci F, et al. Management the talent thoracic stent graft: the Talent Thoracic of abdominal endograft infection. J Cardiovasc Surg. Retrospective Registry. J Thorac Cardiovasc Surg. 2010;51:33–41. 2006;132:332–9. 26. Verhelst R, Lacroix V, Vraux H, et  al. Use of cryo 11. Alsac JM, Khantalin I, Julia P, et al. The significance preserved arterial homografts for management of of endoleaks in thoracic endovascular aneurysm infected prosthetic grafts: a multicentric study. Ann repair. Ann Vasc Surg. 2011;25:345–51. Vasc Surg. 2000;14:602–7. 12. Parmer SS, Carpenter JP, Stavropoulos SW, et  al. 27. Saito A, Motomura N, Hattori O, et  al. Outcome of Endoleaks after endovascular repair of thoracic aortic surgical repair of aorto-eosophageal fistulas with aneurysms. J Vasc Surg. 2006;44:447–52. cryopreserved aortic allografts. Interact Cardiovasc 13. Bavaria JE, Appoo JJ, Makaroun MS, et  al. Thorac Surg. 2012;14:532–7. Endovascular stent grafting versus open surgical 28. Vogt PR, Brunner-La Rocca HP, Carrel T, et  al. repair of descending thoracic aortic aneurysms in Cryopreserved arterial allografts in the treatment of low-risk patients: a multicenter comparative trial. J major vascular infection: a comparison with convenThorac Cardiovasc Surg. 2007;133:369–77. tional surgical techniques. J Thorac Cardiovasc Surg. 14. Chiesa R, Melissano G, Marone EM, et  al. Aorto-­ 1998;116:965–72. oesophageal and aortobronchial fistulae following 29. Czerny M, von Allmen R, Opfermann P, et  al. Self-­ thoracic endovascular aortic repair: a national survey. made pericardial tube graft: a new surgical concept Eur J Vasc Endovasc Surg. 2010;39:273–9.

83  Secondary Procedures After Primary Thoracic Endovascular Aortic Repair (TEVAR): Pathologies… for treatment of graft infections after thoracic and abdominal aortic procedures. Ann Thorac Surg. 2011;92:1657–62. 30. Okita Y, Yamanaka K, Okada K, et  al. Strategies for the treatment of aorto-oesophageal fistula. Eur J Cardiothorac Surg. 2014;46:894–900. 31. Yonago RH, Iben AB, Mark JB. Aortic bypass in the management of aortoesophageal fistula. Ann Thorac Surg. 1969;7:235–7. 32. Van Doorn RC, Reekers J, de Mol BA, et  al. Aortoesophageal fistula secondary to mycotic thoracic aortic aneurysm: endovascular repair and transhiatal esophagectomy. J Endovasc Ther. 2002;9:212–7. 33. Civilini E, Bertoglio L, Melissano G, Chiesa R.  Aortic and esophageal endografting for secondary aortoenteric fistula. Eur J Vasc Endovasc Surg. 2008;36:297–9. 34. Jonker FH, Heijmen R, Trimarchi S, et al. Acute management of aortobronchial and aortoesophageal fistulas using thoracic endovascular aortic repair. J Vasc Surg. 2009;50:999–1004. 35. Dorweiler B, Weigang E, Duenschede F, et  al. Strategies for endovascular aortic repair in aortobronchial and aortoesophageal fistulas. Thorac Cardiovasc Surg. 2013;61:575–80. 36. Munakata H, Yamanaka K, Okada K, Okita Y.  Successful surgical treatment of aortoesophageal fistula after emergency thoracic endovascular aortic repair: aggressive debridement including esophageal resection and extended aortic replacement. J Thorac Cardiovasc Surg. 2013;146:235–7. 37. von Segesser LK, Tkebuchava T, Niederhauser U, et  al. Aortobronchial and aortoesophageal fistulae as risk factors in surgery of descending thoracic aortic aneurysms. Eur J Cardiothorac Surg. 1997;12:195–201. 38. Kubota S, Shiiya N, Shingu Y, et al. Surgical strategy for aortoesophageal fistula in the endovascular era. Gen Thorac Cardiovasc Surg. 2013;61:560–4.

