Nuclear Medicine Textbook: Methodology and Clinical Applications [1st ed. 2019] 978-3-319-95563-6, 978-3-319-95564-3

Building on the traditional concept of nuclear medicine, this textbook presents cutting-edge concepts of hybrid imaging

2,288 336 95MB

English Pages XVI, 1331 [1321] Year 2019

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Nuclear Medicine Textbook: Methodology and Clinical Applications [1st ed. 2019]
 978-3-319-95563-6, 978-3-319-95564-3

Table of contents :
Front Matter ....Pages i-xvi
Front Matter ....Pages 1-1
Fundamentals of Natural and Artificial Radioactivity and Interaction of Ionizing Radiations with Matter (Alberto Del Guerra, Daniele Panetta)....Pages 3-19
Single-Photon-Emitting Radiopharmaceuticals (Federica Orsini, Erinda Puta, Federica Guidoccio, Giuliano Mariani)....Pages 21-56
Positron-Emitting Radiopharmaceuticals (Piero A. Salvadori, Elena Filidei, Assuero Giorgetti)....Pages 57-98
Radiopharmaceuticals for Therapy (Federica Orsini, Federica Guidoccio, Giuliano Mariani)....Pages 99-116
Methods and Instrumentation for Measuring Radioactivity (Maria Evelina Fantacci)....Pages 117-135
General-Purpose Gamma Cameras, Dedicated Gamma Cameras, and Gamma-Probes for Radioguided Surgery (Roberto Pani, Federica Guidoccio, Raffaele Scafè, Pat Zanzonico, Giuliano Mariani)....Pages 137-171
Image Acquisition and Processing with Gamma Cameras Including Integrated SPECT/CT and Dedicated Gamma Cameras (Duccio Volterrani, Federica Guidoccio, Giulia Puccini, Sara Mazzarri)....Pages 173-186
Principles of CT and MR imaging (Christian Bracco, Daniele Regge, Michele Stasi, Michela Gabelloni, Emanuele Neri)....Pages 187-198
PET/CT and PET/MR Tomographs: Image Acquisition and Processing (Nicola Belcari, Ronald Boellaard, Matteo Morrocchi)....Pages 199-217
Essentials of Quantitative Imaging with PET (Adriaan A. Lammertsma)....Pages 219-233
Principles of Radiation Biology and Dosimetry for Nuclear Medicine Procedures (Massimo Salvatori, Amedeo Capotosti, Luca Indovina)....Pages 235-260
Radiation Protection for Patients (Lawrence T. Dauer, Sören Mattsson, Eliseo Vañó, Yoshiharu Yonekura)....Pages 261-272
Radiation Protection for Personnel and the Environment (Antonio C. Traino)....Pages 273-280
Essentials of CT Image Interpretation (Davide Caramella, Matteo Revelli, Alessandro Villa)....Pages 281-316
Essentials of MR Image Interpretation (Davide Caramella, Fabio Chiesa)....Pages 317-350
Image-Guided and Radioguided Surgery (Francesco Giammarile, Sergi Vidal-Sicart, Federica Orsini, Renato A. Valdés Olmos, Giuliano Mariani)....Pages 351-388
Front Matter ....Pages 389-389
Radionuclide Imaging for Non-tumor Diseases of the Brain (Duccio Volterrani, Giampiero Giovacchini, Andrea Ciarmiello)....Pages 391-412
Hybrid Imaging for Tumors of the Brain (Giampiero Giovacchini, Mattia Riondato, Patrizia Lazzeri, Elisa Borsò, Valerio Duce, Rossella Leoncini et al.)....Pages 413-429
Hybrid Imaging of the Head and Neck Region (Alejandro Fernández, Valle Camacho)....Pages 431-447
Radionuclide Imaging of Cardiovascular Disease (Matteo Bauckneht, Flavia Ticconi, Roberta Piva, Riemer H. J. A. Slart, Alberto Nieri, Silvia Morbelli et al.)....Pages 449-497
Radionuclide Imaging of Benign Pulmonary Diseases (Federica Guidoccio, Edoardo Airò, Giuliano Mariani)....Pages 499-521
Hybrid Imaging for Tumours of the Chest (Roberto C. Delgado Bolton, Adriana K. Calapaquí Terán)....Pages 523-542
Hybrid Imaging for Breast Malignancies (Federica Padovano, Giuliano Mariani, Marco Ferdeghini)....Pages 543-570
Hybrid Imaging and Radionuclide Therapy of Musculoskeletal Diseases (Paola Anna Erba, Martina Sollini, Roberta Zanca, Roberto Boni, Lesley Flynt, Elena Lazzeri et al.)....Pages 571-644
Hybrid Imaging of Melanoma and Other Cutaneous Malignancies (Montserrat Estorch)....Pages 645-653
Hybrid Imaging and Radionuclide Therapy in Hemato-oncology (Paola Anna Erba, Martina Sollini, Roberto Boni, Sara Galimberti)....Pages 655-705
Hybrid Imaging and Radionuclide Therapy for Thyroid Disorders (Federica Guidoccio, Gayane Aghakhanyan, Mariano Grosso)....Pages 707-747
Hybrid Imaging in Non-thyroidal Endocrinological Disorders (Duccio Volterrani, Federica Guidoccio, Giuliano Mariani)....Pages 749-765
Hybrid Imaging and Radionuclide Therapy of Neuroendocrine Tumors (Duccio Volterrani, Lisa Bodei, Federica Guidoccio)....Pages 767-784
Radionuclide Imaging of the Nephro-Urinary Tract (Duccio Volterrani, Federica Orsini, Federica Guidoccio)....Pages 785-808
Functional Imaging for Benign Conditions of the Gastrointestinal Tract (Gayane Aghakhanyan, Elisa Fiasconaro, Elisa Tardelli, Mariano Grosso, Italia Paglianiti)....Pages 809-839
Hybrid Imaging for Malignant Conditions of the Gastrointestinal Tract (Joan Duch, Albert Flotats)....Pages 841-857
Radionuclide Therapy for Tumors of the Liver and Biliary Tract (Federica Guidoccio, Giuseppe Boni, Duccio Volterrani, Giuliano Mariani)....Pages 859-879
Hybrid Imaging for Gynecologic Malignancies (Elisa Lodi Rizzini, Elena Tabacchi, Cristina Nanni)....Pages 881-898
Hybrid Imaging for Male Malignancies (Akram Al-Ibraheem, Abdullah S. Al Zreiqat, Serena Chiacchio, Abedallatif A. AlSharif)....Pages 899-924
Nuclear Medicine in Pediatrics (Pietro Zucchetta, Diego De Palma)....Pages 925-949
Hybrid Imaging of the Peripheral Lymphatic System (Paola Anna Erba, Roberto Boni, Martina Sollini, Andrea Marciano, Rossella Di Stefano, Giuliano Mariani)....Pages 951-975
Molecular Guidance for Planning External Beam Radiation Therapy (Federica Orsini, Giovanna Pepe, Arturo Chiti, Giuseppe Roberto D’Agostino, Annibale Versari, Carlo Cavedon et al.)....Pages 977-1006
Front Matter ....Pages 1007-1007
Radiopharmacy/Radiochemistry for Conventional Single-Photon Emitting and Therapeutic Radiopharmaceuticals (Alessandra Boschi, Adriano Duatti)....Pages 1009-1018
Radiopharmacy/Radiochemistry for Positron-Emitting Radiopharmaceuticals, Including Quality Assurance and Process Validation Principles (Piero A. Salvadori)....Pages 1019-1037
Recommendations for Conducting Clinical Trials with Radiopharmaceuticals (Clemens Decristoforo, Serge K. Lyashchenko)....Pages 1039-1050
Standard Operating Procedures for Quality Control of Gamma Cameras (Angela Vaiano)....Pages 1051-1059
Standard Operating Procedures for Quality Control of PET/CT and PET/MR Tomographs (Alessandra Zorz, Alessandro Scaggion)....Pages 1061-1078
Standard Operating Procedures for Quality Control of Instrumentation for Radioguided Surgery (Angela Vaiano)....Pages 1079-1084
Technical Aspects of Dual-Energy X-Ray Absorptiometry and Quantitative Computed Tomography for Assessment of Bone Mineral Density and Body Composition (Pat Zanzonico)....Pages 1085-1098
Current Practical Guidelines for the Most Common Nuclear Medicine Procedures (Irene Marini, Onelio Geatti, H. William Strauss)....Pages 1099-1138
Teaching Cases in Nuclear Medicine: Non-oncological Applications (Venanzio Valenza, Isabella Bruno, Daniela Di Giuda, Germano Perotti, Fabrizio Cocciolillo, Valerio Lanni et al.)....Pages 1139-1198
Teaching Cases in Nuclear Medicine: Oncological Applications (Laura Evangelista, Lucia Setti, Anna Rita Cervino, Gianluigi Ciocia, Lea Cuppari, Riccardo Vicinelli et al.)....Pages 1199-1239
Radiological Anatomy with CT: What the Nuclear Physician Should Know When Reading a PET/CT Scan (Davide Caramella, Matteo Revelli, Alessandro Villa)....Pages 1241-1256
Radiological Anatomy with MR: What the Nuclear Physician Should Know When Reading a PET/MR Scan (Davide Caramella, Fabio Chiesa)....Pages 1257-1278
Interpreting and Reporting the Results of Radionuclide Tests (H. William Strauss, Federica Orsini)....Pages 1279-1296
Back Matter ....Pages 1297-1331

Citation preview

Duccio Volterrani Paola Anna Erba Ignasi Carrió H. William Strauss Giuliano Mariani Editors

Nuclear Medicine Textbook Methodology and Clinical Applications

123

Nuclear Medicine Textbook

Duccio Volterrani  •  Paola Anna Erba Ignasi Carrió  •  H. William Strauss Giuliano Mariani Editors

Nuclear Medicine Textbook Methodology and Clinical Applications

Editors Duccio Volterrani Regional Center of Nuclear Medicine Department of Translational Research and Advanced Technologies in Medicine and Surgery University of Pisa Pisa Italy

Paola Anna Erba Regional Center of Nuclear Medicine Department of Translational Research and Advanced Technologies in Medicine and Surgery University of Pisa Pisa Italy

Ignasi Carrió Hosp. Sant Pau, Nuclear Medicine Department Autonomous University of Barcelona Barcelona Spain

H. William Strauss Molecular Imaging and Therapy Department Cornell University Weill New York, NY USA

Giuliano Mariani Regional Center of Nuclear Medicine Department of Translational Research and Advanced Technologies in Medicine and Surgery University of Pisa Pisa Italy

ISBN 978-3-319-95563-6    ISBN 978-3-319-95564-3 (eBook) https://doi.org/10.1007/978-3-319-95564-3 © Springer Nature Switzerland AG 2019 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Preface

There are few up-to-date books dealing with the Nuclear Medicine specialty as a discipline per se, i.e., intended as complete textbooks providing the rationale for understanding the physical, chemical, biological, and pathophysiological bases of all the applications of nuclear medicine in the routine clinical practice, including diagnosis and therapy. In addition to the expertise in therapy with radionuclides, the modern specialist in nuclear medicine must be a specialist in multimodality imaging. This requirement is the guiding concept throughout this multiauthor textbook (written by expert authors from Europe and the United States), which presents the spectrum of information required to become a practitioner of nuclear medicine and molecular imaging. The text presents the basic science of radiation physics, interaction of radiation with biologic systems, measuring devices for radiation detection and radionuclide imaging, synthesis and biodistribution of radiopharmaceuticals, and approaches to data recording and analysis that comprise the clinical applications of nuclear medicine in diagnosis and therapy. Each of the 51 chapters in this 1300 page text contains numerous graphs, tables, diagrams, and patient illustrations designed to meet the needs of students training for a career as practitioners, technologists, physicists, and/or data analysts. Each chapter provides detailed learning objectives, comprehensive text, and visual material to provide the reader with the information necessary to understand the topic. In addition to the basic science and clinical material, chapters on Computed Tomography and Magnetic Resonance Imaging provide foundation knowledge to employ these important complementary, imaging techniques in hybrid imaging. To supplement the basic science and clinical applications of radionuclide imaging, the authors present current guidelines from international organizations on approved indications to perform specific procedures, recommendations on appropriate techniques to perform and interpret the procedures as well as recommendations from the International Commission on Radiation Protection to perform these procedures with the lowest recommended radiation burden to the patient and the clinical staff. In addition, case presentations in the clinical chapters illustrate the specific application of the radiopharmaceutical procedures. The clinical case presentations are supplemented by additional patient material, presented in Chaps. 47 and 48. These cases allow the reader to combine the clinical history, laboratory findings, radiographic and radionuclide findings to get acquainted with the most typical presentations for diagnosis or therapeutic plan to manage the patient. Developing this text has required the work of many people with different areas of expertise. The authors and editors want to offer special thanks to the numerous individuals from our publisher, Springer Publishing Co., for their work in typesetting, layout, and printing, to bring this project to fruition. Pisa, Italy Pisa, Italy Barcelona, Spain New York, NY Pisa, Italy April 2019

Duccio Volterrani Paola Anna Erba Ignasi Carrió H. William Strauss Giuliano Mariani v

Contents

Part I Basic Science 1 Fundamentals of Natural and Artificial Radioactivity and Interaction of Ionizing Radiations with Matter���������������������������������������������    3 Alberto Del Guerra and Daniele Panetta 2 Single-Photon-Emitting Radiopharmaceuticals�����������������������������������������������������   21 Federica Orsini, Erinda Puta, Federica Guidoccio, and Giuliano Mariani 3 Positron-Emitting Radiopharmaceuticals �������������������������������������������������������������   57 Piero A. Salvadori, Elena Filidei, and Assuero Giorgetti 4 Radiopharmaceuticals for Therapy�������������������������������������������������������������������������   99 Federica Orsini, Federica Guidoccio, and Giuliano Mariani 5 Methods and Instrumentation for Measuring Radioactivity�������������������������������  117 Maria Evelina Fantacci 6 General-Purpose Gamma Cameras, Dedicated Gamma Cameras, and Gamma-Probes for Radioguided Surgery�������������������������������������������������������  137 Roberto Pani, Federica Guidoccio, Raffaele Scafè, Pat Zanzonico, and Giuliano Mariani 7 Image Acquisition and Processing with Gamma Cameras Including Integrated SPECT/CT and Dedicated Gamma Cameras�������������������  173 Duccio Volterrani, Federica Guidoccio, Giulia Puccini, and Sara Mazzarri 8 Principles of CT and MR imaging���������������������������������������������������������������������������  187 Christian Bracco, Daniele Regge, Michele Stasi, Michela Gabelloni, and Emanuele Neri 9 PET/CT and PET/MR Tomographs: Image Acquisition and Processing�����������  199 Nicola Belcari, Ronald Boellaard, and Matteo Morrocchi 10 Essentials of Quantitative Imaging with PET �������������������������������������������������������  219 Adriaan A. Lammertsma 11 Principles of Radiation Biology and Dosimetry for Nuclear Medicine Procedures�������������������������������������������������������������������������������������������������  235 Massimo Salvatori, Amedeo Capotosti, and Luca Indovina 12 Radiation Protection for Patients ���������������������������������������������������������������������������  261 Lawrence T. Dauer, Sören Mattsson, Eliseo Vañó, and Yoshiharu Yonekura 13 Radiation Protection for Personnel and the Environment�����������������������������������  273 Antonio C. Traino 14 Essentials of CT Image Interpretation�������������������������������������������������������������������  281 Davide Caramella, Matteo Revelli, and Alessandro Villa vii

viii

15 Essentials of MR Image Interpretation�������������������������������������������������������������������  317 Davide Caramella and Fabio Chiesa 16 Image-Guided and Radioguided Surgery���������������������������������������������������������������  351 Francesco Giammarile, Sergi Vidal-Sicart, Federica Orsini, Renato A. Valdés Olmos, and Giuliano Mariani Part II Organ/Apparatus-Specific Applications 17 Radionuclide Imaging for Non-tumor Diseases of the Brain�������������������������������  391 Duccio Volterrani, Giampiero Giovacchini, and Andrea Ciarmiello 18 Hybrid Imaging for Tumors of the Brain���������������������������������������������������������������  413 Giampiero Giovacchini, Mattia Riondato, Patrizia Lazzeri, Elisa Borsò, Valerio Duce, Rossella Leoncini, Elisabetta Giovannini, and Andrea Ciarmiello 19 Hybrid Imaging of the Head and Neck Region �����������������������������������������������������  431 Alejandro Fernández and Valle Camacho 20 Radionuclide Imaging of Cardiovascular Disease�������������������������������������������������  449 Matteo Bauckneht, Flavia Ticconi, Roberta Piva, Riemer H. J. A. Slart, Alberto Nieri, Silvia Morbelli, Paola Anna Erba, Cecilia Marini, H. William Strauss, and Gianmario Sambuceti 21 Radionuclide Imaging of Benign Pulmonary Diseases �����������������������������������������  499 Federica Guidoccio, Edoardo Airò, and Giuliano Mariani 22 Hybrid Imaging for Tumours of the Chest�������������������������������������������������������������  523 Roberto C. Delgado Bolton and Adriana K. Calapaquí Terán 23 Hybrid Imaging for Breast Malignancies���������������������������������������������������������������  543 Federica Padovano, Giuliano Mariani, and Marco Ferdeghini 24 Hybrid Imaging and Radionuclide Therapy of Musculoskeletal Diseases ���������  571 Paola Anna Erba, Martina Sollini, Roberta Zanca, Roberto Boni, Lesley Flynt, Elena Lazzeri, Giuliano Mariani, and Torsten Kuwert 25 Hybrid Imaging of Melanoma and Other Cutaneous Malignancies�������������������  645 Montserrat Estorch 26 Hybrid Imaging and Radionuclide Therapy in Hemato-oncology�����������������������  655 Paola Anna Erba, Martina Sollini, Roberto Boni, and Sara Galimberti 27 Hybrid Imaging and Radionuclide Therapy for Thyroid Disorders�������������������  707 Federica Guidoccio, Gayane Aghakhanyan, and Mariano Grosso 28 Hybrid Imaging in Non-thyroidal Endocrinological Disorders���������������������������  749 Duccio Volterrani, Federica Guidoccio, and Giuliano Mariani 29 Hybrid Imaging and Radionuclide Therapy of Neuroendocrine Tumors�����������  767 Duccio Volterrani, Lisa Bodei, and Federica Guidoccio 30 Radionuclide Imaging of the Nephro-­Urinary Tract���������������������������������������������  785 Duccio Volterrani, Federica Orsini, and Federica Guidoccio 31 Functional Imaging for Benign Conditions of the Gastrointestinal Tract�����������  809 Gayane Aghakhanyan, Elisa Fiasconaro, Elisa Tardelli, Mariano Grosso, and Italia Paglianiti 32 Hybrid Imaging for Malignant Conditions of the Gastrointestinal Tract�����������  841 Joan Duch and Albert Flotats