1199

39. Murphy EH, Szeto WY, Herdrich BJ, et al. The management of endograft infections following endovascular thoracic and abdominal aneurysm repair. J Vasc Surg. 2013;58:1179–85. 40. Cernohorsky P, Reijnen MM, Tielliu IF, et al. The relevance of aortic endograft prosthetic infection. J Vasc Surg. 2011;54:327–33. 41. Idrees J, Arafat A, Johnston DR, et  al. Repair of retrograde ascending dissection after descending stent grafting. J Thorac Cardiovasc Surg. 2014;147:151–4. 42. Grabenwoger M, Alfonso F, Bachet J, et al. Thoracic Endovascular Aortic Repair (TEVAR) for the treatment of aortic diseases: a position statement from the European Association for Cardio-Thoracic Surgery (EACTS) and the European Society of Cardiology (ESC), in collaboration with the European Association of Percutaneous Cardiovascular Interventions (EAPCI). Eur Heart J. 2012;33:1558–63. 43. Eggebrecht H, Thompson M, Rousseau H, et  al. Retrograde ascending aortic dissection during or after thoracic aortic stent graft placement: insight from the European registry on endovascular aortic repair complications. Circulation. 2009;120:S276–81. 44. Jonker FH, Schlosser FJ, Geirsson A, et al. Endograft collapse after thoracic endovascular aortic repair. J Endovasc Ther. 2010;17:725–34. 45. Buth J, Harris PL, Hobo R, et  al. Neurologic complications associated with endovascular repair of thoracic aortic pathology: incidence and risk factors. A study from the European Collaborators on Stent/Graft Techniques for Aortic Aneurysm Repair (EUROSTAR) registry. J Vasc Surg. 2007;46: 1103–10. 46. Noor N, Sadat U, Hayes PD, et al. Management of the left subclavian artery during endovascular repair of the thoracic aorta. J Endovasc Ther. 2008;15: 168–76.

Stenting of the Descending Aorta

84

A. D. Godfrey, N. J. Cheshire, and C. D. Bicknell

84.1 Introduction Over the past few years, there has been a significant rise in the number of patients treated for thoracic aneurysm [1, 2]. This is presumably largely secondary to improvements in diagnostic imaging and an appreciation by referring physicians that a minimally invasive approach is now possible, resulting in the reporting of an increasing number of occult descending thoracic aortic aneurysms (TAA) in an ageing population. Recent data suggests that 25% of patients with abdominal aortic aneurysm also have, or will develop, a TAA [3]. So the number of patients, in whom TEVAR could be used, may well rise further—albeit in a high-risk group. Since Miller and Dake implanted the first thoracic aortic stent graft in 1992 [4], major technical improvements have been accomplished. Current generation devices are smaller and more conformable, with the ability to deploy more safely and reproducibly. The obvious reduction in perioperative morbidity and mortality from stenting spawned a rapid clinical adoption of thoracic endovascular aortic repair (TEVAR). As a result, the referral pattern and management of thoracic aortic disease have already started to move from its tradiA. D. Godfrey · N. J. Cheshire · C. D. Bicknell (*) Department of Surgery and Cancer, Imperial College London, London, UK e-mail: [email protected]; [email protected]

tional base in cardiac surgery to clinicians with less experience in open thoracic procedures. The indication for TEVAR is varied. The commonest indication at present is for degenerative descending thoracic aortic aneurysm (dTAA). The population incidence and size distribution of dTAA remain unknown, and evidence describing at what size the rupture risk rises is weak [4]. Indications for TEVAR in dTAA therefore generally lack consensus but currently largely shadow criteria for open surgical repair. Repair is suggested for thoracic aneurysms of 5.5–6  cm or greater in current guidelines. Acute type B dissection is common and rising in incidence. In those where the acute stage is complicated (visceral, limb, spinal malperfusion, rupture, and persisting pain or hypertension), the results from the analysis of the International Registry of Aortic Dissection (IRAD) database and meta-analyses suggest that the mortality rate is significantly reduced with treatment by TEVAR, rather than open surgery [5, 6]. Uncomplicated acute dissection (type B) is still largely managed with a nonoperative approach with long-term aggressive BP control and surveillance. With aneurysmal change treatment may become necessary but is often more complex than the acute stage requiring more extensive surgery. A number of other pathologies, both acute (such as other acute aortic syndromes) and elective, may be treated using endovascular technologies. The table below (Table 84.1) outlines a range

© Springer-Verlag GmbH Austria, part of Springer Nature 2019 O. H. Stanger et al. (eds.), Surgical Management of Aortic Pathology, https://doi.org/10.1007/978-3-7091-4874-7_84

1201

A. D. Godfrey et al.

1202 Table 84.1  Outline of indications for TEVAR Emergency/urgent dTAA rupture Complicated acute type B dissection Penetrating aortic ulcer pain/rupture Mycotic aneurysm Traumatic transection Aortic fistulae—bronchial, oesophageal

Elective Aneurysm (because of increased risk of rupture, rapid increase in size or symptomatic) Late dilatation in type B dissection Penetrating aortic ulcer Pseudoaneurysm formation (post-coarctation repair)

of pathologies, which are currently considered treatable with use of TEVAR—however, a number of controversies still exist. This chapter describes the planning, placement and post-operative treatment of patients undergoing TEVAR with a focus on key issues related to short- and long-term outcomes.

84.2 A  ssessment and Preoperative Optimisation Preoperative assessment is performed before thoracic stent grafting. The assessment is vital in determining whether the benefits of surgery outweigh the risks. In addition, it allows a unique opportunity to optimise the cardiorespiratory function of the patient before surgery. The 5-year survival in patients who undergo TEVAR is between 38 and 53% [7, 8]. The MOTHER registry [9] identified in a large cohort of registry patients that non-aortic mortality was high in the midterm for patients with dTAA, such that the management of modifiable risk factors would appear vital to maximise patient outcomes over the long term. In our practice and many other units performing high-risk aortic repairs, the preoperative “workup” for aortic intervention patients includes as standard: • Pulmonary function testing—to guide the decision to operate, preoperative optimisation and perioperative analgesia choice • Dynamic/stress echocardiography—to guide the decision to operate and those who require optimisation prior to intervention, either by maximising drug therapy or coronary revascularisation

• Renal function (glomerular filtration rate and creatinine clearance) and radionuclide renograms—to guide the decision to operate and to plan in fenestrated or branched aortic endograft cases As always, the results of these assessments should be viewed in combination with imaging at the planning multidisciplinary team meeting and, if required, patients referred on for speciality review prior to being considered for intervention.

84.2.1 Imaging Cross-sectional imaging should extend from the supra-aortic vessels to the common femoral arteries. Detailed thin-slice imaging is mandatory. Multiplanar reconstructions are needed for the best assessment of the pathology, landing zones and access options—particularly in the presence of aortic tortuosity where aortic diameter can be overestimated, sometimes by several centimetres [10]. High-resolution computed tomographic (CT) scanning is the most widely accepted investigative tool for preoperative planning and TEVAR preparation. Other cross-sectional modalities may be used, however, and new MR imaging techniques are developing rapidly in cardiovascular care. CT imaging has been demonstrated to be easily interpreted across the MDT and is readily available in most centres and relatively cheap—enabling transparent planning discussions. A variety of additional planning-specific software packages are commercially available to plan graft selection and deployment strategy.

84  Stenting of the Descending Aorta

1203

84.2.2 Preoperative Planning Planning treatment of thoracic aneurysms must be thorough and include a number of considerations. Although many focus on the technical aspects of graft deployment, most of the errors made in TEVAR are secondary to failures in the planning stage. In the longer term, the secure sealing of the endograft and aneurysm exclusion is vital to reduce the risk of long-term rupture and need for intervention. Consideration may be given to aspects of graft landing zones, access, extraanatomical hybrid revascularisation of arch branches and graft selection.

84.2.3 Selection of Landing Zones The landing zone of the endograft determines whether there is a seal at the proximal and distal end of the graft and hence whether the aneurysm is excluded from the circulation. The quality of the landing zone determines the risk of long-term graft migration, subsequent endoleak and late failure. When assessing suitability of proximal landing zones, criteria such as neck length, angulation and aortic diameter are key and should ideally be within respective instructions for use (IFU) of the selected TEVAR device. In our experience, patients with disease limited to the descending aorta, the minimum healthy landing zones—both proximally and distally—with parallel walls and minimal adherent thrombus, will usually suffice for a seal with commonly available stent graft systems and rarely migrate in the presence of active fixation. It is important, however, that the segment does not contain significant angulation if a short or a diseased landing zone is to be used. Steep angles, adherent mural thrombus or both require greater lengths as a sufficient sealing zone. This is particularly relevant at the proximal sealing zone where highly angulated “gothic” arch shapes are common with large aneurysms. The atheromatous disease burden of the aortic arch should be considered for suitability of

1 0

2 3

4

Fig. 84.1  Modified anatomical map of arch for stent graft planning, showing the landing zones for TEVAR

l­anding zone and predicting stroke risk. This leads to cerebral embolisation, which may increase the rates of cerebrovascular event (stroke) and reduction in psychomotor ability post-procedure. To assist in planning, an anatomical map modified by Balm et al. [11] divides the arch into five zones (Fig. 84.1). In a review of 400 consecutive TEVAR cases, Lee et al. [12] identified the distribution of proximal landing zones in their experience as 41% within zone 2 and a further 49% landing almost equally in zone 3 or 4. With the goal of extending the range of patients treated with endovascular techniques, adjunct surgical techniques have been developed. Extraanatomical arterial debranching procedures create a proximal landing zone for stent grafts in the aortic arch, at the level of the left subclavian artery (in zone 2) with carotid subclavian bypass and the left common carotid artery (zone 1) after carotid-carotid bypass and in zone 0 after neoinnominate reconstruction. Recent technology advances have enabled a wholly endovascular treatment of aneurysms which require proximal stent placement in the arch—using scalloped or branched grafts, in situ fenestration techniques or custom-made fenestrated devices [13, 14]. The use of such devices should, at present, be restricted to research and development programmes and results strictly audited.