Contents

Contents

ix

33 Radionuclide Therapy for Tumors of the Liver and Biliary Tract�����������������������  859 Federica Guidoccio, Giuseppe Boni, Duccio Volterrani, and Giuliano Mariani 34 Hybrid Imaging for Gynecologic Malignancies�����������������������������������������������������  881 Elisa Lodi Rizzini, Elena Tabacchi, and Cristina Nanni 35 Hybrid Imaging for Male Malignancies�����������������������������������������������������������������  899 Akram Al-Ibraheem, Abdullah S. Al Zreiqat, Serena Chiacchio, and Abedallatif A. AlSharif 36 Nuclear Medicine in Pediatrics�������������������������������������������������������������������������������  925 Pietro Zucchetta and Diego De Palma 37 Hybrid Imaging of the Peripheral Lymphatic System �����������������������������������������  951 Paola Anna Erba, Roberto Boni, Martina Sollini, Andrea Marciano, Rossella Di Stefano, and Giuliano Mariani 38 Molecular Guidance for Planning External Beam Radiation Therapy���������������  977 Federica Orsini, Giovanna Pepe, Arturo Chiti, Giuseppe Roberto D’Agostino, Annibale Versari, Carlo Cavedon, Marco Ferdeghini, Paola Anna Erba, and Martina Sollini Part III Practice and Procedures (with Supplemental Online Files) 39 Radiopharmacy/Radiochemistry for Conventional Single-Photon Emitting and Therapeutic Radiopharmaceuticals������������������������������������������������� 1009 Alessandra Boschi and Adriano Duatti 40 Radiopharmacy/Radiochemistry for Positron-Emitting Radiopharmaceuticals, Including Quality Assurance and Process Validation Principles����������������������������������������������������������������������������������� 1019 Piero A. Salvadori 41 Recommendations for Conducting Clinical Trials with Radiopharmaceuticals ��������������������������������������������������������������������������������������������� 1039 Clemens Decristoforo and Serge K. Lyashchenko 42 Standard Operating Procedures for Quality Control of Gamma Cameras ����������������������������������������������������������������������������������������������������� 1051 Angela Vaiano 43 Standard Operating Procedures for Quality Control of PET/CT and PET/MR Tomographs��������������������������������������������������������������������������������������� 1061 Alessandra Zorz and Alessandro Scaggion 44 Standard Operating Procedures for Quality Control of Instrumentation for Radioguided Surgery������������������������������������������������������������� 1079 Angela Vaiano 45 Technical Aspects of Dual-Energy X-Ray Absorptiometry and Quantitative Computed Tomography for Assessment of Bone Mineral Density and Body Composition����������������������������������������������������������������� 1085 Pat Zanzonico 46 Current Practical Guidelines for the Most Common Nuclear Medicine Procedures������������������������������������������������������������������������������������������������� 1099 Irene Marini, Onelio Geatti, and H. William Strauss

x

47 Teaching Cases in Nuclear Medicine: Non-oncological Applications������������������� 1139 Venanzio Valenza, Isabella Bruno, Daniela Di Giuda, Germano Perotti, Fabrizio Cocciolillo, Valerio Lanni, Lucia Leccisotti, Daria Maccora, and Valentina Scolozzi 48 Teaching Cases in Nuclear Medicine: Oncological Applications������������������������� 1199 Laura Evangelista, Lucia Setti, Anna Rita Cervino, Gianluigi Ciocia, Lea Cuppari, Riccardo Vicinelli, and Emilio Bombardieri 49 Radiological Anatomy with CT: What the Nuclear Physician Should Know When Reading a PET/CT Scan������������������������������������������������������� 1241 Davide Caramella, Matteo Revelli, and Alessandro Villa 50 Radiological Anatomy with MR: What the Nuclear Physician Should Know When Reading a PET/MR Scan ����������������������������������������������������� 1257 Davide Caramella and Fabio Chiesa 51 Interpreting and Reporting the Results of Radionuclide Tests����������������������������� 1279 H. William Strauss and Federica Orsini Glossary����������������������������������������������������������������������������������������������������������������������������� 1297 Index����������������������������������������������������������������������������������������������������������������������������������� 1305

Contents

Contributors

Gayane Aghakhanyan  Regional Center of Nuclear Medicine, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy Edoardo Airò  Fondazione CNR/Regione Toscana Gabriele Monastero, Pisa, Italy Akram  Al-Ibraheem Department of Nuclear Medicine, King Hussein Cancer Center, Amman, Jordan Abedallatif  A.  AlSharif Department of Radiology and Nuclear Medicine, School of Medicine, Jordan University Hospital, University of Jordan, Amman, Jordan Abdullah S. Al Zreiqat  Department of Nuclear Medicine, Jordanian Royal Medical Services, Amman, Jordan Matteo Bauckneht  Nuclear Medicine, IRCCS Policlinico San Martino, Genoa, Italy Nicola Belcari  Department of Physics “E. Fermi”, University of Pisa and INFN, Pisa, Italy Lisa  Bodei  Molecular Imaging and Therapy Service, Department of Radiology, Memorial Sloan Kettering Cancer Center, New York, NY, USA Ronald Boellaard  Department of Radiology and Nuclear Medicine, VU University Medical Center, Amsterdam, The Netherlands Emilio Bombardieri  Nuclear Medicine Department, Humanitas Gavazzeni, Bergamo, Italy Scientific Direction, Humanitas Gavazzeni, Bergamo, Italy Giuseppe Boni  Regional Center of Nuclear Medicine, University Hospital of Pisa, Pisa, Italy Roberto Boni  Nuclear Medicine Service, “Papa Giovanni XXIII” Hospital, Bergamo, Italy Elisa Borsò  Nuclear Medicine Department, “S. Andrea” Hospital, La Spezia, Italy Alessandra  Boschi Department of Morphology, Surgical and Experimental Medicine, University of Ferrara, Ferrara, Italy Christian  Bracco Department of Medical Physics, Institute for Cancer research and Treatment, Candiolo, Turin, Italy Isabella  Bruno Nuclear Medicine Unit, Fondazione Policlinico Universitario A. Gemelli IRCCS, Roma, Italy Adriana  K.  Calapaquí  Terán  Department of Pathology, University Hospital Marques de Valdecilla (HUMV), Santander, Cantabria, Spain Valle  Camacho Nuclear Medicine Department, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain Amedeo Capotosti  Department of Physics, Catholic University of the Sacred Heart, Rome, Italy

xi

xii

Davide  Caramella Diagnostic and Interventional Radiology, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy Carlo  Cavedon  Nuclear Medicine Unit and Medical Physics Unit, University Hospital of Verona, Verona, Italy Anna  Rita  Cervino Nuclear Medicine and Molecular Imaging Unit, Veneto Institute of Oncology IOV – IRCCS, Padua, Italy Serena Chiacchio  Regional Center of Nuclear Medicine, University Hospital of Pisa, Pisa, Italy Fabio Chiesa  Unit of Radiology, ASL 5 “Spezzino”, Sarzana, La Spezia, Italy Arturo Chiti  Nuclear Medicine, Humanitas Cancer and Research Center IRCCS, Milan, Italy Department of Biomedical Sciences, Humanitas University, Milan, Italy Andrea Ciarmiello  Nuclear Medicine Department, “S. Andrea” Hospital, La Spezia, Italy Gianluigi Ciocia  Nuclear Medicine Department, Humanitas Gavazzeni, Bergamo, Italy Fabrizio Cocciolillo  Nuclear Medicine Unit, Fondazione Policlinico Universitario A. Gemelli IRCCS, Roma, Italy Lea Cuppari  Nuclear Medicine and Molecular Imaging Unit, Veneto Institute of Oncology IOV – IRCCS, Padua, Italy Giuseppe  Roberto  D’Agostino Radiotherapy and Radiosurgery, Humanitas Cancer and Research Center IRCCS, Milan, Italy Lawrence  T.  Dauer Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, NY, USA Department of Radiology, Memorial Sloan Kettering Cancer Center, New York, NY, USA Clemens Decristoforo  Department of Nuclear Medicine, Medical University of Innsbruck, Innsbruck, Austria Roberto  C.  Delgado  Bolton Department of Diagnostic Imaging and Nuclear Medicine, University Hospital San Pedro and Center for Biomedical Research of La Rioja (CIBIR), Logroño, La Rioja, Spain Diego De Palma  Nuclear Medicine Department, “H. Circolo” Varese, Varese, Italy Alberto  Del Guerra Department of Physics “E.  Fermi”, University of Pisa and National Institute of Nuclear Physics (INFN), Pisa, Italy Daniela Di Giuda  Institute of Nuclear Medicine, Università Cattolica del Sacro Cuore, Rome, Italy Nuclear Medicine Unit, Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy Rossella  Di Stefano Cardiovascular Disease Unit, Department of Surgical, Medical, Molecular Patology and Medical Emegencies, University of Pisa, Pisa, Italy Adriano Duatti  Department of Chemical and Pharmaceutical Sciences, University of Ferrara, Ferrara, Italy Valerio Duce  Nuclear Medicine Department, “S. Andrea” Hospital, La Spezia, Italy Joan Duch  Nuclear Medicine Department, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain Paola  Anna  Erba Regional Center of Nuclear Medicine, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy

Contributors

Contributors

xiii

Montserrat  Estorch Nuclear Medicine Department, Sant Pau Hospital, Autonomous University of Barcelona, Barcelona, Spain Laura  Evangelista Nuclear Medicine and Molecular Imaging Unit, Veneto Institute of Oncology IOV – IRCCS, Padua, Italy Maria  Evelina  Fantacci  Department of Physics “Enrico Fermi”, University of Pisa, Pisa, Italy Marco Ferdeghini  Nuclear Medicine Unit and Medical Physics Unit, University Hospital of Verona, Verona, Italy Alejandro Fernández  Nuclear Medicine Department, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain Elisa  Fiasconaro  Regional Center of Nuclear Medicine, University Hospital of Pisa, Pisa, Italy Elena  Filidei Nuclear Medicine Unit, Fondazione CNR/Regione Toscana Gabriele Monasterio, Pisa, Italy Albert Flotats  Universitat Autònoma de Barcelona, Barcelona, Spain Nuclear Medicine Department, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain Lesley Flynt  Department of Nuclear Medicine, M.D. Anderson Cancer Center, Houston, TX, USA Michela  Gabelloni Diagnostic and Interventional Radiology, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy Sara  Galimberti Hematology Unit, Department of Clinical and Experimental Medicine, University of Pisa, Pisa, Italy Onelio Geatti  Nuclear Medicine Center, University Hospital of Udine, Udine, Italy Francesco  Giammarile Nuclear Medicine and Diagnostic Imaging Section, Division of Human Health, Department of Nuclear Sciences and Applications, International Atomic Energy Agency, Vienna, Austria Assuero  Giorgetti Nuclear Medicine Unit, Fondazione CNR/Regione Toscana Gabriele Monasterio, Pisa, Italy Giampiero  Giovacchini  Nuclear Medicine Department, “S.  Andrea” Hospital, La Spezia, Italy Elisabetta Giovannini  Nuclear Medicine Department, “S. Andrea” Hospital, La Spezia, Italy Mariano  Grosso  Regional Center of Nuclear Medicine, University Hospital of Pisa, Pisa, Italy Federica  Guidoccio Regional Center of Nuclear Medicine, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy Luca  Indovina Department of Medical Physics, Fondazione Policlinico Universitario A. Gemelli, Rome, Italy Torsten  Kuwert  Clinic of Nuclear Medicine, Friedrich-Alexander-University of ErlangenNürnberg, Erlangen, Germany Adriaan  A.  Lammertsma Department of Radiology and Nuclear Medicine, Amsterdam UMC, Vrije Universiteit Amsterdam, Amsterdam, The Netherlands

xiv

Valerio  Lanni Nuclear Medicine Unit, Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy Elena Lazzeri  Regional Center of Nuclear Medicine, University Hospital of Pisa, Pisa, Italy Patrizia Lazzeri  Nuclear Medicine Department, “S. Andrea” Hospital, La Spezia, Italy Lucia  Leccisotti  Nuclear Medicine Unit, Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy Rossella Leoncini  Nuclear Medicine Department, “S. Andrea” Hospital, La Spezia, Italy Elisa Lodi Rizzini  Metropolitan Nuclear Medicine, AOU S. Orsola-Malpighi, Bologna, Italy Serge K. Lyashchenko  Department of Radiology, Memorial Sloan Kettering Cancer Center, New York, NY, USA Daria Maccora  Institute of Nuclear Medicine, Università Cattolica del Sacro Cuore, Rome, Italy Nuclear Medicine Unit, Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy Andrea  Marciano Regional Center of Nuclear Medicine, Department of Translational Research and Advanced Technologies in Medicine, University of Pisa, Pisa, Italy Giuliano  Mariani Regional Center of Nuclear Medicine, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy Cecilia  Marini Nuclear Medicine, Department of Health Sciences, University of Genoa, Genoa, Italy CNR Institute of Bioimages and Molecular Physiology, Milan, Italy Irene Marini  Nuclear Medicine Institute, Università Cattolica del Sacro Cuore and Nuclear Medicine Center, Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy Sören Mattsson  Medical Radiation Physics, Lund University, Malmö, Sweden Sara Mazzarri  Regional Center of Nuclear Medicine, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy Silvia Morbelli  Nuclear Medicine, IRCCS Policlinico San Martino, Genoa, Italy Nuclear Medicine, Department of Health Sciences, University of Genoa, Genoa, Italy Matteo Morrocchi  Department of Physics “E. Fermi”, University of Pisa and INFN, Pisa, Italy Cristina Nanni  Metropolitan Nuclear Medicine, AOU S. Orsola-Malpighi, Bologna, Italy Emanuele  Neri Diagnostic and Interventional Radiology, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy Alberto  Nieri Nuclear Medicine, Department of Health Sciences, University of Genoa, Genoa, Italy Federica  Orsini Nuclear Medicine Unit, “Maggiore della Carità” University Hospital, Novara, Italy Federica  Padovano Nuclear Medicine Unit and Department of Diagnostics and Public Health, University of Verona, Verona, Italy Italia  Paglianiti Regional Center of Nuclear Medicine, University Hospital of Pisa, Pisa, Italy Daniele Panetta  CNR Institute of Clinical Physiology, National Research Council, Pisa, Italy

Contributors

Contributors

xv

Roberto Pani  Department of Medical and Surgical Sciences and Biotechnologies, Sapienza University of Rome, Rome, Italy Giovanna Pepe  Nuclear Medicine, Humanitas Cancer and Research Center IRCCS, Milan, Italy Germano Perotti  Nuclear Medicine Unit, Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy Roberta  Piva Nuclear Medicine, Department of Health Sciences, University of Genoa, Genoa, Italy Giulia Puccini  Regional Center of Nuclear Medicine, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy Erinda Puta  Nuclear Medicine Unit, “Maggiore della Carità” University Hospital, Novara, Italy Daniele  Regge  Department of Surgical Sciences, University Hospital “Città della Salute e della Scienza”, University of Turin, Turin, Italy Department of Radiology, Institute for Cancer Research and Treatment, Candiolo, Turin, Italy Matteo Revelli  Department of Diagnostic Imaging and Laboratory Medicine, AUSL Reggio Emilia – IRCCS, Reggio Emilia, Italy Mattia Riondato  Nuclear Medicine Department, “S. Andrea” Hospital, La Spezia, Italy Piero A. Salvadori  PET/Cyclotron Unit, CNR Institute of Clinical Physiology, Pisa, Italy Massimo  Salvatori Nuclear Medicine Institute, Fondazione Policlinico Universitario A. Gemelli, Rome, Italy Nuclear Medicine Institute, Catholic University of the Sacred Heart, Rome, Italy Raffaele  Scafè  Department of Molecular Medicine, Sapienza University of Rome, Rome, Italy Gianmario Sambuceti  Nuclear Medicine, IRCCS Policlinico San Martino, Genoa, Italy Nuclear Medicine, Department of Health Sciences, University of Genoa, Genoa, Italy Alessandro  Scaggion Medical Physics Department, Veneto Institute of Oncology IOV  – IRCCS, Padua, Italy Valentina Scolozzi  Nuclear Medicine Unit, Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy Lucia Setti  Nuclear Medicine Department, Humanitas Gavazzeni, Bergamo, Italy Riemer H. J. A. Slart  Biomedical Photonic Imaging Group, University of Twente, Enschede, The Netherlands Martina Sollini  Department of Biomedical Sciences, Humanitas University, Milan, Italy Michele Stasi  Department of Medical Physics, Institute for Cancer research and Treatment, Candiolo, Turin, Italy H.  William  Strauss Molecular Imaging and Therapy Service, Memorial Sloan Kettering Cancer Center, New York, NY, USA Elena Tabacchi  Metropolitan Nuclear Medicine, AOU S. Orsola-Malpighi, Bologna, Italy Elisa Tardelli  Regional Center of Nuclear Medicine, University Hospital of Pisa, Pisa, Italy

xvi

Flavia  Ticconi Nuclear Medicine, Department of Health Sciences, University of Genoa, Genoa, Italy Angela Vaiano  Health Physics Unit of Prato and Pistoia, “San Jacopo” Hospital, Pistoia, Italy Renato A. Valdés Olmos  Nuclear Medicine Section and Interventional Molecular Imaging Laboratory, Department of Radiology, Leiden University Medical Centre, Leiden, The Netherlands Nuclear Medicine Department, Onze Lieve Vrouwe Gasthuis Hospital, Amsterdam, The Netherlands Venanzio  Valenza Institute of Nuclear Medicine, Università Cattolica del Sacro Cuore, Rome, Italy Nuclear Medicine Unit, Fondazione Policlinico Universitario A. Gemelli IRCCS, Roma, Italy Eliseo Vañó  Department of Radiology, Complutense University of Madrid, Madrid, Spain Annibale  Versari Nuclear Medicine Unit, Department of Oncology and Advanced Technologies, AUSL-IRCCS Reggio Emilia, Reggio Emilia, Italy Riccardo  Vicinelli  Postgraduate Specialty School in Nuclear Medicine, “Milano Bicocca” University, Milan, Italy Sergi  Vidal-Sicart Nuclear Medicine Department, University Hospital Clinic, Barcelona, Spain Alessandro Villa  Unit of Radiology 2, ASL 5 “Spezzino”, Sarzana, La Spezia, Italy Duccio  Volterrani Regional Center of Nuclear Medicine, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy Yoshiharu Yonekura  National Institute of Radiological Sciences, Chiba, Japan Roberta Zanca  Regional Center of Nuclear Medicine, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy Pat Zanzonico  Memorial Hospital Research Laboratories, Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, NY, USA Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, NY, USA Alessandra Zorz  Medical Physics Department, Veneto Institute of Oncology IOV – IRCCS, Padua, Italy Pietro Zucchetta  Nuclear Medicine Department, University Hospital of Padua, Padua, Italy

Contributors

Part I Basic Science

1

Fundamentals of Natural and Artificial Radioactivity and Interaction of Ionizing Radiations with Matter Alberto Del Guerra and Daniele Panetta

Contents 1.1   Brief History of Radiation and Radioactivity 1.1.1  Wilhelm Röentgen 1.1.2  Henry Becquerel 1.1.3  Marie and Pierre Curie 1.1.4  Ernest Rutherford 1.1.5  International Commission on Radiological Protection

                 

1.2   Physical Basis of a Radiation Field

   5

1.3   Radioactive Decays 1.3.1  Electron Energy Levels 1.3.2  Physical Principles of the Radiation Sources 1.3.3  The Nuclear Decays

           

1.4   Natural Sources of Radiation

   9

1.5   Artificial Sources of Radiation

 12

1.6   Interaction of Charged Particles with Matter

 14

1.7   Interaction of Neutral Particles with Matter

 15

References

 19

Learning Objectives

• To build up the competence to describe the main discoveries that paved the way to the modern science of radiation and radioactivity and their numerous applications • To become acquainted with the ICRU and SI units for quantitative description of radiation sources, radiation fields, and interaction of radiation fields with matter

A. Del Guerra (*) Department of Physics “E. Fermi”, University of Pisa and National Institute of Nuclear Physics (INFN), Pisa, Italy e-mail: [email protected] D. Panetta CNR Institute of Clinical Physiology, National Research Council, Pisa, Italy

4 4 4 4 4 4

6 6 6 7

• To acquire the knowledge of the various types of decays (alpha, beta, gamma), laws of decays, decay chains, and equilibrium • To become familiar with the concept and the description of natural occurring radioisotopes, terrestrial radioactivity, radioactive families, and cosmic rays • To be able to understand the working principles and the application of the generators of electromagnetic radiation, nuclear reactions and generators of radionuclides, and cyclotrons • To understand the physical principles of Coulomb interactions, light particles versus heavy particles, stopping power, and range • To be able to describe the interaction of photons with matter (photoelectric effect, Compton scattering, e−/e+ pair production) and the interaction of neutrons

© Springer Nature Switzerland AG 2019 D. Volterrani et al. (eds.), Nuclear Medicine Textbook, https://doi.org/10.1007/978-3-319-95564-3_1

3

4

1.1

A. Del Guerra and D. Panetta

 rief History of Radiation B and Radioactivity

1.1.1 Wilhelm Röentgen X-rays were discovered by Wilhelm Röentgen on Christmas Eve 1895, and the official announcement was made on December 28, 1895. The use of X-rays for medical purposes was rapidly accepted: in Europe many hospitals used X-rays for the diagnosis of trauma and fractures and for locating bullets in soldiers. X-rays were used in absence of knowledge of the potential detrimental effects of ionizing radiation and their interaction with living matter. Nevertheless, the idea that X-rays could be used to treat tumors was introduced in January 1896 (within 1 month after the announcement of the discovery), when a Chicago electrician, Emil Grubbe, started using X-rays to radiate a woman with relapsing breast cancer, becoming the first radiotherapist in history.