A. D. Godfrey et al.

1204

84.2.4 Determining Best Access Route for Stent Graft Deployment For the delivery of the stent graft from a peripheral vessel, the access route must be suitable. In general the common femoral artery (CFA) is chosen as the most easily accessible vessel. In the assessment of access vessels, particular attention should be paid to the diameter of the external iliac arteries, where healthy, non-­ calcified vessels can usually accommodate a delivery system slightly larger in diameter than the vessel size. Calcified highly tortuous vessels will only permit device passage if their native diameter is larger than the device delivery system. Great care should always be exercised when planning TEVAR in female patients with small diseased iliac arteries, as this group of patients may easily sustain injury through traumatic insertion or withdrawal of stent graft systems. For those with long strictures in the iliac access vessels, consideration should be given to the formation of an iliac conduit, through an extraperitoneal approach. This is relatively easy to perform, and while it may minimally increase post-operative morbidity rates and length of stay, the rates of iliac injury are reduced significantly [15]. We prefer end-to-side anastomosis of a 10 mm Dacron graft onto the common iliac artery, which can be used for access. At the case end, we oversew a short stump of the conduit unless there are concerns regarding infection. In cases where there is sufficient concern that re-­ intervention will be required, it can be useful to anastomose the distal end to the common femoral artery (as an iliofemoral bypass), which can later be used for access through the groin. In our practice, however, this is rarely necessary. Focal iliac disease can be considered for preoperative angioplasty; this is most valuable in the common iliac artery. The accepted process is to balloon dilate the vessel first, followed by stent graft passage. If a stent is required, this should be placed at the end of the procedure. There are a number of advanced endovascular techniques available, which may be used to overcome heavily diseased and calcified EIAs. These include multiple sites of angioplasty,

either directly to the vessel or from within a covered stent graft—the “paving and cracking” technique [16]. Highly angulated segments of the aorta can be particularly challenging. Technological advances continue to provide improved stent graft devices by most if not all manufacturers, which ease graft delivery.

84.2.5 Graft Selection A detailed account of the graft sizing and planning strategies for TEVAR are out of the scope of this chapter. However, in general once the proximal and distal landing zones have been determined, the diameter of the aorta at the level of the landing zone is measured in a true cross-sectional plane. The graft is oversized approximately 15–20% in relation to the aortic size in aneurysmal disease to ensure sealing. In dissections, an aggressive approach to oversizing may lead to proximal type A dissection. There are a number of different graft manufacturers. Selection of an appropriate device depends on a number of factors including availability, familiarisation with the device, flexibility and pushability of the aortic delivery device as well as conformability of the proximal end of the graft to reduce the risk of leak.

84.2.6 Extraanatomical Bypass of the Left Subclavian Artery When the aneurysm of the descending thoracic aorta extends proximally up to the left subclavian artery, an adequate landing zone can be achieved by covering the origin of the left subclavian artery. When TEVAR was first introduced, the left subclavian artery was covered by many with impunity, and the need for revascularisation because of upper limb ischaemia was rare. The indications to LSCA revascularisation are considered absolute in patients presenting with specific clinical situations; these include: • Left internal bypass

mammary

artery-coronary

84  Stenting of the Descending Aorta

• Isolated left cerebral hemisphere (i.e., incomplete Circle of Willis) • Functioning left upper extremity arteriovenous dialysis fistula or bypasses A dominant left vertebral artery is considered by some to also be a mandatory case for revascularisation. Many will now revascularise as a matter of routine to reduce the neurological ischaemia risk. There appears to be an increased risk of stroke when the left subclavian artery is covered supported by registry evidence [9]. This is our current practice, but not all units have the same policy—a selective revascularisation approach has been suggested by many based on ischaemic risk stratification, including the extent of coverage, prior aortic surgery or an occluded hypogastric artery [17].

84.3 The TEVAR Procedure There are many techniques used to perform TEVAR [18]; a standard approach is described here. TEVAR is usually performed under general anaesthetic with the patient in the supine position. Access to the circulation is most often achieved via the common femoral artery, which may be percutaneous or facilitated by an open cutdown of the artery. When using a percutaneous access route, a small incision is made, wire access is secured, and the closure device placed before the stent graft is introduced. Through the contralateral groin, access is secured, a sheath is placed and an imaging catheter (usually a pigtail catheter) is introduced to lie proximally to the aneurysm to image during the procedure. The patient is heparinised adequately using the activated clotting time as a guide to heparin dose, and the graft is introduced in its delivery device to the level of the proximal landing zone. An intraarterial digital subtraction angiogram is performed. The level of the landing zone is clearly marked. This often requires adjustment of the angle of the C-arm to reduce parallax error. The graft is then accurately deployed taking care to maximise the seal (landing) zone while avoid covering vessels unintentionally. With grafts that

1205

are deployed at or near the aortic arch (zones 2 and 3), the blood pressure can be lowered substantially by pharmacological means in order to avoid graft malplacement. In those cases where very proximal grafts are placed after debranching of the aortic arch vessels, rapid overdrive pacing may be necessary before deployment. Further stents may be placed to extend the repair to the distal landing zone ensuring an adequate stent-to-­ stent overlap. Once completed the proximal and distal extent of the graft and the stent overlap zones are then ballooned to mould the stent adequately. A check angiogram is then performed to ensure adequate graft placement and to ensure there is no direct leak of contrast into the aneurysm sac (see below for description of endoleak). At this stage, with aneurysm exclusion attention must be paid to spinal cord protection measures (also below). At the end of procedure, the device is removed carefully. For percutaneous procedures, the closure device is deployed to secure the vessel. If an open cutdown has been used to expose the CFA, the device is removed, clamps are placed proximally and distally and the arteriotomy is formally closed.

84.4 Post-operative Management After TEVAR patients, most patients are extubated after the procedure and cared for in a high dependency unit for at least 24 h and often longer where spinal drain protection is used. Assessment for early complication including vascular and neurological lower limb observations is vital to prevent significant sequelae. After discharge from hospital, long-term follow-­up is mandatory for all those who have undergone thoracic stent grafting. Assessment must be made of stent graft position (assessing migration), kinking and endoleak. Routine imaging at this time is most often performed with CT scanning after stent graft placement and then at regular intervals (initially at 1 year). Figure 84.2 shows a pre- and post-operative CT angiogram for one of our more ­routine patients.

1206

A. D. Godfrey et al.

a

b

Fig. 84.2 (a) Preoperative CT angiogram for one of our more standard TAA cases. (b) Post-operative CT angiogram demonstrating stent graft position with no evidence of endoleak and thrombosed aneurysm sac

84.5 Complications 84.5.1 Spinal Cord Ischaemia Spinal cord ischaemia is a significant problem with all thoracic aortic repair strategies. The endovascular repair of thoracic aortic pathology has significantly decreased the overall incidence

of complications when compared with open surgery. Nevertheless, the risk of paraplegia remains an important concern, with rates ranging from 2% up to 8% [19]. Blood flow to the spinal cord relies not only on branches of the vertebral, deep cervical, intercostal and lumbar arteries feeding the anterior spinal artery but also on more distant arteries

84  Stenting of the Descending Aorta

such as the hypogastric and subclavian (via the vertebral and anterior spinal network) which also feed into the same network. Specific determinants of spinal cord ischaemia include: Length of coverage of aorta Coverage of LSCA Coverage of internal iliac vessels Hypotension peri- or post-operatively Previous aortic repair For procedures limited to the upper thoracic segments, the risk of spinal cord ischaemia is relatively low but is believed to rise sharply when more than eight segmental arteries are sacrificed [20]. Spinal cord perfusion is directly related to mean arterial BP and CSF pressure in the same way as cerebral perfusion pressure after head injury. Most units advise a raised mean arterial pressure after stenting (>85 mmHg). Prophylactic placement of a cerebrospinal fluid (CSF) drainage catheter is recommended in selected TEVAR cases, which have a high risk of spinal cord ischaemia, as described above. Avoidance of anaemia and hypoxia will also improve oxygen delivery to the spinal cord. In those with multiple aneurysms (typically thoracic and abdominal segments), staging the treatments is also advantageous in terms of spinal cord risk. CSF drainage should aim to keep the spinal cord pressure less than 15 mmHg. Units performing aortic interventions—and the concurrent need for CSF drainage—should utilise strict evidence-­based protocols to avoid errors which may have disastrous consequences.

84.5.2 Stroke There is a growing realisation that the incidence of cerebrovascular events occurs in up to 20% of TEVAR cases [19]. Predictors of occurrence include a more proximal landing zone, arch atheroma burden and LSCA occlusion. The psychomotor impact of such events is expectedly more severe in advancing age or in patients with a prior history of cerebrovascular events.