1.1.2 Henry Becquerel Henry Becquerel in 1896 was carrying out a series of experiments to investigate phenomena related to the luminescence and/or phosphorescence of certain materials. In particular, he was looking for a possible relationship with the X-ray phenomenon, which he had come to know a few months earlier. In fact, the presentation of the work on X-rays performed by W.C. Röentgen had taken place at the beginning of 1896 at the Académie Des Sciences in Paris. During his experiments, Henry Becquerel found that uranium salts placed casually beside photographic plates, completely protected from light, caused their darkening. He hypothesized that such salts would emit unknown rays that were able to penetrate the plate container and very similar to X-rays. He had thus discovered the phenomenon of radioactivity.

1.1.3 Marie and Pierre Curie Marie and Pierre Curie, in continuing the experiments of Henry Becquerel, discovered 2 years later that uranium salts were not the only ones to emit penetrating radii but that other substances, such as thorium, had the same property. Marie Curie suggested these substances to be defined “radioactive” (from the Latin word “radium,” i.e., ray). In the continuation of the studies with a new mineral, the pitchblende, the Curies succeeded in finding new elements, polonium, and radium, whose “radioactivity” was much higher than that of thorium. In particular, using radium, for the first time, they identified and classified three types of radiation:

the first electrically positive charged, the second negative charged, and the third neutral. The first three letters of the alphabet α (alpha), β (beta), and γ (gamma) were associated with these rays.

1.1.4 Ernest Rutherford Ernest Rutherford, during his studies for the understanding of the atomic structure, discovered that alpha particles penetrate only a few tens of micrometers in aluminum, while beta particles are characterized by a 100-fold more penetrating power. Using electric and magnetic fields, he verified that in the electric field the beta particles are strongly deflected toward the positive pole, the alpha particles are deflected to a lesser degree toward the negative pole, while the gamma ray trajectory is not affected by the presence of the field. As a result, it was understood that the beta particles have a negative charge, the alpha particles have a positive charge (and a greater mass than the beta particles), and the gamma rays are electrically neutral. The deflection studies of such particles crossing thin layers of material (gold sheets) then led to the formulation of the atomic structure with a central positive charge core (i.e., the nucleus), where almost all the mass is concentrated and charged (negative) electrons rotating around it: the Rutherford’s model of the atom was established.

1.1.5 International Commission on Radiological Protection Several decades after their discovery, it was realized that X-rays could produce far superior effects than those originally hypothesized: it was understood that their “administration” for medical purposes could not be prolonged because of safety issues and had to be regulated. The process of evaluating radiation damage was started through the most apparent effects: the emergence of a radiation erythema was considered the typical warning signal. However, this alarm index was only a qualitative element, and it was necessary to wait until 1928 to have a standardized unit of measure of the intensity of a radiation beam by using the induced ionization analysis. The International Commission on Radiological Protection (ICRP) was established in 1928 at the International Radiology Congress (the original name being “International X-ray and Radium Protection Committee”). The ICRP is still an emanation of the International Society of Radiology, but its range covers not only the medical field but also all aspects of protection against ionizing radiation. The committee issues recommendations on the basic principles of radioprotection. These recommendations, in turn, are transposed by other international and national organizations in the form of specific scientific documents and,

1  Fundamentals of Natural and Artificial Radioactivity and Interaction of Ionizing Radiations with Matter

then, by international, national, and regional government ­bodies, through applicable laws and decrees.

Key Learning Points

• Wilhelm Röentgen, Henry Becquerel, and Marie and Pierre Curie together with Ernest Rutherford are the authors of the main discoveries that paved the way to the modern science of radiation and radioactivity and to their numerous applications. • The International Commission for Radiation Protection is the scientific body responsible for issuing the recommendations on the basic principles of radioprotection.

1.2

Physical Basis of a Radiation Field

The term radiation is applicable to every physical quantity that is able to transport energy through space and time. The source of the radiation could be due to transformations of unstable nuclei or interactions of particles with nuclei. If the particles have enough energy, the electromagnetic (EM) radiation and charged particles can ionize the matter they interact with. For this reason, these types of radiation are called ionizing radiation. The radiation could be emitted from either the transformations of an unstable nucleus (such as beta electrons, α-particles, γ photons, neutrons, and fission fragments) or by secondary processes (such as internal conversion electrons or Auger electrons, X-rays, and UV photons). The radiation is emitted in all directions (4π geometry). Particles with an electric charge can directly ionize atoms through multiple Coulomb collisions: they are called directly ionizing radiation. Neutral particles and EM radiation (e.g., photons) can also ionize atoms but only through sparse interactions. These latter interactions produce secondary charged particles that, in turn, lead to direct ionization of the surrounding atoms. Hence, neutral particles are called indirectly ionizing radiation. A photon is a quantum of electromagnetic energy, i.e., the elementary unit of the energy transported by an electromagnetic wave (in analogy, the electron charge is a quantum of electric charge). Because of the wave-particle duality of quantum mechanics, the photon can be treated as a particle with its own energy (E = hν). For a quantitative description of a radiation field, we will follow the convention adopted by the International Commission of Radiation Units and Measurements (ICRU), in Ref. [1]. A radiation field could be thought of as a portion of space in which particles move along individual trajectories. At a point P in the field, the particle fluence (or flux), Φ, is defined as the number N of particles passing in a time interval

5

Δt through a sphere of cross-sectional area S centered in P. When S becomes infinitely small, Φ becomes

F = dN / dS éë m -2 ùû 

(1.1)

Since the number N is a stochastic quantity, in the above equation, the expectation value of N must be considered for the computation of Φ. The time derivative of Eq. (1.1) represents the number of particles passing through P in the unit time, i.e., the fluence rate (or flux density) which is indicated by ϕ:

f = d / dt ( dN / dS ) = d 2 N / ( dt dS ) éë m -2 s -1 ùû . 

(1.2)

Sometimes the information about the number of particles is not sufficient. Also the total energy transported by particles requires a specific quantitative parameter. Let us denote by R the total kinetic energy of the N particles through P: R = åEi ni , i = 1, ¼, M , (1.3)  i where ni is the number of particles having energy Ei, M is the total number of energies available in the field, and åni = N i

is the total number of particles. The distribution of particles ni in each energy level is referred to as the energy spectrum of the radiation. Since a finite number of energies are present among all the N particles, the spectrum is a discrete energy spectrum. In a most general case, the energy of the particles in a field can have a value in a continuous interval between 0 and Emax. In this case, the summation in Eq. (1.3) is replaced by an integral, and the energy spectrum is called continuous. Examples of discrete energy spectra are the photon field of a γ-emitting radionuclide (e.g., 99mTc) or the particle field of an α-emitter (e.g., 210Po), whereas examples of continuous energy spectra are the bremsstrahlung field, the particle field of a β-emitter (e.g., 3He), or the particle field of a positron emitter (e.g., 11C). The energy fluence Ψ and energy fluence rate ψ, measured in the SI in J m−2 and J m−2 s−1, are defined, respectively, as:

Y = dR / dS éë Jm -2 ùû , 

y = d / dt ( dR / dS ) = d 2 R / ( dt dS ) éë Jm -2 s -1 ùû . 

(1.4) (1.5)

Key Learning Points

• Radiation with sufficient energy can ionize matter, whether directly or indirectly according to its physical nature. • Several quantities are identified by ICRU to define quantitatively a radiation field in terms of the number of particles passing through a given portion of space or in terms of the energy that they carry.

6

1.3

A. Del Guerra and D. Panetta

Radioactive Decays

1.3.1 Electron Energy Levels To understand the motion of electrons around the nucleus, it may be useful to start from the atomic model of Rutherford. In this model it is assumed that the electrons move on elliptical orbits around the nucleus, with a motion similar to that of the planets around the Sun. If the eccentricity of the orbits is neglected, the motion of each electron can be assimilated to a uniform circular motion. In this case, the centripetal force is given by the Coulomb’s force between the nucleus and the electron. For a given radius of the orbit, the electron potential energy in an atom with atomic number Z is given by

Ep = -1 / 4pe 0 Ze 2 / r



(1.8)

In addition, for each value of ℓ, there are (2ℓ + 1) possible values of the quantum magnetic number, m (the projection of ℓ on an arbitrary axis z). Finally, for each m, the electrons can take two different spin states. It follows that for each value of ℓ, the maximum number of electrons is given by N e (  ) = 2 ( 2 + 1) 



(1.9)

By combining Eqs. (1.8) and (1.9) and knowing that the sum of the first k numbers is equal to k (k + 1)/2, it is found that the maximum number of electrons in a given energy level is n -1

(1.6)

where the minus sign indicates that the electron is attracted by the nucleus. According to the previous equation, the negative potential energy of the electron increases by increasing r until it goes to zero when its distance from the nucleus becomes very large with respect to the atom size. Suppose now that one wants to “extract” an electron from the atom, i.e., to take it from an initial distance r from the nucleus (where the potential energy is equal to Ep) to a very large final distance, that is, to the “infinite” (where the potential energy is zero). If we assume that the kinetic energy of the electron is the same at the two positions, both initial and final, the energy required to extract the electron is Efinal - Ep = 0 - Ep = Ep 

 = 0,1, ¼, n - 1 



(1.7)

To displace the electron from the atom, that is, to ionize the atom, it is necessary to provide an energy equal to or greater than the absolute value of Ep. For this reason, this energy is called electron energy. The atomic model of Rutherford is a classical model: in fact, it cannot explain many phenomena that can only be described by a quantum model. In the early twentieth century, it was replaced by the Bohr model (semiclassical model), which in turn was further refined by modern quantum mechanics. In the present atomic model, each electron is in a particular quantum state, characterized by a certain energy value and other physical quantities (such as angular momentum and spin). Not all energies are allowed: only a few “discrete” or, better yet, “quantized” values ​​are available. These energy values represent ​​ the so-called energy levels uniquely distinguished from the main quantum number. The energy of electrons in each energy level increases with increasing n. For historical reasons (inherited by spectroscopy), the energy levels with n = 1, 2, 3, 4, 5 are given by the letters K, L, M, N, P, respectively. For each n value, the electrons can assume different values of ​​ the orbital quantum number ℓ:

namely • • • • •

N e ( n ) = å2 ( 2 + 1) = 2n 2  =0



(1.10)

Ne = 2 for K orbitals Ne = 8 for L orbitals Ne = 18 for M orbitals Ne = 32 for N orbitals Ne = 50 for P orbitals

1.3.2 P  hysical Principles of the Radiation Sources The transfer of energy from the radiation to the atoms of a medium causes its excitation or ionization. In the excitation process, the transferred energy is sufficient only to change the atom state to an energized energy level. On the other hand, ionizing radiation can produce the ionization (electron expulsion) of the atoms of the medium. A biological medium exposed in a field of ionizing radiation becomes the site of a series of processes, originating from the transfer of energy from radiation to the medium, which manifest themselves with various effects. It is necessary to relate the effects produced with the physical characteristics of the radiation field and those of the irradiated medium. To do this, the characteristics of the various types of particles and radiations, their interactions with the material crossed, and the physical quantities and units used to describe the characteristics of the radiation fields and material media must be known with respect to the interactions with ionizing radiation. The term “radiation” is usually used to describe apparently very different phenomena such as light emitting by a lamp, heat from a flame, or X-ray from a medical diagnostic machine. A common characteristic of all these phenomena is the transport of energy into space without a material propagation medium. Energy can be transported in vacuum or in dielectric materials through electromagnetic waves, i.e., oscillations of the electric field and the magnetic field that

1  Fundamentals of Natural and Artificial Radioactivity and Interaction of Ionizing Radiations with Matter Frequency (Hz)

Type of radiation Wavelength (m)

104

108

Radio frequency (RF) ∼10–2

1012

Microwaves

1015

Infrared

∼10–3

∼10–5

Visible ∼5 × 10–7

1016

Ultraviolet ∼10–8

7 1018

X-rays

1020

Gamma rays

∼10–10

∼10–12

Fig. 1.1  The electromagnetic spectrum. Longer wavelengths correspond to smaller frequency due to the inverse relationship stated in Eq. (1.12). In this figure, the waves are not drawn in scale for graphical reasons (adapted from Ref. [2])

propagate in space. Electromagnetic waves are characterized by a constant speed in the vacuum and in a medium as

v vacuum = c = 2.99729 ´ 108 m / s (1.11a) v medium = c / n (1.11b)

respectively, where n is the refractive index of the medium. The frequency of an electromagnetic wave, indicated by the symbol ν, is defined as the number of field oscillations in the time unit at a given location of the space. The w ­ avelength, indicated by λ, is the distance between two consecutive ­maximum field strength (peaks) positions at a given instant (see Fig. 1.1). The relation that correlates speed, frequency, and wavelength is

n =c/l 

(1.12)

For the atom to be ionized due to wave interaction, the wavelength must have a similar value to the size of the atom. Therefore the wavelength λ must be of the order of 10−10 m. The frequency of such a wave can be calculated with Eq. (1.12), obtaining ν = c/λ ≈ 3 × 1018 Hz. Considering that the visible light frequency is in the order of 1014  Hz (about 10,000 times lower), it is clear that it is not possible to ionize the atoms with this type of radiation. It is useful to reformulate the concept in terms of energy transmitted by electromagnetic radiation. Electromagnetic waves propagate within “packets,” called photons, which can be treated as energy particles equal to

E = hn 

(1.13)

where h = 6.62 × 10−34 [J s] is the Planck constant. For a photon to ionize an atom, it is necessary that its energy is greater than or equal to the binding energy of the atomic electrons involved in the interaction. To express numerically the energies involved in the ionization and excitation processes, it is convenient to use a special unit called electronvolt (eV). One electronvolt is defined as the kinetic energy acquired by an electron that is

accelerated by an electrical potential difference of 1  V.  The kinetic energy acquired by an electric charge that is accelerated by a uniform and constant electric field is given by

Ec = q ´ DV (1.14)

Therefore, by knowing that the charge of the electron is ≈1.602 × 10−19 C, using Eq. (1.14) and the definition of electron volt, we get the following conversion value:

1eV = 1.602 ´ 10-19 [C ´ V ] = 1.602 ´ 10-19 J 

(1.15)

1.3.3 The Nuclear Decays Atoms in nature can be stable or unstable. An atom is stable if its nucleus does not undergo spontaneous transformations. This means that in a stable atom, the number of protons and neutrons contained in the nucleus maintains their nuclear energy levels unchanged over time. Unstable atoms are those in which the nucleus spontaneously undergoes transformations. Such spontaneous nuclear transformations are called “decays.” A nuclear decay is always accompanied by the emission of some form of radiation: for this reason, unstable elements are also called radioactive. Following one or more transformations, a radioactive nucleus always decays into a stable nucleus, which can belong to a chemical species equal to or different from the initial one. The nuclear decay is a probabilistic phenomenon. It is not possible to determine the instant in which a given unstable nucleus will undergo a transformation. However, we can quantitatively describe the temporal evolution of a very large set of nuclei by the following law of radioactive decay:



N ( t ) = N 0 exp ( -l t ) 

(1.16)

8

A. Del Guerra and D. Panetta

In the previous equation, N0 represents the number of radioactive nuclei contained in the material (sample) considered at a given initial moment t0 = 0; N(t) represents the number of radioactive nuclei contained in the sample at a time t from the initial instant (i.e., the number of nuclei that at time t has not yet undergone any transformation); and λ is the decay constant, which represents the average number of nuclear decays occurring in the unit of time. The constant λ is characteristic of each radioactive isotope and does not depend on the amount of matter contained in the radioactive sample under examination. It has the inverse dimension of time and is measured in s−1. Equation (1.16) describes the following concept: the number of unstable nuclei, contained in a given radioactive material sample, decreases exponentially over time; τ is defined as the average lifetime of a given nuclear species, i.e., the inverse of the decay constant λ:

t = 1/ l 



(1.17)

“Average lifetime” means the average time between the production (either natural or artificial) and the decay of a given radioactive nucleus. This time varies considerably from isotope to isotope and can assume values between fractions of second and billions of years. Another frequently used quantity, strictly related to the average lifetime, is the half-life, t1/2, indicating the time required to halve the number of radioactive nuclei contained in the sample (see Fig. 1.2): t1/ 2 = t ´ ln 2 (1.18)



The activity A(t) is defined as the average number of nuclear decays occurring at time t in the unit of time. According to the definition, the activity coincides with the decay rate (i.e., the derivative) of N(t), changed by sign: N(t)/N0

1.0 0.9 0.8 0.7



A ( t ) = -dN / dt = - N 0 ( -l ) exp ( -l t ) = l N ( t ) = N ( t ) / t (1.19a) A ( t ) = A0 exp ( -l t ) 

(1.19b)

The activity measurement unit is the Becquerel (Bq). An activity of 1 Bq corresponds to a nuclear transformation (disintegration) per second. Another measure of activity, of historical interest but also very much used in practice, is the Curie (Ci), defined as the activity of 1 g of radium-226. The conversion between Bq and Ci is obtained by the following equivalence:

1 Ci = 3.7 ´ 1010 Bq (1.20)

There are various types of nuclear decay. The α-decay is the spontaneous particle spraying by heavy nuclei, characterized by an excess of protons; α is a positively charged heavy particle, formed by two protons and two neutrons, identical to a helium-4 nucleus (4He). Some examples of α-decaying radionuclides are 226Ra, 222Rn, and 210Po. The β− and β+ decays consist of the emission of an electron (e−) and of a positron (positive electron, indicated by e+) by a nucleus, respectively. Some elements that undergo decay β are 90Sr (β−) and 18F (β+). The γ-decay consists in the emission of a photon by the nucleus originally in an excited state. Photons emitted in the γ-decay are physically indistinguishable from X-rays, though the latter are produced by de-energizing processes of atomic electrons. X photons have generally lower energies than γ photons. The distinction between photons of type X or γ is based only on the mechanism by which they were generated (atomic or nuclear) and not on the energy they carry. Radionuclides may have different allowed routes of nuclear de-excitation. When more decays are possible for a given nuclear species, we use the term “branching ratio” (BR) of a decay mode (or decay channel) to denote the ratio between the number of decays of that mode and the total number of disintegrations. For instance, the BR for β+ decay of 18F is 96.86%, whereas the same nuclide can decay also by electron capture (EC) with a BR of 3.14%.