1207

Analysis of large registry-based datasets has allowed risk stratification by LSCA management [9]. This identified a stroke rate of 9% in those where the LSCA was covered, 4.9% when revascularised and lowest overall when the artery was uncovered (2.2%). Coverage of the left subclavian artery without revascularisation would arguably appear to be the single most important, modifiable predictor of post-TEVAR stroke and is potentially avoidable with appropriate revascularisation procedures [1].

84.5.3 Endoleak Failure of aneurysm sealing, with leakage of circulating blood into the aneurysm sac, is termed endoleak. The risk of endoleak is that of continuous pressurisation of the aneurysm sac and further expansion of the aneurysm and rupture. Rupture following apparently successful endovascular repair is well described in the infrarenal aorta [21]. There are currently five types of endoleak described: Type 1 Due to proximal (1a) or distal (1b) failure of stent graft sealing and direct pressurisation of the sac. Type 2 These endoleaks occur as there is leak into the sac from an intercostal vessel with retrograde flow. Type 3 This endoleak type occurs secondary to a leak between stent graft components or because of a tear in the graft fabric and loss of integrity of the graft. Type 4 This is due to porosity of the graft, which was a common feature in early stent graft designs. Type 5 This classification of endoleak is reserved for aneurysms treated by TEVAR which continue to expand, but no demonstrable leak is identified. Type 1 and 3 endoleaks imply that there is direct communication between the aortic blood flow and the aneurysm sac. These are therefore at significant risk of rupture. If either of these

A. D. Godfrey et al.

1208

endoleak types are identified at the time of surgery on check angiograms, all reasonable steps to abolish endoleak should be taken at this time. This may include extending the graft proximally or distally, relining the graft and re-ballooning. These endoleaks may also occur as a result of endograft migration or expansion of the proximal sealing zone diameter. Re-intervention is necessary in all cases to prevent the risk of rupture. Type 2 endoleaks are often present after TEVAR, and many resolve spontaneously when merely observed on follow-up scanning. Current practice is to observe these endoleaks and intervene only where there is continued sac expansion (and a presumption that the aneurysm sac is still pressurised).

After large-calibre access for TEVAR, the clinical team should remain aware of the risk of groin vessel bleeding—overtly or covertly. The latter possibility should always be considered if a patient is “unstable” following TEVAR, at which time urgent CT scanning should be performed. Surgical exploration and repair are usually indicated in such circumstances. Large-calibre systems passing through the iliac system can result in lower limb ischaemia, most usually as a consequence of dissection. Suspicion should arise during removal of the system if there is no adequate inflow or on the final iliac, imaging runs. A bare metal stent can be used in cases of iliac dissection.

84.6 Summary 84.5.4 Access Complications These formed the single most common complication after endovascular aortic repairs when the EUROSTAR database was analysed [22]. When imaging suggests small or complex EIA access, or when intraoperative findings suggest iliac difficulty, it is important to maintain a high level of suspicion of injury. Every surgeon who has been involved in significant numbers of TEVAR procedures has seen the iliac artery, usually as the device is removed at the end of the procedure, bringing a length of the EIA with it and significant bleeding. The key is early detection and prevention of haemorrhage. Before removing the stiff wire—but with the delivery device withdrawn into the very distal EIA—angiography should be performed to look for contrast extravasation. If the EIA has ruptured, the device must not be removed. Instead, control should be obtained through a contralateral approach with inflation of a large-calibre occlusion balloon at the aortic bifurcation. The delivery system may now be removed, and with sufficient endovascular experience, a covered endograft can often be used to repair the disruption. It should be standard practice to have aortic occlusion balloons and covered stent grafts present as emergency equipment stock if performing endovascular interventions.

TEVAR is a significant advance in the treatment of descending thoracic aneurysms reducing the morbidity and short-term mortality. Improvements in the technology and techniques available for TEVAR continue to expand the number of patients that can undergo a minimally invasive approach safely. The technology of TEVAR will continue to expand. Due care must be taken in planning and performing TEVAR to reduce the significant complications that occur and reduce risk of long-term graft failure.

References 1. Olsson C, et  al. Thoracic aortic aneurysm and dissection: increasing prevalence and improved outcomes reported in a nationwide population-based study of more than 14,000 cases from 1987 to 2002. Circulation. 2006;114:2611–8. 2. von Allmen RS, Anjum A, Powell JT.  Incidence of descending aortic pathology and evaluation of the impact of thoracic endovascular aortic repair: a population-­ based study in England and Wales from 1999 to 2010. Eur J Vasc Endovasc Surg. 2013;45:154–9. 3. Hultgren R, et  al. Female and elderly abdominal aortic aneurysm patients more commonly have concurrent thoracic aortic aneurysm. Ann Vasc Surg. 2012;26:918–23. 4. Dake MD, et  al. Transluminal placement of endovascular stent-grafts for the treatment of descending thoracic aortic aneurysms. N Engl J Med. 1994;331:1729–34.