0.6 N=N0/2

0.5 0.4

N=N0/4 N=N0/8

0.3

Key Learning Points

0.2

• Atomic electrons are arranged in orbitals with discrete energy levels and angular momentum according to quantum mechanics. • The atomic nuclei can be stable or unstable. • Unstable nuclei can reach stability by undergoing single or (more frequently) chains of transformations (i.e., decays) leading to the emission of alpha particles (4He nuclei), beta particles (electrons or positrons), or gamma photons (i.e., EM waves).

0.1 0.0 0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0 t/τ

t=t1/2

t=2t1/2

t=3t1/2

Fig. 1.2  Exponential decay law. N0 is the number of radioactive nuclei contained in a sample at the initial instant t0 = 0; N(t) is the number of non-decayed nuclei at time t; and τ and t1/2 indicate the average lifetime and the half-life of the considered nuclear species, respectively (reproduced from Ref. [2])

1  Fundamentals of Natural and Artificial Radioactivity and Interaction of Ionizing Radiations with Matter

1.4

Natural Sources of Radiation

Radioactive isotopes reach stability by radiation emission. Such isotopes can be naturally occurring in rocks, in the atmosphere, or in water, or they can be artificially produced. In the first case, they are classified as natural sources of radiation. Some natural sources of interest in the radioprotection field are uranium-238 (238U), potassium-40 (40K), radium-226 (226Ra), and radon-222 (222Rn). In particular, 222Rn is one of the major sources of radiation risk for the population, as it is a colorless and tasteless gas that can emerge from the subsoil to reach high concentrations in basements and poorly ventilated areas. The radioactive isotopes present in the rocks are referred to as primordial sources of radiation, as they have been produced in cosmic events prior to the formation of Earth. Other natural-type radiations are cosmic radiation, mainly consisting of high-energy protons and cosmogenic radiation such as tritium (hydrogen-3) and carbon-14 being produced by the interaction of cosmic radiation with the stable nuclei present in the atmosphere or on the Earth. The main properties of all known isotopes, whether they are stable or unstable, natural or artificially produced, are organized in the Nuclide Table (see Fig. 1.3a). In this table, the various nuclides are ordered in a matrix where the rows are characterized by an equal number of ­protons and the columns are characterized by an equal number of neutrons. Moving from left to right, the N value increases, while going from the bottom up, there are Z increasing numbers. Stable isotopes in the Nuclide Table form the so-called stability curve (see Fig. 1.3b). By observing this curve, it is noted that for lightweight elements (Z   92 (transuranic elements) are not naturally occurring and can only be produced artificially (see next section). Bismuth-209 (Z = 83), which has been long regarded as the heaviest stable nucleus, has been recently found to be a metastable isotope undergoing α-decay, even though with an incredibly long half-life of 2 × 1019 years, i.e., more than one billion times the age of the universe [4]. Elements from polonium to uranium are linked to each other by decay chains, also called radioactive “families” or radioactive series (see Fig. 1.4). Each family has a radioactive progenitor, decaying into radioactive daughters and so on until a stable configuration is reached. The three main radioactive families have a naturally occurring radioactive progenitor. The thorium family starts from thorium-232, undergoing α-decay

9

with half-life of 1.4  ×  1010  year and ending to the stable isotope lead-208. All elements of the radioactive series undergo α or β decay; hence all the modifications of the mass number through the series occur in multiples of 4. For this reason, the thorium family is sometimes referred to as the A = 4n series, meaning that all the elements of the series have a mass number that can be obtained by multiplying 4 by an integer number (e.g., n = 58 for thorium-232, n = 57 for radium-228, and so on). With this convention, the socalled uranium family is referred to as A = 4n + 2 family. This series has uranium-238 as progenitor (half-life 4.47 × 109 year) and ends into the stable isotope lead-206. Radium-226 and its daughter radon-222 belong to this decay chain. Also the uranium-actinium (or simply actinium) family (A = 4n + 3) ends to a stable isotope of lead (lead-207); the name of this series follows from the historical name of 235U, “actinouranium” (half-life 7 × 108 year), with lower natural abundance with respect to 238U. It also helps to avoid ambiguity with the uranium family that starts from a uranium isotope as well. A family with A = 4n + 1 also exists (the neptunium series), but it cannot be included in the present list of natural sources of radiation because their members can only be produced artificially by nuclear reactions. The rate of production of members of a radioactive family depends on the availability of their parents; hence, in a given segment of a decay chain A → B where A and B are both unstable, the rate of emission of radiation of the daughter nucleus B is depending on the decay rate of its parent A. When t1/2,A is much longer than t1/2,B, after a sufficient amount of time, the amount of daughter nuclei NB reaches a constant value, due to the fact that the rate of its production equals the rate of its decay. In formulae, this can be show by considering that the time derivative of NB is given by dN B = lA N A - lB N B  dt



(1.21)

The first term on the right-hand side has a positive sign because NB grows up as the parent nucleus A undergoes disintegration. The more general solution of the above differential equation is



N B = N 0, A

lA ( e-lAt - e-lBt ) + N0,Be-lBt  lB - lA

(1.22)

where N0,A and N0,B are the initial numbers of parent and daughter nuclei, respectively. Let us suppose to start from a pure sample of A, so that N0,B = 0. If t1/2,A ≫ t1/2,B then λA ≪ λB (see Eqs. (1.17) and (1.18)) and hence λB − λA ~ λB. Furthermore, if the observation time t is long enough (i.e., t ≫ t1/2,B), we have e - lBt  e - lA t . Under these approximations, the solution for NB can be rewritten as follows:

10

A. Del Guerra and D. Panetta

a Z 100

N

80

Z=

60 40

Stable nuclei

20

0

20

40

60

b

80

100

N

Fig. 1.3 (a) Excerpt from “Table of Radioactive Isotopes” (Ref. [3]). (b) The stability curve. For increasing Z, the stable nuclei are those characterized by an excess of neutrons (adapted from Ref. [2]). The plotted data mimics the real distrubution of the stable nuclei



NB = NA

lA lB 

(1.23)

In the above equation, we have exploited the equation N A = N 0, A e - lA t . This situation is called secular equilibrium. In terms of activity, it’s easy to show that multiplying both sides of Eq. (1.23) by λB we get

lB N B = lA N A (1.24)

and hence AB = AA. In other words, at secular equilibrium the activity of the daughter equals the activity of the parent. Because we have assumed a very long half-life of the parent species, at secular equilibrium, this activity remains constant with good approximation until some processing is done to break the equilibrium condition (e.g., uranium ores are extracted during mining). As an example, radon-222 (t1/2 = 3.82 days) is in equilibrium with its parent radium-226

1  Fundamentals of Natural and Artificial Radioactivity and Interaction of Ionizing Radiations with Matter

α β–

A

X

Ancestor

A

X

Stable Z 238

234

U

235

U

U

234M 228Th

232Th

234Th

92 (Uranium) 231Pa

230Th

91 (Protactinium) 227Th

231Th

90 (Thorium)

227Ac

228Ac 228Ra

11

224Ra

226Ra

89 (Actinium) 223Ra

88 (Radium)

223Fr 220Rn

222Rn

219Rn 218At

216Po

212Po

218Po

212Bi 212Pb

219At 214Po

214Bi 214Pb

208Pb 208Tl

87 (Francium)

210Po 210Bi

210Pb 210Tl

215At 215Po

215Bi 206Pb

86 (Radon) 85 (Astatine) 211Po 211Bi

211Pb

206Tl

83 (Bismuth) 207Pb

207Tl

206Hg

Thorium family (A=4n)

84 (Polonium)

82 (Lead) 81 (Thallium) 80 (Mercury)

Uranium family (A=4n+2)

Actinium family (A=4n+3)

Fig. 1.4  The three radioactive families with naturally occurring ancestors (thorium-232, uranium-238, and uranium-235). The half-lives and branching ratios are not reported for graphical reasons. The reader can refer to Ref. [5] for a complete decay scheme of all these elements

(t1/2  =  1620  years) if the radium-containing sample is left undisturbed for t ≫ t1/2 (222Rn) (typically 4–5 weeks). When the parent’s half-life is not much greater than the observation time, the daughter’s half-life is not treated as negligible, and its activity does not reach a constant value after long times. In this case, Eq. (1.23) (which is valid after long times of observation) must be rewritten as follows:



NB = NA

lA . lB - lA 

(1.25)

Multiplying both sides by λB, we get the following equation in terms of activity:



AB = AA

lB . lB - lA 

(1.26)

The previous equation implies that, after a sufficient amount of time, the daughter’s activity becomes greater than

the parents’ one even though the ratio between them remains constant over time. This condition is called transient equilibrium. By starting from a pure A sample, the time required to reach the maximum activity of the daughter B is given by



tmax

æl ö ln ç B ÷ l = è A ø. lB - lA 

(1.27)

A common example of transient equilibrium in nuclear medicine is that of molybdenum-99 (the production of 99Mo is discussed in the next section, as it is only produced artificially), undergoing β− decay with half-life of 65.94  h. Its daughter 99m Tc is in an excited state, and its stability is reached by γ-decay with a half-life of 6 h. By starting from a pure sample of 99Mo, the maximum activity of 99mTc is reached after 22.8 h. Another important primordial radioactive element is potassium-40. This isotope has a half-life of 1.25 × 109 year and undergoes either β+ decay to calcium-40 (branching ratio

12

A. Del Guerra and D. Panetta

89.28%) or electron capture to argon-40 (branching ratio 10.72%). Approximately 120  ppm of the total amount of potassium found in nature is in form of 40K. This is important from the point of view of nuclear medicine because for a man weighting 70  kg, about 16  mg of 40K are contained in the body giving rise to a background body activity of ca 4.3 kBq. Besides radiation coming from radioactive elements present in rocks and soil, conveniently organized in radioactive families as explained above, radiation also comes from outside the Earth’s atmosphere and can be either originated by the Sun or from outside the solar system. The solar component of such cosmic radiation is mainly composed by high-­ energy protons (primary radiation) that produce secondary radiation (photons, electrons, neutrons, pions, and muons) when interacting with the molecules of the atmosphere. The primary cosmic ray component coming from outside the solar system has in general higher intensity, and its spatial distribution is isotropic.

Key Learning Points

• Radioactive sources can be naturally occurring, such as those present in rocks (terrestrial radioactivity) or coming from outside the Earth (cosmic rays). • Naturally occurring radioisotopes all belong to radioactive decay chains (or families) starting from thorium-232 (thorium family), uranium-238 (uranium family), or uranium-235 (actinium family).

1.5

Artificial Sources of Radiation

The natural sources of radiation are of little interest in medicine. Most natural radioisotopes have a lifetime too long to be safely used in therapeutic or diagnostic fields. Furthermore, it is also difficult to obtain them with the levels of purity required for this type of application. It is preferable to artificially produce the desired radioisotopes using particle accelerators that allow the nuclei contained in a target to be activated in a controlled manner. A very common artificial radiation source in the medical field is the radiogenic tube, used in radiodiagnostics to produce X-rays (Fig. 1.5). In a radiogenic tube, electrons emitted by a cathode by thermal effect are initially accelerated by an electrical potential difference (typically 40–140  kV depending on the specific diagnostic application) and, subsequently, are abruptly braked in a collision with a heavy metal anode. The cathode consists of one or two tungsten filaments that are heated by Joule effect at temperatures of 1500– 2600  °C.  The anode can be tungsten or molybdenum (the latter is used for mammography tubes) and is designed to optimize the X-ray escape from the exit window and maximize thermal dispersion. X-rays are emitted following the

abrupt abrasion of electrons on the anode. The spectrum of the emitted radiation is composed of a continuous component (braking radiation, in German “bremsstrahlung”) and another discrete (“lines”) component (characteristic of the anode material). In the case of sources used in computed tomography (CT), radiogenic tubes must be able to withstand substantial workloads. In these cases, the anode is rotated very quickly so that the electron beam does not always hit the same metal area so as to prevent the overheating of a single disk area and to promote thermal dissipation. X-ray tubes for medical applications are limited to accelerating voltages  Eb. Photoelectric absorption has a low probability to occur on weakly bound electrons. In this case, the interaction can be modelled as a relativistic collision between a photon and a free (i.e., unbound) electron at rest. In most cases, this type of interaction occurs with electrons of external orbitals, and the process is called Compton scattering (or incoherent scat-



hn

1+

hn (1 - cos q ) me c 2 

(1.32)

where me is the mass of the electron. At higher energies, photons can interact with the electric field of the atomic nucleus leading to the production of an electron-positron pair (Fig.  1.8), in agreement with the Einstein’s equation for mass-energy equivalence, E = mc2. This process is allowed only if the initial energy hν of the primary photon is greater than the total rest mass energy of the two output particles:

hn > me + c 2 + me - c 2 = 2me c 2 = 1022 keV (1.33)

The difference between the initial photon energy and the threshold energy is shared between the electron and the positron being created in form of kinetic energy. The nucleus also

1  Fundamentals of Natural and Artificial Radioactivity and Interaction of Ionizing Radiations with Matter

shares a small (in most cases, negligible) fraction of recoil energy, in such a way that the conservation laws for energy and momentum are preserved. All the three interaction processes described above (photoelectric absorption, Compton scattering, and e+e− pair production) result in the emission of one or more charged particles, which in turn can ionize the surrounding atoms via Coulomb-type interactions as explained in the previous section. In this sense, photons in the energy range covered in the context of nuclear medicine and diagnostic radiology are called “indirectly ionizing” radiations. Let us now define some useful parameter for a macroscopic description of the photon beam interaction with matter. The atomic or nuclear cross section, σ, represents the probability for a photon to undergo a given interaction on a single atom (or nucleus). The overall probability of interaction is the sum of the cross sections for all the possible types of interaction:

s tot = ås i



i



(1.34)

17

The cross section is expressed in m2, although a common unit is the barn (1 b = 10−28 m2). The linear attenuation coefficient is obtained by multiplying the total cross section by the number of atoms per unit volume:

m=

NA r s tot  A

(1.35) where A is the mass number, ρ is the material density, and NA is the Avogadro number. The dependence of the linear attenuation coefficient on Z and hν is given by the interaction cross section for the individual interaction types and can be seen in Fig. 1.9 for photon beams interacting in water (top) and lead (bottom). The linear attenuation coefficient (μ) has the dimension of the inverse of a length, so its SI unit is m−1. The parameter μ is a function of the incident photon energy and of the Z and ρ of the target material. When a monoenergetic photon field with initial fluence Φ0 traverses a homogeneous slab of material with linear attenuation coefficient μ and thickness x, the resulting

a

H2O, Zeff=7.51

10

3

10

2

10

1

Total Rayleigh Photoelectric Compton Pair production

µ/ρ (cm2/g)

100

10-1

10-2 Nuclear Medicine

10-3

Diagnostic radiology

Radiotherapy

10-4

10-5

10-6 0 10

101

102

103

104

hν (keV)

Fig. 1.9  Energy dependence of the total and interaction-specific linear attenuation coefficient for photons in water (a) and lead (b) [11]. In lead, the electron binding energies are within the photon energy range, leading to sharp edges in the photoelectric absorption cross section

18

A. Del Guerra and D. Panetta

b 103

102

101

M-edges M1: 3.9 keV M2: 3.6 keV M3: 3.1 keV M4: 2.6 keV M5: 2.5 keV

Pb, Z=82

L-edges L1: 15.9 keV L2: 15.2 keV L3: 13.0 keV

Total Rayleigh Photoelectric

K-edge 88.0 keV

Compton Pair production

µ/ρ (cm2/g)

100

10-1

10-2 Nuclear Medicine

10-3

Diagnostic radiology

Radiotherapy

10-4

10-5

10-6 100

101

102

103

104

hv (keV)

Fig. 1.9 (continued)

­(attenuated) fluence Φ after the slab is given by the BeerLambert law of exponential attenuation (see Fig. 1.10):

F = F0 exp ( - m x ) 

(1.36)

Neutrons can also ionize atoms in an indirect way, but, instead of interacting with atomic electrons, they are more likely to interact with atomic nuclei via elastic or inelastic nuclear scattering. Nuclear reactions often occur when the target nucleus absorbs the neutron with the subsequent emission of various kinds of secondary radiations. The atomic cross section of each type of interaction strongly depends on the energy of the incident neutron, as well as on the target material. For fast neutrons (with kinetic energy in the order of kiloelectronvolt or higher), the dominating interaction is the scattering. In elastic scattering (n,n), the mechanism of indirect ionization comes from the fact that the incident neutron transfers kinetic energy to the recoil nucleus. If instead the collision is inelastic (n,n′), the neutron enters the target nucleus which then re-emits it at a different angle with respect to the original direction. When the neu-

tron enters the target nucleus, this latter one is set to an excited state; the subsequent de-excitation is followed by the emission of a prompt γ photon. For fast neutrons, most of the energy loss of the primary particle is due to elastic scattering. The highest fraction of neutron energy is lost through elastic collision with target particles of approximately the same mass (i.e., hydrogen nuclei). More generally, the average energy loss for elastic collision is given by DE =

2 Ekin A

( A + 1)

2



(1.37)

where A is the mass number of the target atom and Ekin is the initial kinetic energy of the neutron. Hence, the most effective materials for slowing down neutrons (or to moderate them) are those with high hydrogen content. At room temperature, i.e., with mean kinetic energies of kT ≃ 0.025 eV for T = 300 K (where k is the Boltzmann constant), the neutron absorption dominates over the scattering in some materials. The absorption cross section of thermal

1  Fundamentals of Natural and Artificial Radioactivity and Interaction of Ionizing Radiations with Matter

a

neutrons is particularly high for Boron-10. For instance, nuclear reactions like 10B(n,α)7Li are often exploited in a special type of radiation therapy called Boron neutron capture therapy (BNCT).

dx

µ attenuated beam

Incident beam

References x

b

Φ(x)/Φ0

1.0 0.9 0.8 0.7 0.6 Φ=Φ0/2

0.5 0.4

Φ=Φ0/4 Φ=Φ0/8

19

0.3 0.2 0.1 0.0 0.0

0.5

1.0

1.5

x/HVL=1

2.0

2.5

x/HVL=3

3.0

4.0

3.5

µx

x/HVL=2

Fig. 1.10 (a) Operative definition of the linear attenuation coefficient (see text for details). (b) Exponential attenuation curve as a function of the attenuation (μx). The half-value layer (HVL) is the thickness for which the initial photon fluence is halved

Key Learning Points

• In the energy range relevant for medical applications, neutral particles interact with atomic electrons and nuclei of the target material mainly by coherent (Rayleigh) or incoherent (Compton) scattering, photoelectric absorption, or e+/e− pair production. • Some of these interactions lead to the production of secondary charged particles, which in turn are responsible for almost all the ionizations: for this reason, neutral particles are called indirectly ionizing radiations. • Neutrons interact mainly with the nuclei of the target material through elastic or inelastic collisions (in case of fast neutrons) or by neutron capture or absorption (for thermal neutrons).

1. ICRU.  Report 85. Fundamental quantities and units for ionizing radiation (revised). J ICRU. 2011;11(1):1–31. 2. Volterrani D, Erba PA, Mariani G, editors. Fondamenti di medicina nucleare. Tecniche e applicazioni. Milan: Springer; 2010. 3. Firestone RB, Baglin CM, Chu SYF.  Table of Isotopes. 8th ed. New York, NY: Wiley and Sons, Inc; 1999. 4. de Marcillac P, Coron N, Dambier G, Leblanc J, Moalic J. Experimental detection of α-particles from the radioactive decay of natural bismuth. Nature. 2003;422:876–8. 5. Chu SYF, Ekström LP, Firestone RB.  The Lund/LBNL Nuclear Data Search, Version 2.0, February 1999. http://nucleardata. nuclear.lu.se/toi/nucSearch.asp. 6. Podgorsak EB, editor. Radiation oncology physics. A handbook for teachers and students. Vienna: International Atomic Agency; 2005. 7. IAEA. Technical report series no. 465. Cyclotron produced radionuclides: principles and practice. Vienna: International Atomic Energy Agency; 2008. 8. IAEA. Technical document no. 1065, Production technologies for molybdenum-99 and tecnetium-99m. Vienna: International Atomic Agency; 1999. 9. Gould P.  Medical isotope shortage reaches crisis level. Nature. 2009;460:312–3. 10. Jødal L, Le Loirec C, Champion C. Positron range in PET imaging: an alternative approach for assessing and correcting the blurring. Phys Med Biol. 2012;57:3931–43. 11. Boone JM, Chavez AE.  Comparison of x-ray cross sections for diagnostic and therapeutic medical physics. Med Phys. 1996;23:1997–2005.

2

Single-Photon-Emitting Radiopharmaceuticals Federica Orsini, Erinda Puta, Federica Guidoccio, and Giuliano Mariani

Contents 2.1     Overall Background

 22

2.2     General Localization Mechanisms for Single-Photon-Emitting Radiopharmaceuticals

 23

99m 2.3      Tc-Labeled Radiopharmaceuticals 2.3.1  99mTc-Sodium Pertechnetate 99m 2.3.2    Tc-Bisphosphonates 99m 2.3.3    Tc-Diethylene Triamine Penta-Acetic Acid (99mTc-DTPA) 99m 2.3.4    Tc-Mercapto-Acetyl-Triglycine (99mTc-MAG3) 99m 2.3.5    Tc-Dimercaptosuccinic Acid (99mTc-DMSA) 99m 2.3.6    Tc-Radiocolloids 99m 2.3.7    Tc-Mannosyl-DTPA-Dextran (99mTc-Tilmanocept) 99m 2.3.8    Tc-Macroaggregated Albumin (99mTc-MAA) 99m 2.3.9    Tc-Hexa-Methyl-Propylene-AmineOxime (99mTc-HMPAO) 2.3.10 99mTc-Ethylenediylbis-Cysteine-Diethylester (99mTc-ECD) 2.3.11 99mTc-Hexakis-2-Methoxy-2-Isobutyl-Isonitrile (99mTc-Sestamibi) 2.3.12 99mTc-6,9-Bis(2-Ethoxyethyl)-3,12-Dioxa-6,9-Diphospha-Tetradecane (99mTc-Tetrofosmin) 2.3.13 99mTc-Labeled Iminodiacetic Acid (IDA) Derivatives (99mTc-IDA) 2.3.14 99mTc-Sulesomab 2.3.15 99mTc-EDDA/HYNIC-[Tyr3]-Octreotide (TOC) and Other Somatostatin Analogs

 28  29  30  31  32  33  33  36  36  37  38  39  40  41  42  42

2.4     Radiopharmaceuticals Labeled with Indium-111 2.4.1  111In-DTPA-Octreotide (111In-Pentetreotide) 111 2.4.2    In-DTPA

 44  44  45

2.5     Radioiodinated Imaging Agents 2.5.1  123I/131I-Sodium Iodide 123 2.5.2    I-Ortho-Iodo-Hippuric Acid (123I-Hippuran) 2.5.3  Radioiodinated Meta-Iodo-Benzyl-Guanidine (MIBG) 123 2.5.4    I-N-ω-FluoroPropyl-2βCarbomethoxy-3β-(4-Iodophenyl)-nor-Tropane (123I-FP-CIT) 123 2.5.5    I-Iodobenzamide (123I-IBZM)

 45  45  46  47  49  50

201 2.6      Tl-Chloride

 51

67 2.7      Ga-Citrate

 51

2.8     Lung Ventilation Radiopharmaceuticals 2.8.1  99mTc-DTPA Aerosol 99m 2.8.2    Tc-Technegas 2.8.3  Xenon-133

 52  52  53  53

F. Orsini (*) · E. Puta Nuclear Medicine Unit, “Maggiore della Carità” University Hospital, Novara, Italy e-mail: [email protected] F. Guidoccio · G. Mariani Regional Center of Nuclear Medicine, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy © Springer Nature Switzerland AG 2019 D. Volterrani et al. (eds.), Nuclear Medicine Textbook, https://doi.org/10.1007/978-3-319-95564-3_2

21

22

F. Orsini et al. 2.9     Other Radiopharmaceuticals 2.9.1  Perspectives in Molecular Imaging Based on Single-Photon-Emitting Radiopharmaceuticals

 54  54

References

 55

Learning Objectives

• Learn the general definition of “radiopharmaceutical” and “radiolabeling.” • Learn the main physical and chemical characteristics of radionuclides emitting γ-rays commonly employed for single-photon imaging, including 99mTc, 111In, 123I, 131I, 201 Tl, and 67Ga. • Distinguish the different mechanism(s) of localization of the most important radiopharmaceuticals employed in conventional nuclear medicine imaging. • Understand the favorable physical characteristics of 99mTc for gamma camera imaging and of the 99Mo/99mTc generator for distribution logistics. • Become familiar with the wide spectrum of available radiopharmaceuticals commonly used in clinical practice for gamma camera imaging, with special attention to their clinical indications and in vivo pharmacokinetics. • Become familiar with the estimates of radiation dosimetry to normal tissues/organs for each diagnostic radiopharmaceutical. • Understand the concept of “molecular target” and of “theranostics.”

2.1

Overall Background

Nuclear medicine images depict anatomic, functional, and metabolic processes in tissue. To create these images, specific compounds called radiopharmaceuticals are administered to the patient. A radiopharmaceutical generally consists of a biochemical core, which confers certain biological properties that dictate its biodistribution. This biochemical core is chemically linked to the radionuclide emitting the radioactive signal that can be detected from outside the body, usually in the form of γ-rays (also called “photons”); in some cases, it is the ionic form per se of the radionuclide that determines its biodistribution. Imaging with single-photon-emitting radionuclides (radioisotopes of elements) produces both planar images and single-photon emission computed tomography (SPECT) using a gamma camera. Radiopharmaceuticals labeled with positron-emitting radionuclides are used for positron emission tomography imaging (as described in Chap. 3 of this book “Positron-emitting radiopharmaceuticals”), while radiopharmaceuticals emitting predominantly β- or α-particles are used for therapeutic purposes (as described in Chap. 4 of this book).

Radiopharmaceuticals consisting of only the radionuclide in its ionic form usually concentrate in specific target tissues/ organs because either they are radioisotopes of a naturally occurring element of biologic interest (such as thyroid imaging with 123I− or 131I−—which undergoes the same physiologic distribution as nonradioactive native iodine-127) or they share chemical similarities/analogies with native elements of biologic interest (such as the 201Tl+ ion—mirroring the physiologic distribution of the native K+ ion). For more complex radiopharmaceuticals, the radionuclide is linked to the main core/carrier molecule through a chemical process denominated “radiolabeling.” The diagnostic information provided by scintigraphic images derives from the specific distribution of a radiopharmaceutical within the body, usually injected intravenously (less often by inhalation, orally, interstitially, intracavitarily, or intra-arterially). The main parameters derived from serial scintigraphic images are delivery and clearance of the tracer from the organ or tissue. Pathophysiologic changes induce modifications of these parameters (especially retention and/or the absence of retention of the radiopharmaceutical in the tissue/ organ under examination). The radionuclides most frequently used for diagnostic applications in conventional nuclear medicine are the isotopes of technetium, iodine, indium, gallium, and thallium. Their main physical characteristics are summarized in Table 2.1.

Table 2.1  Main physical characteristics of radionuclides used for ­single-photon imaging, arranged according to increasing half-life of physical decay Radionuclide 99m Tc 123 I 111 In 201 Tl 67 Ga 133 Xe 131 I 51 Cra 125 a I

Decay T1/2 6.0 h 13.2 h 2.80 days 3.04 days 3.26 days 5.25 days 8.0 days 27.7 days 59.9 days

Energy of major photon emissions (keV) 141 (89%) 159 (83%); 528 (1%) 171 (90%); 245 (94%) 68–82 (88%); 135 (3%); 167 (10%) 93 (39%); 185 (21%); 300 (17%) 81 (36.5%) 284 (6%); 365 (82%); 637 (7%) 5 (20%); 320 (10%) 27 (14%); 31 (26%); 36 (7%)

Not useful for actual external imaging in humans; used for specific application involving counting of biological samples with well-type γ-counters

a

2  Single-Photon-Emitting Radiopharmaceuticals

Radiopharmaceuticals labeled with 99mTc are used for about 85% of diagnostic in single-photon imaging. This radionuclide is widely employed clinically because: • It can be obtained from commercial 99Mo/99mTc generators in local radiopharmacies. • Many kits are available to produce the radiopharmaceutical on-site with readily available equipment. • The metastable radionuclide produces a 140 keV photon in 88% of its nuclear decays (a conversion electron is produced in 12%). • Its γ-energy (140  keV) is favorable for detection by gamma cameras (whose detectors have excellent efficiency for γ-energies between 100 and 200 keV). • Its relatively short half-life of 6 h represents a good compromise between the quality of images and radiation burden to the patient. Biological-metabolic alterations typically occur at an early stage of disease prior to morphologic alterations in shape, size, structure, or motion of the tissue/organ. On the other hand, this high sensitivity of nuclear medicine investigations for early changes is rarely matched by equally high specificity for a given disease. For instance, the increased metabolism of glucose in a particular region of the body can indicate the presence of neoplastic cells with high proliferative activity, but it may also indicate a condition of intense inflammation due to infection or immunologic attack. Since radiopharmaceuticals are typically used in very small mass amounts (in the order of micro-, nano-, or picomoles), they generally do not cause any disturbance of the biological system under investigation (except for possible radiobiological effects due to the radioactive emission). Quality control procedures applied to nonradioactive drugs are applicable to radiopharmaceuticals. Quality control tests that ensure the purity, potency, product identity, biologic safety, and efficacy of radiopharmaceuticals are mandatory for their clinical use. Moreover, because of the radiation component, radionuclidic and radiochemical purity must also be tested. The equipment and procedures necessary for these quality control tests are described in details in Chaps. 39 and 40 of this book (for single-photon emitting and for positronemitting radiopharmaceuticals, respectively). The quantity of radioactivity administered varies according to the procedure. Examinations based on the use of ionizing radiation must be optimized so as to deliver a radiation dose “as low as reasonably achievable” (ALARA) consistent with the diagnostic goal. Metabolism, biodistribution, and excretion of drugs are different in children from those in adults; the European Association of Nuclear Medicine (EANM) and the Society of Nuclear Medicine and Molecular Imaging (SNMMI) have developed pediatric dosage cards that take into account the age and body weight for determining the activity of radiopharmaceuticals for administration to children [1].

23

The chemical and physical characteristics of a radiopharmaceutical are the main factors determining its accumulation and retention in normal and diseased tissues of the organism. This chapter classifies radiopharmaceuticals according to radionuclide used for labeling and to the main mechanism(s) of tissue localization responsible for the specific distribution properties of the imaging agent.

Key Learning Points

• Radiopharmaceuticals usually consist of a combination of a biologic directing agent and of a radionuclide emitting photons that can be detected outside the body using gamma cameras for imaging. • The diagnostic information provided by radiopharmaceuticals derives from their specific distribution within the body in physiologic and pathologic conditions. • Alterations occurring at the molecular level detectable by functional imaging typically occur at an early stage of disease, prior to morphologic alterations in shape, size, structure, or motion of the tissue/organ. • The most common radionuclide used for singlephoton imaging is 99mTc; other radionuclides employed in conventional nuclear medicine include 131 123 111 I, I, In, 67Ga, and 201Tl. • Since radiopharmaceuticals are typically used in very small mass amounts, they generally do not cause any disturbance to patients, except for possible radiobiological effect. • Quality control procedures applied to nonradioactive drugs must be applied also to radiopharmaceuticals. • The quantity of radiopharmaceutical used in a specific procedure should be “as low as reasonably achievable” (ALARA), especially in pediatric patients. • Radiopharmaceuticals can be classified according to the radionuclide used for labeling and to the main mechanism(s) of tissue localizations.

2.2

 eneral Localization Mechanisms G for Single-Photon-Emitting Radiopharmaceuticals

Localization of most radiopharmaceuticals is not limited to one simple mechanism but also depends on the mode of administration and delivery to the tissue. For some radiopharmaceuticals there is more than one single localization mechanism involved.

24

The principle of compartmental localization or dilution is at the basis of the applications of nondiffusible radiopharmaceuticals. This definition refers to the situation where the molecules of interest are distributed in an enclosed volume (called a compartment). In case of uniform radiopharmaceutical dispersion, the principle of compartmental dilution is exploited for calculation of the whole-body erythrocyte mass (red blood cells labeled with 51 Cr) or of whole-body plasma volume (125I-radioiodinated serum albumin). These methods are based on the principle of dilution of a known amount of radioactive substance in an unknown distribution volume; in particular, by measuring the degree of dilution (i.e., change in concentration) of the amount of radioactivity administered, it is possible to calculate the volume in which the radiopharmaceutical has been diluted. Red blood cells labeled with a radionuclide (e.g., 99mTc) also allow to record in vivo images of the heart at systole and diastole and to measure chamber volume at the end of diastole and systole determinig the left ventricular ejection fraction. In some other conditions, there can be an abnormal leakage of contents from the compartment. For instance, scintigraphy with 99mTc-labeled red blood cells can identify the site of gastrointestinal bleeding as an area of increased radioactivity accumulation at some point in the gastrointestinal lumen—where the radiolabeled erythrocytes do not normally accumulate. Similarly, scintigraphy with a radiopharmaceutical that normally undergoes hepatobiliary clearance and excretion (e.g., 99mTc-mebrofenin) can visualize abnormal biliary leakage into the abdominal cavity, e.g., after biliary surgery. In other conditions, images can show, within a certain tissue/organ, areas of radioactive concentration that are reduced with respect to the surrounding (supposedly normal) tissue/organ. Such areas are often the result of an obstruction in a compartmental space. For instance, ventilation lung scintigraphy after inhalation of the radioactive gas xenon-133 can indicate the presence of an obstruction in the lung airways as an area of reduced/absent radioactivity accumulation in the bronchi/bronchioles beyond the point of obstruction. Macroaggregated albumin (MAA) particles, ranging in diameter from 10 to 90 μm (with >90% of particles ranging from 10 to 40 μm), can be radiolabeled (usually with 99mTc) and injected intravenously to depict the distribution of regional pulmonary blood flow. The particles become mechanically trapped in the capillary bed of the lungs, thus enabling to utilize the principle of microembolization to identify the regions of pulmonary parenchyma with normal perfusion. Regions with reduced perfusion (typically in case of thromboembolic disease) are depicted as areas with decreased/absent radioactivity.