84  Stenting of the Descending Aorta 5. Eggebrecht H, et  al. Endovascular stent-graft placement in aortic dissection: a meta-analysis. Eur Heart J. 2006;27:489–98. 6. Fattori R, et  al. Complicated acute type B dissection: is surgery still the best option? A report from the International Registry of Acute Aortic Dissection. JACC Cardiovasc Interv. 2008;1:395–402. 7. Shah AA, et  al. Results of thoracic endovascular aortic repair 6 years after United States Food and Drug Administration approval. Ann Thorac Surg. 2012;94:1394–9. 8. Goodney PP, et al. Survival after open versus endovascular thoracic aortic aneurysm repair in an observational study of the Medicare population. Circulation. 2011;124:2661–9. 9. Patterson B, et  al. Aortic pathology determines midterm outcome after endovascular repair of the thoracic aorta: report from the Medtronic thoracic Endovascular Registry (MOTHER) database. Circulation. 2013;127:24–32. 10. Rudarakanchana N, et  al. Variation in maximum diameter measurements of descending thoracic aortic aneurysms using unformatted planes versus images corrected to aortic centerline. Eur J Vasc Endovasc Surg. 2014;47:19–26. 11. Balm R, Reekers JA, Jacobs MJ.  Classification of endovascular procedures for treating thoracic aortic aneurysms. In: Jacobs MJ, Branchereau A, editors. Surgical and endovascular treatment of aortic aneurysms. New York: Futura Publishing Company; 2000. p. 19–26. 12. Lee WA, et al. Late outcomes of a single-center experience of 400 consecutive thoracic endovascular aortic repairs. Circulation. 2011;123:2938–45.

1209 13. Alsafi A, et  al. Endovascular treatment of tho racic aortic aneurysms with a short proximal landing zone using scalloped endografts. J Vasc Surg. 2014;60:1499–506. 14. Riga CV, et  al. In vitro fenestration of aortic stent-­ grafts: implications of puncture methods for in situ fenestration durability. J Endovasc Ther. 2013; 20:536–43. 15. Abu-Ghaida AM, et al. Broadening the applicability of endovascular aneurysm repair: the use of iliac conduits. J Vasc Surg. 2002;36:111–7. 16. Kpodonu J, et  al. “Cracking and paving”: a novel technique to deliver a thoracic endograft despite ilio-­ femoral occlusive disease. J Card Surg. 2009; 24:188–90. 17. Riambau V, et al. Spinal cord protection and related complications in endovascular management of B ­dissection: LSA revascularization and CSF drainage. Ann Cardiothorac Surg. 2014;3:336–8. 18. Cheshire N, Bicknell N.  Thoracic endovascular aortic repair: the basics. J Thorac Cardiovasc Surg. 2013;145:S149–53. 19. Bicknell CD, Powell JT. Thoracic aortic aneurysms. Br J Surg. 2013;100:850–2. 20. Griepp EB, Luozzo GD, Schray D, et  al. The anatomy of the spinal cord collateral circulation. Ann Cardiothorac Surg. 2012;1:350–7. 21. Wyss TR, et al. Rate and predictability of graft rupture after endovascular and open abdominal aortic aneurysm repair: data from the EVAR Trials. Ann Surg. 2010;252:805–12. 22. Harris PL, Buth J. An update on the important findings from the EUROSTAR EVAR registry. Vascular. 2004;12:33–8.

Complications of TEVAR

85

Rana O. Afifi, Ali Azizzadeh, and Anthony L. Estrera

85.1 Introduction Open surgical repair has been the treatment of choice for degenerative thoracic aortic aneurysms since the 1950s. It includes thoracotomy, aortic cross-clamping, and replacement of the affected part with a prosthetic graft. Despite improvement in surgical techniques and perioperative care, mortality and morbidity remain significant [1–4]. Thoracic endovascular aortic repair (TEVAR) was first used in 1987 in Ukraine by Volodos et al. to repair a traumatic thoracic aortic aneurysm. In 1994, Dake et al. reported 13 successful cases of TEVAR, proving that it is a safe and feasible treatment method for selected patients with thoracic aortic aneurysms [5]. Commercial use of TEVAR increased after a few devices were approved by the US Food and Drug Administration (FDA) in 2005. Since that time, TEVAR has been reported to lower thoracic aortic aneurysm repair mortality and morbidity [6–9]. Endovascular repair became the predominant method of treatment for most thoracic aortic pathologies, including aneurysms, traumatic injury, penetrating aortic ulcers, and aortic dissection. R. O. Afifi · A. Azizzadeh · A. L. Estrera (*) Department of Cardiothoracic and Vascular Surgery, McGovern Medical School at The University of Texas Health Science Center at Houston (UTHealth), Houston, TX, USA e-mail: [email protected]; [email protected]; [email protected]; [email protected]