F. Orsini et al.

The mechanical trapping mechanism is also used for the preliminary evaluation of liver perfusion before trans-arterial radionuclide therapy with 90Y-labeled microspheres in patients with inoperable liver malignancies; in this case, 99m Tc-MAA is injected directly into the arterial branches of the hepatic artery. The mechanism of chemisorption is used in bone scans. For example, 99mTc-labeled diphosphonates accumulate at the surface of newly formed calcium hydroxyapatite crystals and in the amorphous bone mineral matrix; their uptake is increased in areas with increased new bone formation (e.g., fracture healing, infection, primary and metastatic tumors) (Fig.  2.1). Chemisorption is quite strong, intermediate between chemical covalent binding and hydrogen binding (adsorption); hence the term chemisorption is used. In addition to chemisorption on the bone surfaces, there can also be chemisorption onto calcium phosphate crystals in dystrophic calcium deposits that precipitate in certain soft tissues, e.g., as a consequence of severe hyperparathyroidism and hypercalcemia. Autologous radiolabeled leukocytes localize at sites of infection based on chemotactic signals. The chemotaxix is the characteristic movement or orientation of an organism or cell (including leukocytes) along a chemical concentration gradient either toward or away from the chemical stimulus and tipically occurs in the sites of inflammation and infection, involving immune cells, blood vessels, and molecular mediators. Phagocytosis is the process whereby microparticles are internalized from the extracellular space into cells, a process that is especially active in macrophages. Although macrophages are ubiquitous cells, they are concentrated mostly in the reticuloendothelial system, especially in the liver, spleen, bone marrow, and lymph nodes, as well as at sites of infection/inflammation. Phagocytosis is facilitated by several properties of the particles, such as size between 2.5 nm and 1 μm, negative charge, and possible opsonization by a class of macromolecules that include the complement fractions C3, C4B, and C5, as well as some α- and β-globulins. Phagocytosis is responsible for the accumulation of 99mTclabeled particles in the nanocolloidal size range (e.g., albumin nanocolloids or sulfur colloid preparations) at different sites depending on the route of administration. In fact, these radiocolloids can be used for imaging the liver, spleen, and bone marrow following systemic administration (i.e., an i.v. injection). On the other hand, radiocolloids injected interstitially migrate through the lymphatic system. This mechanism is the basis of lymphoscintigraphy, a preliminary step to, e.g., radioguided sentinel lymph node biopsy. Many radiopharmaceuticals accumulate in cells through transmembrane transport mechanisms (either passive or active transport). While passive transport does not require any energy expenditure and only involves diffusion of a substance from areas of high concentration to areas of low concentration (concentration gradient), on the other hand,

2  Single-Photon-Emitting Radiopharmaceuticals

a

25

b

Fig. 2.1  Examples of planar whole-body acquisitions of bone scintigraphy with 99mTc-MDP. (a) Normal pattern of skeletal scintigraphy showing physiologic turnover of the mineral matrix of the bone, with-

out any areas of abnormally increased tracer uptake. (b) The presence of numerous focal areas with markedly increased tracer uptake throughout the skeleton in a patient with multiple skeletal metastasis

active transport requires energy (in the form of ATP or an electrochemical gradient of Na+ or H+) for the system to function. Passive transport generally occurs by diffusion, i.e., when a substance moves from an area of high concentration to an area of low concentration. A substance, typically lipophilic, crosses the phospholipid bilayer of cell membranes. Simple diffusion is influenced by pH/ionization state. In fact, many molecules may exist in either a neutral state or as a charged ion, depending on pH. Hence, a molecule may be able to diffuse across a membrane in its non-ionized lipophilic form but cannot diffuse across the same membrane in its ionized hydrophilic form. Membranes have small pores that limit the size of molecules that can cross the membrane (molecular weight 90%), without undergoing tubular reabsorption; the remaining amount of 99mTc-MAG3

O O

N O N

Tc

N S

O

OH O

Fig. 2.8  Chemical and tridimensional structure of 99mTc-MAG3. Color codes: yellow =  99mTc; red = O; white = H; light blue = C; blue = N; olive green = S

2  Single-Photon-Emitting Radiopharmaceuticals

is cleared through glomerular filtration, despite a high fraction of plasma protein binding (80%). There is also a minor fraction of excretion through the hepatobiliary tract, mediated by uptake in hepatocytes that is independent from renal function and mainly attributable to the lipophilic properties of 99mTc-MAG3. Some of the minimal quantities (1–2%) of impurities formed during radiolabeling accumulate in the liver and also undergo hepatobiliary excretion. Because of its high first-pass renal extraction (around 60%) due to the active tubular secretion mechanism, 99mTcMAG3 is particularly appropriate for use in pediatric-age patients, in the evaluation of the transplanted kidney, and in subjects with impaired renal function. In these cases, the high extraction fraction results in good images with a high target-to-background ratio with low injected activity. Thanks to the properties of this tracer, quantitative analysis of the data acquired during dynamic renal scintigraphy allows calculation of the tubular extraction rate (TER), an important parameter of renal function. The main radiation dosimetry estimates to patients following intravenous administration of 99mTc-MAG3 are reported here below, normalized to unit of administered activity [3]: • Effective dose 0.0017  mSv/MBq (urinary bladder emptied 30 min after administration) • Tissues/organs with the highest values of absorbed dose: –– Urinary bladder wall 0.11 mGy/MBq –– Uterus 0.012 mGy/MBq –– Lower large bowel wall 0.0057 mGy/MBq

Key Learning Points



Tc-MAG3 (or betiatide or mertiatide) is a chemical compound undergoing partial glomerular filtration and active tubular secretion. • This radiopharmaceutical is characterized by high renal extraction, and it is commonly employed for dynamic renal scintigraphy and tubular extraction rate (TER) calculation. 99m

2.3.5 99mTc-Dimercaptosuccinic Acid (99mTc-DMSA) Tc-DMSA is a chelated radiopharmaceutical used for renal parenchymal imaging (static renal scintigraphy), which provides information on morphology (malformations, ectopies, agenesis), cortical function, and renal injuries or infections. The precise intrarenal handling of 99mTc-DMSA is still debated. Uptake can occur by tubular reabsorption across

33

either the luminal membrane or directly by extraction from the blood of the peritubular capillaries via the basocellular membrane. Following i.v. injection, 99mTc-DMSA initially distributes in the extracellular fluid, with temporary retention in the liver and spleen. About 90% of the radiopharmaceutical circulates in the blood bound to plasma proteins, which prevents its glomerular filtration (limited to 2–3% of total injected activity). Its kidney extraction is 5% at each passage, and 1 h postinjection, approximately 50% of the administered activity is firmly retained in the proximal renal tubular cells (mainly in the cytoplasm), with a 22:1 cortex/medulla uptake ratio. Maximum renal cortical uptake of the radiopharmaceutical is reached at about 3 h. Static renal scintigraphy acquired 1–3 h after injection provides excellent imaging of the renal cortex. Washout of 99mTc-DMSA is slow, so that 6 h after administration about 35% of injected activity is still retained in the renal cortex. Very late images (6–24  h) are generally acquired only in the event of urinary tract obstruction; at this time point, image quality is still suboptimal, due especially to the late concentration of 99mTc-DMSA in the liver, with an inverse proportion to renal uptake and function. About 50% of the injected activity is still bound to the renal tubules 24 h after administration. The main radiation dosimetry estimates to patients following intravenous administration of 99mTc-DMSA are reported here below, normalized to unit of administered activity [3]: • Effective dose 0.0088 mSv/MBq • Tissues/organs with the highest values of absorbed dose: –– Kidney 0.18 mGy/MBq –– Urinary bladder wall 0.018 mGy/MBq –– Spleen 0.013 mGy/MBq

Key Learning Points



Tc-DMSA is a chelated radiopharmaceutical with high renal extraction and intraparenchymal retention commonly employed for static renal scintigraphy. • The precise intrarenal handling of 99mTc-DMSA is still debated. 99m

99m

2.3.6 99mTc-Radiocolloids A “colloid” is defined as a homogeneous mixture in which one substance of microscopically dispersed insoluble particles is suspended throughout another substance (liquid or gas). To qualify as a colloid, the mixture must be one that does not settle or would

34

take a very long time to settle appreciably. Unlike those in a suspension, colloidal particles cannot be separated out by ordinary filtering or by centrifugation. The dispersed substance alone is called the colloid. Therefore, from the biological point of view, a colloid includes all particles ranging from 1 nm to about 4 μm that are found in the body fluids. These particles are generally due to debris of microorganisms, fragments of cells produced during the physiologic processes of tissue renewal, or products of intestinal absorption (especially fat). Colloids are generally removed from circulation by phagocytosis, except for products of intestinal absorption that undergo their physiological metabolism in the liver or at other metabolic sites. Originally introduced to perform hepatosplenic scintigraphy and bone marrow scintigraphy (after systemic i.v. administration), their most common current use is for lymphoscintigraphy. Following its intradermal or subcutaneous administration, a colloidal tracer is usually cleared from the injection site by lymphatic drainage, allowing visualization of lymph nodes protecting that tissue. In patients with breast cancer, melanoma, or penile tumors, it has become a standard of clinical practice to biopsy the first lymph node (called the sentinel node) draining the region of the tumor. Radiopharmaceuticals with colloidal properties include several types of preparations. The most widely used radiocolloids are 99mTc-nanocolloidal human albumin, 99mTc-sulfur colloid, and 99mTc-antimony trisulfide. Additional radiocolloids employed for lymphoscintigraphy (depending on local availability) include 99mTc-rhenium sulfide nanocolloid, 99mTcsulfide nanocolloid, 99mTc-stannous phytate, 99mTc-tin colloid, and 99mTc-microaggregated human albumin [6]. Despite important differences in the size range of the radiolabeled particles (that dictate variable rates of clearance from the site of interstitial injection) (Fig. 2.9), a choice of a certain radiocolloid preparation over others is based more on local availability than on differences in diagnostic performance (e.g., the success rate in sentinel lymph node identification for radioguided sentinel lymph node biopsy, SLNB). It should be noted that smaller-sized particles migrate faster from the site of interstitial injection through the lymphatic channels, a feature that is more favorable for investigating the pattern of lymphatic circulation, e.g., in patients with edema of the limbs. However, small particles are not efficiently retained in the first lymph node they encounter along a certain pathway of lymphatic drainage (i.e., the sentinel lymph node), and they progress in part to visualize other lymph nodes along the same pathway; this feature complicates the gamma-probe-guided intraoperative search of sentinel lymph node(s). On the other hand, larger particles are retained more efficiently in the sentinel lymph node, but at the expense of a slower migration speed from the site of interstitial injection. 99m Tc-Nanocolloidal human albumin is most frequently used in Europe and represents a good compromise between the speed of migration and lymph node retention. The size of its particles ranges from 5 to 100 nm, with at least 95% of human albumin

F. Orsini et al. HSA microcolloid Rhenium sulfide Stannous fluoride Sulfur colloid (unfiltered) Sulfur colloid (prefiltered) HSA nanocolloid Antimony sulfide Dextran 1

10

100

1000

10000

Ranges of particle sizes, nm

Fig. 2.9  Schematic representation of the ranges of particle sizes (in nm) constituting the main radiocolloids employed for lymphoscintigraphy; the logarithmic scale is used to compensate for the wide range in sizes (reproduced with permission from Erba PA, Bisogni G, Del Guerra A, Mariani G.  Methodological aspects of lymphoscintigraphy: radiopharmaceuticals and administration. In: Mariani G, Manca G, Orsini F, Vidal-Sicart S. Valdés Olmos RA, eds. Atlas of Lymphoscintigraphy and Sentinel Node Mapping  – A Pictorial Case-Based Approach. Milan: Springer; 2013: pp 17–26)

colloidal particles with a diameter ≤80 nm, which easily migrate through the lymphatic system but are not totally retained in the sentinel node. Lymphoscintigraphy will therefore usually display multiple lymph nodes along a certain lymphatic route, an occurrence that can complicate the intraoperative search for the “true” sentinel lymph node using the gamma probe. 99m Tc-Sulfur colloid is commonly used in the USA; the range of particle sizes is quite wide (15–5000 nm) depending on the preparation method, with an average size ranging from 305 to 340  nm. The filtered colloidal form of 99mTcsulfur colloid (particle size ranging from 100 to 220 nm) is used for lymphoscintigraphy. In this case radiocolloids are retained in the sentinel lymph node more efficiently than 99m Tc-nanocolloidal human albumin, although their migration is quite slow, so that a longer time can be required to complete lymphoscintigraphy. 99m Tc-Antimony trisulfide, used mostly in Canada and Australia, has a range of particle size of 3–30 nm. It migrates quite fast from the injection site through the lymphatic system; however, it is less efficiently retained in the sentinel node, so that several lymph nodes can typically be visualized along a single lymphatic draining channel. The use of radiocolloids for liver, spleen, and bone marrow imaging relies on the fact that, following i.v. administration, these radiopharmaceuticals are cleared from the circulation through phagocytosis taking place in macrophages of the endothelial-lymphatic system, which are particularly abundant in the liver, spleen, and reticuloendothelial system of the bone marrow. Lymphoscintigraphy is the first phase of radioguided SLNB. Along the route of lymphatic drainage from the site

2  Single-Photon-Emitting Radiopharmaceuticals

of injection around the tumor, radiocolloids reach lymph nodes. The first lymph node is called the sentinel lymph node; this is the lymphatic station where tumor cells possibly entering lymph vessels and migrating along the route of lymphatic drainage encounter the first lymph node (or nodes) of the lymphatic basin draining that specific tumor region. Different parameters such as specific injection site, particle size, and pathophysiology of local lymphatic circulation affect the speed of lymphatic drainage of radiocolloids. Small particles are drained to the extent of about 40–60%, whereas the larger particles are retained preferentially at the injection site. For instance, when using 99mTc-nanocolloidal albumin, lymph node activity reaches a plateau within 2  h after injection, approximately 3% of the injected activity being retained in all visualized nodes. The ideal radiopharmaceutical for sentinel lymph node biopsy should allow rapid visualization of the first lymph node draining from the tumor/injection site and prolonged retention in the sentinel node, possibly without further migration to higher echelon nodes. In principle, radiocolloids with relatively large-sized particles (between 200 and 300  nm) have efficient retention in the first lymph node encountered along the migration route, with very little progression to higher echelon nodes. Nevertheless, particles of this size migrate quite slowly from the injection site. Since the speed of migration of radiocolloids injected interstitially is inversely related to particle size, some compromise must be found between the speed of lymphatic migration and the degree of phagocytosis of the colloidal particles in lymph nodes. Given the different sizes of radiocolloids successfully used for lymph node mapping, the choice is usually based on local availability. Radiocolloids can also be used as nonabsorbable 99mTclabeled radiopharmaceutical for gastroenteric scintigraphy, including esophageal transit and gastric emptying imaging after oral administration. In this case the radiocolloids are generally suspended in some liquid, semisolid, or solid food (such as water, jelly, or eggs, respectively). Since radiocolloids cannot be absorbed within the gastrointestinal tract, it is possible to obtain functional images of the oral, esophageal, and gastric lumen enabling investigation of swallowing disorders, dysmotilities, or gastroparesis, according to the principle of compartmental localization. An additional use of radiocolloids such as 99mTc-nanocolloidal human albumin has been described, i.e., as inert, small-sized particles for lung ventilation studies, by means of a special nebulizer producing a radioactive droplet aerosol [7]. The main radiation dosimetry estimates to patients following intravenous administration of 99mTc-labeled colloids 100–1000  nm in size (including sulfur colloid, tin colloid, microaggregated albumin, and phytate) are reported here below, normalized to unit of administered activity [3]:

35

• Effective dose 0.0091 mSv/MBq • Tissues/organs with the highest values of absorbed dose: –– Spleen 0.074 mGy/MBq –– Liver 0.071 mGy/MBq –– Gallbladder wall 0.02 mGy/MBq The main radiation dosimetry estimates to patients following intra-tumoral administration in breast cancer patients of 99mTc-labeled small colloids (10  mg/dL, hepatobiliary visualization is poor. In hepatocytes these compounds are rendered water-soluble by conjugation with glucuronic acid and are then excreted into the bile through the hepatocyte bile ducts, ending up ultimately into the intestine. About 8–17% of the injected activity of 99mTc-mebrofenin remains in the circulation 30  min after injection. Approximately 1–9% of the administered activity is excreted in the urine over the first 2 h after injection. In fasting individuals, the maximum liver uptake occurs by 10 min postinjection and peak gallbladder activity by 30–60  min after injection. Only a minor fraction of the injected activity is eliminated in the urine (60% at 48  h), while approximately 15% of the injected activity is excreted through the gastrointestinal tract. Although very rarely, some side effects have been reported, such as headache, tingling, dizziness, increased appetite, and pain at the intravenous injection site.

50

F. Orsini et al.

Since uptake of 123I-FP-CIT is receptor-mediated, attention must be paid to possible interference by other medications/drugs involved in the dopaminergic system that may alter the specific tracer binding; if possible, such medications should be discontinued for at least five half-lives. The main substances that may interfere with the results of the scan with 123I-FP-CIT are amphetamines, benzatropine, bupropion, cocaine, and sertraline. The main radiation dosimetry estimates to patients following intravenous administration of 123I-FP-CIT are reported here below, normalized to unit of administered activity: • Effective dose 0.025 mSv/MBq • Tissues/organs with the highest values of absorbed dose: –– Liver 0.085 mGy/MBq –– Colonic wall 0.059 mGy/MBq –– Spleen and gallbladder wall 0.044 mGy/MBq

Key Learning Points



I-FP-CIT (or ioflupane) is a radioiodinated cocaine analog able to bind with high affinity and selectivity to dopamine transporter (DAT) in the presynaptic endings of basal ganglia. • It is commonly used for brain SPECT when investigating patients with movement disorders and/or neurodegenerative diseases. 123

2.5.5 123I-Iodobenzamide (123I-IBZM) I-IBZM (also known as 123I-iolopride; see Fig. 2.19) targets the D2 postsynaptic receptors in the striatum nucleus (putamen and caudate). The main clinical indication for the use of this radiopharmaceutical is for the differential diagnosis between idio123

H N O HO

NH

CH3 O

pathic Parkinson’s disease (with generally preserved—or even upregulated D2 postsynaptic receptor activity) and other Parkinsonisms (with degeneration of the D2 postsynaptic structures). Additional indications include the evaluation of D2 receptor blockade by neuroleptics, suspected Wilson’s disease and Huntington’s disease, and pituitary adenoma (the presence of D2 receptors may have implications for the therapeutic strategy). Furthermore, scintigraphy with 123I-IBZM has also been performed in patients with extra central nervous system conditions, i.e., to image melanoma lesions. The exact mechanism thereby 123I-IBZM is taken up by melanoma cells is not clear yet. In addition to the increased arterial flow typical of tumor neoangiogenesis, 123I-IBZM might bind to D2 receptors expressed on the melanoma cell membrane, and/or it might interact specifically with some intracellular structures, such as melanin and its precursors. Following i.v. injection, about 75% of 123I-IBZM circulating in the blood is bound to plasma proteins; only about 3–4% of the injected activity crosses the blood-brain barrier due to its lipophilic characteristics, neutral charge, and small size. The majority of circulating 123I-IBZM is in fact quickly converted into two hydrophilic metabolites that are therefore no longer able to diffuse through the blood-brain barrier. The fraction of 123I-IBZM entering the brain is taken up by the postsynaptic D2 receptors present in the striatum dopaminergic system. Optimal target/background ratio is reached about 40 min after injection, plateauing up to over 3 h. The radiopharmaceutical undergoes predominant urinary excretion (about 40% at 24 h and 60% at 48 h after administration). As with other 123I-labeled radiopharmaceuticals, adequate patient’s preparation is recommended to prevent thyroid uptake of radioiodide released during the metabolic degradation of the tracer. The main radiation dosimetry estimates to patients following intravenous administration of 123I-IBZM are reported here below, normalized to unit of administered activity [12]: • Effective dose 0.034 mSv/MBq • Tissues/organs with the highest values of absorbed dose: –– Thyroid 0.16 mGy/MBq –– Urinary bladder wall 0.07 mGy/MBq –– Lower large bowel wall 0.064 mGy/MBq

CH3

123I

Fig. 2.19  Chemical and tridimensional structure of 123I-IBZM. Color codes: magenta = 123I; red = O; white = H; light blue = C; blue = N

Key Learning Points



I-IBZM (or iolopride) targets the D2 postsynaptic receptors of the basal ganglia. • It is used for brain SPECT when investigating patients with movement disorders. 123

2  Single-Photon-Emitting Radiopharmaceuticals

2.6

201Tl-Chloride

The monovalent cation Tl+ has biochemical characteristics similar to those of the K+ ion. Following i.v. injection, 201Tl is rapidly cleared from the blood because of diffuse uptake into cells of all organs and tissues, with a pattern of distribution that is largely determined by the magnitude of regional blood flow. Therefore, in excitable cells (typically neurons and muscle cells), 201Tl undergoes the continuous transmembrane flux that physiologically occurs along an electropotential gradient during depolarization (passive transport) and subsequent repolarization (an energy-dependent process linked to the Na+/K+ pump). These sustained transmembrane exchanges explain the high intracellular concentration of Tl+ ions, mimicking the distribution of the K+ ions. Because of the typical high accumulation in myocardial cells (that physiologically undergo continual, sequential depolarization, and repolarization), 201Tl has been used to investigate myocardial perfusion; in fact, myocardial uptake increases proportionally with perfusion, and prolonged retention depends on the integrity of cell membrane, hence on myocardiocyte viability. Since 201Tl is not trapped in myocytes or in other tissues, redistribution in rest condition of 201 Tl occurs over several hour following administration at the peak of a stress test. This redistribution process renders the 201 + Tl ion available for possible myocardial accumulation at rest in regions that were ischemic when 201Tl was injected at peak stress, thus allowing redistribution images to be acquired that are fairly independent of perfusion and mainly reflect viability. Besides its uptake in myocardial tissue, the 201Tl+ ions tend to accumulate in all metabolically active cells, including cancer cells. 201Tl-Chloride has therefore been used for tumor imaging as an indicator of cell viability, particularly for discriminating viable tumor tissue from fibrous scar tissue after surgery and/or radiation therapy (especially in patients with brain gliomas) and also for parathyroid scintigraphy (where this agent has however been completely replaced with 99m Tc-sestamibi). Following i.v. administration, about 85% of 201Tl+ ions are cleared from the blood during a single circulation, so that about 2 min after administration  5 cm or for any hemorrhagic cyst in women in late menopause. Clot in hemorrhagic cysts may mimic a solid nodule, and, if visualized, a MRI exam should be performed. The CT findings of cystic mature teratoma are a hypo-­ attenuating unilocular mass near to fluid density in which a mural nodule (Rokitansky nodule) or calcifications can be observed. The presence of fat tissue (negative density values on ROI measurement) is pathognomonic for mature cystic teratoma diagnosis. The other indeterminate cystic lesions of the adnexa are cystadenoma, cystadenofibroma, mucinous cystadenocarcinoma, and serous ovarian cystadenocarcinoma. Cystadenoma and cystadenofibroma are benign ovarian neoplasms that can be serous (uniloculated) or mucinous (multiloculated). When these lesions are bigger than 7 cm in size, have  multiple septa and demonstrable contrast enhancement on CECT, mural nodules showing contrast enhancement, and  vascularized thick walls, they must be considered as having a high likelihood of neoplastic transformation. Mucinous ovarian cystadenocarcinoma and serous ovarian cystadenocarcinoma are the malignant variant. Advanced-­stage ovarian tumors are often associated with the presence of ascites and peritoneal carcinomatosis, characterized by the thickened appearance of peritoneal sheaths or with small parietal nodules with demonstrable contrast enhancement. Metastatic lesions of the ovaries are more often solid (Krukenberg metastases) and bilateral, but cystic metastases can also occur.

14  Essentials of CT Image Interpretation

14.17.1 Primary Bone Lesions

Key Learning Points

• Ovarian lesions have some characteristic CT patterns that can aid to categorize the lesions as a simple cyst, hemorrhagic cyst, mature cystic teratomas (dermoid cyst), and other indeterminate cystic lesions. • Simple cystic lesion management is based on dimensional criteria and stratified for risk in accordance with premenopausal or postmenopausal status. • The CT findings of cystic mature teratoma are a hypo-attenuating unilocular mass near to fluid density in which a mural nodule (Rokitansky nodule) or calcifications can be observed. • The presence of fat tissue (negative density values on ROI measurement) is pathognomonic for mature cystic teratoma diagnosis. • Metastatic lesions of the ovaries are more often solid (Krukenberg metastases) and bilateral, but cystic metastases can also occur.

14.17 Bone Lesions Bone lesions can be divided into bone neoplasms, bone infections, bone-related lymphoproliferative disorders, bonerelated endocrinological disorders, and miscellaneous. Bone neoplasms can be primary or secondary. Primary bone tumors can be classified in relationship to primary or secondary and with respect to the type of tissue: bone-, cartilage-, or fibrous forming tumors, bone marrow tumors, and miscellanea. Fig. 14.24  Sclerotic primary bone tumor. NECT on coronal plane (a) shows a large sclerotic lesion in the diaphysis of the femur with a “sunburst” periosteal reaction (white arrowhead): the lesion resulted in osteosarcoma at biopsy. NECT on coronal plane (b) exhibit a large lesion with multiple calcifications and scalloped margins (red arrowhead). Low-grade chondrosarcoma was diagnosed at biopsy

313

a

The differential diagnosis of a primary bone lesion is heavily influenced by the age of the patient and the location. Most likely malignant findings at CT are periosteal reaction (Codman’s triangle), sunburst or lamellated periosteal reaction, cortical destruction, and soft tissue involvement. CT benign findings of periosteal reaction are well-defined margins and benign periosteal reaction (thick, continuous, or wavy periosteal thickening). Most radiological characteristics could overlap in many cases, so osteolytic or sclerotic type, age of patient, and site of the lesions can reduce the differential diagnosis. Bone tumor may appear as predominantly lytic lesions if bone destruction is predominant or sclerotic if new bone formation or tumor matrix calcifications are present. Osteolytic lesions include the following: Fibrous dysplasia or fibrous cortical defect, osteoblastoma, giant cell tumor, aneurysmal bone cyst, chondroblastoma or chondromyxoid fibroma, hyperparathyroidism (brown tumor), non-ossifying fibroma (NOF), enchondroma or eosinophilic granuloma, and simple (unicameral) bone cyst. Sclerotic lesions include the following: When solitary, the differential diagnoses include enostosis (bone island), osteosarcoma, chondrosarcoma, calcifying enchondroma, osteoblastoma, osteoid osteoma, Paget’s disease, callus after fracture, and chronic osteomyelitis (Fig. 14.24). When sclerotic lesions appear as multifocal, the differential diagnoses include the metastasis from prostate and breast cancer, bone islands, hemangiomas (especially when located in the spine), multiple infarct (drugs related), and Paget’s disease. b

314

D. Caramella et al.

14.17.2 Secondary Bone Lesions Skeletal metastases account for 70% of all malignant bone neoplasms. Lung cancer, breast cancer, renal cell carcinoma, and prostate cancer account for approximately 80% of all skeletal metastases. Skeletal metastases tend to be multiple when present. The lymphoproliferative disorder may involve the bone

a

usually as part of a disseminated process presenting as lytic infiltrative lesions (with permeative pattern) or bone sclerosis that may mimic other primary and secondary bone neoplasms on CT imaging. The pelvis and spine are commonly involved. Skeletal metastases have three different patterns, depending on the preponderant process of bone resorption (osteolytic) or bone formation (osteoblastic) (Fig. 14.25) or mixed:

b

c

Fig. 14.25  Bone metastases. Sclerotic metastases from prostate carcinoma on dorsal and lumbar spine (a). Osteolytic metastases on dorsal spine (b) and sacrum (c) from lung adenocarcinoma

14  Essentials of CT Image Interpretation

Osteolytic metastases are most likely due to thyroid cancer, lung cancer, renal cell cancer, gastrointestinal carcinomas, melanoma, hepatocellular carcinoma, and squamous cell carcinoma of the skin. Sclerotic metastases are most likely due to prostate carcinoma, breast carcinoma (may be mixed), transitional cell carcinoma (TCC), carcinoid, medulloblastoma, neuroblastoma, mucinous adenocarcinoma of the gastrointestinal tract (e.g., colon carcinoma), and lymphoma. Mixed lytic and sclerotic metastases are related to breast carcinoma and lung carcinoma; metastases from cervical cancer and testicular tumors are typically lytic but rarely can be mixed. Prostate carcinomas are typically sclerotic but can be mixed in a small fraction of patients.

14.17.3 Miscellaneous Bone Lesions Paget’s disease is a common, chronic bone disorder that leads to bone deformity. Paget’s disease usually presents with  all three types of bone alteration (sclerotic, lytic, and mixed) due to different amounts of osteoclastic or osteoblastic activity at different stages of disease: lytic form (initial active moment), mixed form (active form), and sclerotic form (late stage). When the spine is involved, the typical cortical thickening and sclerosis of subchondral and mural bone gives a “picture frame” appearance of the vertebral bodies. When the skull is involved, in the acute phase osteoporosis circumscripta (lytic lesion) may be observed. In long bones, the classic sign is the “candle flame sign” that is an area of a V-shaped lucency extending from the subchondral space toward the diaphysis. Infectious disorders: osteomyelitis is an infection of the bone most likely due to bacterial pathogens (pyogenic) or other pathogens such as tuberculosis, syphilis, or fungus. Osteomyelitis can be acute or chronic. In the  acute phase (after 2 weeks), the most important CT findings are bone marrow lucency areas, cortical thickening with periosteal reaction (periostitis), and endosteal scalloping. In the chronic phase the most relevant sign is sequestrum. The cloaca sign may also be observed. Involvement of the spine (spondylitis) by tuberculosis is also called Pott disease. Usually, both the vertebral body and the disk area are simultaneously affected  by the disease. The  most relevant findings  on CECT are irregularity of affected vertebrae at the level of the end plates (initial phase), while extension into paravertebral soft tissue is  more frequently observed in the late phase. Vertebra plana is a late finding in which the vertebra has completely lost its anterior and posterior height.

315

Key Learning Points

• The differential diagnosis of a primary bone lesion is heavily influenced by the age of the patient and the location. Most likely malignant findings on CT are periosteal reaction (Codman’s triangle), sunburst or lamellated periosteal reaction, cortical destruction, and soft tissue involvement. • Most radiological characteristics could overlap in many cases, so osteolytic or sclerotic type, age of patient, and site of the lesions can reduce the differential diagnosis. • Skeletal metastases account for 70% of all malignant bone neoplasms. Lung cancer, breast cancer, renal cell carcinoma, and prostate cancer account for approximately 80% of all skeletal metastases. Skeletal metastases tend to be multiple when present. • Osteolytic metastases are most likely due to thyroid cancer, lung cancer, renal cell cancer, gastrointestinal carcinomas, melanoma, hepatocellular carcinoma, and squamous cell carcinoma of the skin. • Sclerotic metastases are most likely due to prostate carcinoma, breast carcinoma (may be mixed), transitional cell carcinoma (TCC), carcinoid, medulloblastoma, neuroblastoma, mucinous adenocarcinoma of the gastrointestinal tract (e.g., colon carcinoma), and lymphoma. • Mixed lytic and sclerotic metastases are related to breast carcinoma, lung carcinoma (typically lytic but rarely can be mixed), carcinoma of the cervix, and testicular tumors. Prostate carcinomas are typically sclerotic but mixed in a small fraction of patients. • Paget’s disease is a common, chronic bone disorder that leads to bone deformity. Paget’s disease usually presents with  all three types of bone alteration (sclerotic, lytic, and mixed) due to different amounts of  osteoclastic or osteoblastic activity at different stages of disease.

14.18 Abdominal Lymphadenopathies CT interpretation criteria of lymphadenopathies: CT imaging for the assessment and characterization of pathological lymph node disease takes into account the size, the number, the morphology, the margins, the attenuation characteristics, and the enhancement after contrast. Increase in size is not always pathologic, because often reactive nodes are bigger than 10 mm of transverse diam-

316

eter (considered as benign limit in size). Hyper-attenuating lymphadenopathy on NECT can be observed in Kaposi sarcoma, Castleman disease, carcinoid, and angioimmunoblastic lymphadenopathy. Calcifications are present in tuberculosis, sarcoidosis, lymphoma treated with radiotherapy, papillary thyroid carcinoma, and breast cancer. Lymph node metastatic disease commonly occurs along lymphatic pathways. For example, breast cancer tends to metastasize to axillary lymph nodes and rectal cancer to lymph nodes of the  inferior mesenteric artery and mesorectal space.

Key Learning Point

• CT imaging for the assessment and characterization of pathological lymph node disease takes into account the size, the number, the morphology, the margins, the attenuation characteristics, and the enhancement after contrast.

Further Reading Bernardino ME. Computed tomography of calcified liver metastases. J Comput Assist Tomogr. 1979;3:32–5. Carter BW, Benveniste MF, Madan R, Godoy MC, de Groot PM, Truong MT, Rosado-de-Christenson ML, Marom EM. ITMIG classification of mediastinal compartments and multidisciplinary approach to mediastinal masses. Radiographics. 2017;37(2):413–36. Chen IY, Kats DS, Jeffrey RB, et al. Do arterial phase helical CT image improve detection or characterization of colorectal liver metastases? J Comput Assist Tomogr. 1997;21(3):391–7. El-Sherief AH, Lau CT, Wu CC, Drake RL, Abbott GF, Rice TW.  International association for the study of lung cancer (IASLC) lymph node map: radiologic review with CT illustration. Radiographics. 2014;34(6):1680–91. Erasmus JJ, Connolly JE, McAdams HP, Roggli VL. Solitary pulmonary nodules: Part I. Morphologic evaluation for differentiation of benign and malignant lesions. Radiographics. 2000a;20(1):43–58.

D. Caramella et al. Erasmus JJ, McAdams HP, Connolly JE. Solitary pulmonary nodules: Part II.  Evaluation of the indeterminate nodule. Radiographics. 2000b;20(1):59–66. Grazioli L, Olivetti L, Fugazzola C, et al. The pseudocapsule in hepatocellular carcinoma: correlation between dynamic MR imaging and pathology. Eur Radiol. 1999;9:62–7. Greenspan A, Jundt G, Remagen W. Differential diagnosis in orthopaedic oncology. Philadelphia: Lippincott Williams & Wilkins, c2007. (2006) ISBN:0781779308. Helvie MA, Rebner M, Sickles EA, et al. Calcifications in metastatic breast carcinoma in axillary lymph nodes. AJR Am J Roentgenol. 1988;151(5):921–2. Herts BR, Megibow AJ, Birnbaum BA, et  al. High-attenuation lymphadenopathy in AIDS patients: significance of findings at CT. Radiology. 1992;185(3):777–81. Hoang JK, Vanka J, Ludwig BJ, Glastonbury CM. Evaluation of cervical lymph nodes in head and neck cancer with CT and MRI: tips, traps, and a systematic approach. AJR Am J Roentgenol. 2013;200(1):W17–25. Ichikawa T, Federle MP, Grazioli L, et al. Fibrolamellar hepatocellular carcinoma: imaging and pathologic findings in 31 recent cases. Radiology. 1999;213:352–61. Itai Y, Matsui O.  Blood flow and liver imaging. Radiology. 1997;202:306–14. Jeung MY, Gangi A, Gasser B, Vasilescu C, Massard G, Wihlm JM, Roy C.  Imaging of chest wall disorders. Radiographics. 1999;19(3):617–37. del Pilar Fernandez M, Redvanly RD. Primary hepatic malignant neoplasms. Radiol Clin N Am. 1988;36:333–48. Lucey BC, Stuhlfaut JW, Soto JA. Mesenteric lymph nodes: detection and significance on MDCT. AJR Am J Roentgenol. 2005;184(1):41–4. Nasseri F, Eftekhari F.  Clinical and radiologic review of the normal and abnormal thymus: pearls and pitfalls. Radiographics. 2010;30(2):413–28. Nickell LT Jr, Lichtenberger JP 3rd, Khorashadi L, Abbott GF, Carter BW.  Multimodality imaging for characterization, classification, and staging of malignant pleural mesothelioma. Radiographics 2014;34(6):1692-1706. Nishino M, Ashiku SK, Kocher ON, Thurer RL, Boiselle PM, Hatabu H.  The thymus: a comprehensive review. Radiographics. 2006;26(2):335–48. Truong MT, Ko JP, Rossi SE, Rossi I, Viswanathan C, Bruzzi JF, Marom EM, Erasmus JJ. Update in the evaluation of the solitary pulmonary nodule. Radiographics. 2014;34(6):1658–79. Whitten CR, Khan S, Munneke GJ, Grubnic S. A diagnostic approach to mediastinal abnormalities. Radiographics. 2007;27(3):657–71.

Essentials of MR Image Interpretation

15

Davide Caramella and Fabio Chiesa

Contents 15.1      Introduction

 317

15.2      Signal Characteristics of Tissue Components in MR Images 15.2.1  T1-Weighted Images 15.2.2  T2-Weighted Images 15.2.3  Water-Fat Signal Decomposition, Fat Suppression, and Dual-Phase Chemical Shift Imaging 15.2.4  Diffusion-Weighted Imaging 15.2.5  Contrast-Enhanced Images

 318  318  318  319  320  320

15.3      Fundamentals of Imaging Interpretation: Common Patterns of Disease 15.3.1  Fluid-Containing Lesions 15.3.2  Signal of Blood-Containing Lesions 15.3.3  Fat-Containing Lesions 15.3.4  Solid Lesions

 321  321  322  322  324

15.4      Essentials of Differential Diagnosis by Anatomical Region 15.4.1  Brain Lesions 15.4.2  Neck Masses 15.4.3  Evaluation of Mediastinum and Pleura 15.4.4  Evaluation of Abdomen and Pelvis

 325  325  329  330  333

Further Reading

 350

Learning Objectives

• Understand how to distinguish various tissue components based on signal intensity on T1-weighted and T2-weighted images. • Describe the most common causes of signal alterations on diffusion-weighted images. • Define an approach to lesion characterization based on identification of four main internal components (fluid, fat, blood, and solid tissues).

D. Caramella (*) Diagnostic and Interventional Radiology, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy e-mail: [email protected] F. Chiesa Unit of Radiology, ASL 5 “Spezzino”, Sarzana, La Spezia, Italy

• Correlate the most common signal alterations in brain lesions with pathology. • Correlate the most typical imaging features of cystic and solid lesions in different anatomical regions with pathology. • Describe the most common signal alterations in the musculoskeletal system.

15.1 Introduction The definition “hybrid imaging” refers to the combination of cross-sectional imaging techniques used in radiology  (CT and MRI) with radionuclide imaging used in nuclear medicine. PET/MRI is a rapidly evolving hybrid imaging technique being increasingly employed in clinical practice and may represent the most challenging modality for nuclear medicine specialists. In the simple PET/CT examination,

© Springer Nature Switzerland AG 2019 D. Volterrani et al. (eds.), Nuclear Medicine Textbook, https://doi.org/10.1007/978-3-319-95564-3_15

317

318

low-­dose unenhanced CT images are acquired for both attenuation correction and anatomical localization, with a limited contribution for characterizing lesions detected by tracer uptake in the PET scan. In contrast, even the most generic whole body  PET/MRI hybrid imaging, without specific organ-­ targeted or contrast-enhanced sequences, requires interpretation of multiple images including T1-weighted, T2-weighted, and diffusion-weighted images. In the first part of this chapter, the appearance of various tissue components on pulse sequences commonly used in MR examinations will be described, and the most common imaging patterns of disease  –­ classified on the basis of prevalent tissue composition –­ will be illustrated. In the second part, imaging findings useful for differentiating pathological processes in various organs will be discussed, with particular emphasis on mass lesions commonly evaluated on MRI.