In recent years, multiple publications comparing open repair of thoracic aortic aneurysm to TEVAR have demonstrated favorable shortand midterm results in the TEVAR groups. These have included reduced early mortality and lower paraplegia rates and cardiac and respiratory complications [8, 10–12]. Longterm survival was comparable between groups, with a higher rate of reintervention in patients treated with TEVAR [13, 14].

85.2 Mortality A recent review by Powell et al. [15] summarized the outcome of TEVAR for variant thoracic aortic pathologies. The 30-day mortality following TEVAR was reported between 2.5 and 6.5% [12, 14, 16–21]. The 5-year mortality was related to the indication for TEVAR, ranging from 9.2 to 30% in non-acute aortic dissections [22–24] and 38–53% following repair of descending thoracic aortic aneurysms [14, 18, 20, 21].

85.3 Neurological Complications Neurological complications following TEVAR can be devastating. They include perioperative stroke and spinal cord ischemia, resulting in paraparesis and/or paraplegia.

© Springer-Verlag GmbH Austria, part of Springer Nature 2019 O. H. Stanger et al. (eds.), Surgical Management of Aortic Pathology, https://doi.org/10.1007/978-3-7091-4874-7_85

1211

R. O. Afifi et al.

1212

85.3.1 Stroke Risk of perioperative stroke is similar in open and endovascular repair of the thoracic aorta. Incidence following TEVAR has been reported between 3% and 6% [11, 13, 25]. The exact mechanism of perioperative stroke is still unclear. Many investigators believe that thromboembolism and hypoperfusion are the main two mechanisms contributing to stroke following TEVAR, each independently or in combination with one another [26, 27]. Cerebral embolism has been shown using transcranial Doppler to occur during wire, catheter maneuvers in the aortic arch as well as device deployment [28]. Hypoperfusion may occur following coverage of supraaortic branches by the stent graft. The atheromatous burden in the aortic arch has been reported as a risk factor for stroke in thoracic aortic repair. Gutsche et  al. [27]

6th Rib

reported a 5.8% incidence of stroke in their review of 171 patients who underwent TEVAR.  They identified three risk factors for perioperative stroke, including (a) history of preoperative stroke, (b) CT grade IV atheroma (>5 mm) in the aortic arch or ­proximal descending aorta, and (c) extent A or C coverage (Fig. 85.1). Extension of the endovascular repair into the aortic arch (zone 0–2) (Fig. 85.2) with coverage of supraaortic vessels increases the risk of perioperative stroke [29–32]. Feezor et  al. [29] reported an incidence of 4.6% of perioperative stroke in a series of 196 patients treated with TEVAR. Seven (78%) of these patients had coverage of zones 0–2, while only 2 (22%) had coverage of zones 3–4. All patients with coverage of zone 0 or 1 had an elective arch revascularization performed. Melissano et al. [32] reported a study of 393 patients treated with TEVAR at a single

6th Rib

Fig. 85.1  Classification of descending thoracic aortic aneurysm: (a) distal to left subclavian artery to T6; (b) from T6 to the level of the diaphragm; and (c) distal to left subclavian artery to diaphragm

85  Complications of TEVAR

1213

a

b

c

d

e

f

Fig. 85.2 (a) Distribution of landing zones for TEVAR. Zone 0: repair involving the innominate artery. Zone 1: repair involving the left carotid artery. Zone 2: repair involving the left subclavian artery. Zone 3: repair involving the proximal third of the descending thoracic aorta. Zone 4: involving the distal two thirds of the descending thoracic aorta. (b, c) Placement of a device in

zone 3 or 4 without the need for revascularization. (d) Coverage of the left subclavian artery with a left carotid-­ subclavian bypass. (e) Coverage of the left carotid and subclavian arteries with a carotid-carotid and carotid-­ subclavian bypass. (f) Complete arch coverage with ascending aorta to innominate and left carotid bypass as well as a left carotid-subclavian bypass

institution. In 143 of the cases, the aortic arch was involved (32 zones “0,” 35 zones “1,” and 76 zones “2”). Incidence of stroke between patients treated with TEVAR with or without arch involvement was 2.8% vs. 1.2%, respectively. The stroke rate was 9.4% (P