15.2 S  ignal Characteristics of Tissue Components in MR Images 15.2.1 T1-Weighted Images Pulse sequence parameters of T1-weighted acquisitions (short TR and TE) create images with contrast between fluids and tissues, both normal and pathological, to create images, based on differences in T1 relaxation characteristics (Table  15.1). Shorter T1 relaxation times correspond to a higher signal intensity on T1-weighted images. Normal tissues that show the shortest T1 relaxation times are composed of molecular species capable of rapid exchange of absorbed radio-frequency energy with the surrounding medium. This is usually achieved through dipolar interactions in a thermodynamically irreversible process generating heat. The most important molecular entities with these properties are: –– Fatty acids, the predominant biochemical constituents of mature adipose tissue –– Water solutions containing dispersed macromolecules, more often represented by proteinaceous and/or mucinous substances that slow random movements of water molecules whose tumbling rate matches the Larmor frequency –– Substances with paramagnetic properties, including gadolinium-­based contrast media, specific blood degradation products, and some metallic atoms, such as copper and manganese. Water solutions with no appreciable proteinaceous or paramagnetic content show very low signal intensity on T1-weighted images. Parenchymal organs are composed of a mixture of different tissues, which in turn contain a wide range of intra-

D. Caramella and F. Chiesa Table 15.1 Signal intensities of different body components on T1-weighted and T2-weighted images Signal intensity High

Intermediate

Low Signal void

T1-weighted Fat, highly proteinaceous fluids, gadolinium, methemoglobin, other paramagnetic substances Most parenchymal organs, muscle tissue, proteinaceous fluids Clear fluids, tendons, ligaments, fibrous tissue Calcium, cortical bone, air

T2-weighted Clear fluids (e.g., CSF, bile, urine), fat (in TSE sequences) Most parenchymal organs, muscle tissue Tendons, ligaments, fibrous tissue Calcium, cortical bone, air, fast-­ flowing blood

cellular and extracellular molecules and variable amounts of free water. For this reason all parenchymal organs demonstrate lower T1 signal intensity than fat, with organ-specific differences determined by complex and often incompletely understood biochemical and structural factors. Muscle tissue has intermediate signal on T1-weighted images comparable to that of some parenchymal organs. Dense fibrous tissue, the main constituent of tendons, ligaments, and mature fibrotic scars, exhibits very low signal intensity on T1-weighted images owing to the low content of free water and strong internal dipolar interactions. Similarly, cortical bone and dense calcium deposits present as signal voids due to the very low density of resonating protons. Gaseous components show no appreciable signal. Vascular structures usually appear hypointense on T1-weighted images. In specific pulse sequences called “time of flight” a phenomenon known as flow-related enhancement can be exploited to generate angiographic images in which flowing blood shows a markedly high signal intensity.

15.2.2 T2-Weighted Images Pulse sequence parameters for acquisition of T2-weighted images (long TR and TE) are designed to create image contrast between fluids and tissues, both normal and pathological, based on differences in T2 relaxation time (Table 15.1). Longer T2 relaxation times correspond to a higher signal on T2-weighted images. Molecular tumbling rate and the consequent frequency of dipolar interactions are the main factors influencing the T2 relaxation mechanism. Free water protons, as found in physiological bodily fluids like bile, urine and cerebrospinal fluid, have the longest T2 relaxation time  and thus show the highest signal. In

15  Essentials of MR Image Interpretation

pulse sequences currently used in clinical practice for obtaining T2-weighted images (such as  HASTE, singleshot TSE, and FSE) adipose tissue, mainly composed of fatty acids, shows hyperintense signal, which is  slightly lower than water but higher than parenchymal organs. T2-weighted sequences with very long TE can be used to image viscera containing nearly static fluid, with a common example being magnetic resonance cholangiopancreatography (MRCP). Parenchymal organs demonstrate low to intermediate signal intensity on T2-weighted images. Organ-specific differences are due to variations in tissue composition, especially free water content and concentration of mineral elements such as iron. Muscle tissue shows lower signal intensity than most parenchymal organs on T2-weighted images. Similar to T1-­ weighted images, fibrous tissue, cortical bone, and dense calcifications appear dark. Gaseous components exhibit no signal. Substances with superparamagnetic properties and highly concentrated paramagnetic molecules cause local magnetic field inhomogeneities that accelerate T2 relaxation processes and cause a loss of signal on T2-weighted images. The most notable examples are hemosiderin, the main storage form of iron in tissues, deoxyhemoglobin packed in red blood cells, concentrated gadolinium normally excreted in the urinary tract, and by-products of melanin metabolism in melanocytic tumors. Vascular structures can show variable signal on T2-weighted images depending on flow characteristics. Rapid flowing blood in arterial vessels usually appears dark because of a flow-related phenomenon denominated “flow void”. However, arterial vessels may show high signal intensity similar to static fluid because of random synchronization of acquired data with the  diastolic phase of cardiac cycle. This finding can be typically seen in the aorta and other large arteries running perpendicular to the imaging plane on rapidshot sequences. In fluid-attenuated inversion recovery (FLAIR) sequences, signal of static fluids is nulled by precisely timed inversion pulses. FLAIR sequences are used in brain MRI to suppress cerebrospinal fluid signal in order to increase the conspicuity of parenchymal lesions. Gradient-echo pulse sequences with relatively long TE are weighted for a parameter named T2* (“T2 star”) and are  very sensitive to magnetic field inhomogeneities. T2*weighted sequences are  useful to identify small calcifications and superparamagnetic deposits such as hemosiderin, often not detectable on T2-weighted sequences. On gradientecho T2*-weighted images, calcium and iron deposits appear as very low signal intensity areas with exaggerated size compared to their real volume (“blooming” effect).

319

15.2.3 Water-Fat Signal Decomposition, Fat Suppression, and Dual-Phase Chemical Shift Imaging Fat is predominantly composed of medium and long chain fatty acids containing numerous hydrogen atoms linked to a carbon framework which is responsible for the characteristic spectral behavior (difference in precession frequency compared to water) called “chemical shift.” This physical property can be exploited by using particular pulse sequences to achieve specific effects in image contrast. In Dixon technique, water and fat signal is decomposed to generate three-dimensional images containing signal from a single proton species (“water-only” and “fat-only” images) or a combination of the two. This method is commonly applied to three-dimensional gradient-echo sequences for generating attenuation maps used for correction of PET transmission scan in hybrid imaging. On fat-saturated sequences and STIR sequences, the signal from fat is suppressed by spectral selective saturation and inversion recovery techniques respectively. On T1-weighted images, fat saturation is commonly applied to three-dimensional fast gradient-echo sequences for breath-­hold  studies of the chest and abdomen before and after administration of gadolinium-based contrast agents. On T2-weighted fat-saturated sequences, nulling of fat signal increases the contrast resolution of other tissue components and increases the conspicuity of fluid-containing structures. STIR sequences achieve similar results with more complex effects on tissue contrast due to the T1 dependency of inversion recovery technique. The main practical difference between inversion recovery and fat saturation techniques is the low susceptibility of the former to magnetic field inhomogeneities. STIR sequences are therefore advantageous in the study of  anatomically complex regions such as the neck and whole body imaging of the skeletal system. In MR imaging protocols for the evaluation of thoracic and abdominal regions, the most commonly used pulse sequence exploiting the chemical shift effect is dual-phase T1-weighted gradient echo. Two different types of images are obtained at specific TE, depending on magnetic field intensity. In the first, named “out-of-phase” image, water and fat null each other if equally represented in the same image voxel. In the second, named “in-phase” image, water and fat contained in the same voxel combine  their signal. Dualphase chemical shift imaging with in-phase and out-of-phase acquisitions allows recognition of microscopic quantities of fat that occupy less than a voxel, undetectable by selective fat suppression techniques that act on voxels containing predominantly fat. These sequences may also be used to detect substances typically associated with susceptibility artifacts such as iron and calcium deposits; the resulting blooming

320

D. Caramella and F. Chiesa

effect is more conspicuous on “in-phase” compared to “out-­ –– High cellularity that correlates microscopically with the of-­phase” images due to the longer TE. presence of many cellular barriers that hinder water diffusion, as found in many malignant and some benign tumors. 15.2.4 Diffusion-Weighted Imaging –– Intracellular edema with relative reduction of extracellular volume as seen in cerebral ischemia. Diffusion-weighted sequences are obtained with the applica- –– Fluid collections with a highly compartmentalized struction of symmetrical gradient waveforms before and after a ture like purulent material in abscesses and clotted blood spin-echo refocusing pulse with an echo planar data readout. in hematomas. The extrinsic parameter “b-value” determines the sensitivity to molecular diffusion phenomena: a b-value of 0 means no diffusion weighting, while progressively increasing b-values 15.2.5 Contrast-Enhanced Images correspond to higher diffusion sensitivity. In the typical diffusion-­weighted imaging protocol, two or three b-values Gadolinium-based contrast agents are water-soluble moleare set (from 0 up to a maximum b-value of 1000 for most cules with strong paramagnetic properties that shorten applications), and apparent diffusion coefficient (ADC) T1 relaxation time of fluid compartments in which they dismaps are generated. tribute, resulting in signal enhancement. Immediately after Tissues or pathological processes characterized by a intravenous administration, the plasma concentration of restricted diffusion of water molecules show high signal gadolinium rapidly rises and reaches a peak in arterial intensity on high b-value diffusion-weighted images. On the blood vessels followed by a gradual decline as contrast other hand, free water (as found in simple fluid-containing passes in the intravascular compartment and then diffuses structures) appears hyperintense on images with a b-value of in the extracellular space. Multiphase dynamic evaluation zero, which are predominantly T2-weighted, and shows of contrast enhancement is commonly performed in MR no signal on high b-value images. In some cases, lesions with imaging protocols using fat-saturated three-dimensional high signal intensity on images with a b-value of zero may T1-weighted gradient-­echo sequences that can be acquired appear bright on moderate to high b-value images even if in less than 20  s during a single breath-hold. Turbo spinthey do not significantly restrict diffusion, a phenomenon echo T1-weighted sequences are typically used in static known as “T2 shine-through”. ADC maps synthetically and anatomical regions such as the head, neck, spine, pelvis, quantitatively convey the information of tissue diffusion and limbs. characteristics at different b-values and prove very useful to Images acquired during the maximum arterial concentraavoid the pitfall of T2 shine-through effect. tion of gadolinium are used in contrast-enhanced MR angiAt low b-values (b 75%) with 75% of patients with 50% regression), especially in the subgroup of patients with Breslow thickness between 0.75 and 0.99 mm. It may also be considered in patients where exact thickness cannot be reliably assessed because of inadvertent shaving or cauterization before definitive biopsy of the suspected lesion. SLNB can be considered also for melanomas with >4 mm Breslow thickness, in order to obtain more accurate information and for local control. Contraindications for SLNB include poor general health status, local or systemic spread of disease ascertained by other diagnostic procedures, and prior extensive surgery in the region of the primary tumor or of the lymph node basin. In case of concurrent primary melanoma and satellitosis or “in-transit” metastases, SLNB should not be considered due to the fact that these patients are staged as having stage III disease [28, 43].

Key Learning Points

• Sentinel lymph node biopsy identifies about 20% of patients with clinically occult nodal metastasis at diagnosis and avoids futile surgery in patients with negative sentinel lymph node(s).

373

• There might be a non-negligible rate of false-­ negative sentinel lymph node biopsies in patients with cutaneous melanoma. • Clinical management based on sentinel lymph node findings prolongs disease-free survival and melanoma-­specific overall survival in patients with intermediate-thickness cutaneous melanoma, whereas evidence is not stringent in patients with higher Breslow thickness. • Sentinel lymph node biopsy is indicated in melanoma patients with clinically lymph node-negative lymphatic stations and intermediate Breslow thickness (1.01–4 mm). • Controversy remains when the primary melanoma is 4  mm; different factors should be considered in these patients, based on histology of the primary tumor.

16.4 R  adioguided Sentinel Lymph Node Biopsy in Head and Neck Cancers 16.4.1 The Clinical Problem New cases of oral cavity, pharynx, and larynx cancers in the United States in 2017 were estimated at about 63,000 (73% men, 27% women), 27% of them being located in the pharynx, 26% in the tongue, 21% in the mouth, 21% in the larynx, and 5% at other sites. These tumors account for about 3.5% of all new cancers cases, and about 13,350 deaths were expected for these cancers in 2017 [44]. Squamous cell carcinoma (or its variants) is the most frequent histologic type of these tumors (over 90% of the cases). Metastasis to neck lymph nodes is a major determinant for the prognosis of oral, oropharyngeal, and other head and neck cancers, as the disease-free survival rate decreases to approximately 50% when even a single lymph node harbors metastasis. The extent of lymph node involvement can be considered as an indirect index of the systemic tumor burden and is a crucial factor of tumor staging, being closely correlated to overall survival and constituting a main determinant of treatment planning [45]. Preoperative evaluation of patients with newly diagnosed head and neck cancers includes physical examination, ­ultrasound, CT, MRI, and/or PET/CT; all these modalities have suboptimal sensitivity for the detection of microscopic lymph node involvement, as occult metastasis is found in 15–30% of patients with clinically node-negative head and neck squamous cell carcinoma [46]. Although debate is still ongoing about performing prophylactic lymph node neck adenectomy versus a wait-and-­

374

see approach [47], selective neck dissection is generally recommended in all patients with clinically negative lymph node status. However, lymph node metastasis is found in only about 30% of the patients with early oral squamous cell carcinoma so treated, thus resulting in overtreatment in over 70% of patients. Prophylactic neck lymphadenectomy has been developed based on the common pathways for spread of all head and neck cancers to regional nodes, and it consists of surgical removal of those nodes most commonly involved with metastasis specifically from head and neck cancers. The extent of lymphadenectomy varies according to location of the primary tumor. For example, in case of oral cancers, it includes all lymph nodes of levels I, II, and III and sometimes also the superior part of level V. In case of pharyngeal and laryngeal cancers, selective neck dissection includes levels II, III, IV, and VI when appropriate [48]. Due to the anatomical complexity and unpredictable individual variability of lymphatic drainage in the individual patients (that can be found in up to 20–30% of patients), the therapeutic value of selective neck dissection is limited by the occurrence of 10–20% of “skip metastases” that have bypassed the expected first nodal basin. In this scenario, radioguided SLNB constitutes an alternative to selective neck dissection, as it guides surgery to a personalized identification of lymphatic neck drainage and it allows detection of occult cervical metastasis in patients with early head and neck cancers. In particular, a negative SLNB prevents the unnecessary removal of functional lymph nodes and limits the extent of neck dissection surgery. Compared to selective neck dissection, patients also usually experience fewer complications such as sensory disturbances (skin numbness); overall duration of surgery is also considerably shortened [49]. Application of SLNB technique has been investigated mostly for those head and neck cancers whose anatomical location allows direct and easy access to the tumor for injection of the lymphatic mapping agents. For this reason the procedure has already been sufficiently standardized in patients with oral and oropharyngeal squamous cell carcinoma with accessible subsites, while it is still experimental in head and neck cancers arising at other sites. In patients with early oral and oropharyngeal cancer, SLNB has shown results comparable to those of selective neck dissection in terms of regional control, as demonstrated by multiple validation trials [50, 51].

Key Learning Points

• Metastasis to neck lymph nodes is a major determinant for the prognosis and for treatment planning in patients with oral, oropharyngeal, and other head and neck cancers.

F. Giammarile et al.

• Preoperative evaluation with even the most advanced imaging techniques has suboptimal sensitivity for the detection of microscopic lymph node involvement. • Although selective neck dissection is recommended in patients with clinically negative lymph node status, lymph node metastasis is found in only about 30% of the patients with early oral squamous cell carcinoma, thus resulting in overtreatment in over 70% of such patients. • Radioguided sentinel lymph node biopsy constitutes a valid alternative to selective neck dissection, as it personalizes identification of lymphatic neck drainage and it allows detection of occult cervical metastasis in patients with early head and neck cancers. • The procedure has already been sufficiently standardized in patients with oral and oropharyngeal squamous cell carcinoma with accessible subsites, whereas it is still under experimental in head and neck cancers arising at other sites.

16.4.2 Lymphatic System of the Head and Neck The head and neck region contains over 300 lymph nodes, constituting approximately 30% of lymph nodes in the whole human body. Their anatomy is quite complex due to the close proximity with different tissues and vital organs—often of small size. Concerning in particular lymphatic anatomy of the cervical region, a useful schematic classification into specific anatomic subsites groups them into seven levels on each side of the neck, Robbins’ classification [52] (Fig. 16.17). In the anterosuperior region of the neck, submental and submandibular lymph nodes are included in level I, respectively, as sublevel IA and sublevel IB, divided by the anterior belly of the digastric muscle. In the lateral region of the neck, upper jugular nodes extend from the base of the skull up to the inferior border of the hyoid bone and are classified as level II; the vertical plane defined by the posterior surface of the submandibular gland divides lymph nodes located anteriorly (sublevel IIA) from nodes located posteriorly (sublevel IIB). The middle jugular lymph nodes are included in level III, the space from the hyoid bone up to the inferior border of the cricoid cartilage. Level IV includes lower jugular nodes, from the inferior border of the cricoid cartilage to the clavicle. The posterior border of the sternocleidomastoid muscle constitutes the posterior border of levels II, III, and IV. Those lymph nodes located posteriorly to the posterior border of the sternocleidomastoid muscle are classified as level V; in turn, sublevel

16  Image-Guided and Radioguided Surgery

375

Fig. 16.17  Lymph node levels of the neck according to Robbins’ classification

VA includes the spinal accessory nodes, whereas sublevel VB includes the nodes following the transverse cervical vessels and the supraclavicular nodes. Pretracheal, paratracheal, precricoid, and perithyroidal nodes (including the lymph nodes along the recurrent laryngeal nerves) constitute level VI. Finally, level VII includes the superior mediastinal lymph nodes [53]. The patterns of lymph drainage from the oral cavity, ­oropharynx, larynx, and hypopharynx exhibit numerous interand intraindividual variations, even from the same primary tumor site. For example, drainage from the anterior floor of the mouth and lingual apex is expected to submandibular nodes; however, bilateral drainage to higher-level lymph nodes and to the middle jugular chains is not infrequently observed. Moreover, lymphatic drainage from a primary tumor located in the midline may be directed to either the left or the right side. On the other hand, lateralized malignancies of the tongue or floor of the mouth often show pure contralateral drainage [54–56].

Key Learning Points

• Anatomy of the lymphatic system in the head and neck region (which constitutes about 30% of all lymph nodes in the whole human body) is quite complex. • Robbins’ classification levels for each side of the neck constitute a well-validated basis for staging patients with cancers of the head and neck.

• The patterns of lymph drainage from the oral cavity, oropharynx, larynx, and hypopharynx exhibit numerous inter- and intraindividual variations, even from the same primary tumor site. • Lymph originating in midline structures (tongue and floor of the mouth) frequently drains to contralateral lymph nodes.

16.4.3 Modalities of Tracer Injection A variety of radiopharmaceuticals have been used for imaging the pattern of lymph flow in the head and neck region. In addition to the established radiocolloid lymphatic mapping agents (primarily 99mTc-albumin nanocolloid and 99mTc-­sulfur colloid), a new tracer targeting the mannose receptors located on macrophages and other cells in lymph nodes, 99mTc-tilmanocept, has more recently been approved; the particularly small particles of this agent result in rapid clearance from the site of interstitial injection, while avid binding to the CD206 receptors expressed on macrophages’ membranes results in reduced or absent drainage to second-­echelon lymph nodes and efficient SLN retention, thus facilitating SLN detection both preoperatively and intraoperatively [57, 58]. As an alternative to the radioactive label, near-infrared fluorescent tracers (such as indocyanine green—ICG) have been used that are potentially helpful for SLN biopsy [59]; moreover, the combination of fluorescent tracers with conventional 99mTc-

376

radiocolloids is an attractive option for improving the SLN detection rate in patients with head and neck malignancies, by combining radioguidance with optical guidance [60]. The technique of tracer injection may have a relevant impact on the acquisition time and image quality during preoperative lymphatic mapping. In case of oropharyngeal cancers, the tracer should be injected intramucosally in the healthy mucosa surrounding the malignant lesion or scar margin, considering that in the subepithelial stroma there is a high concentration of lymphatic capillaries, which provides a larger area for faster lymph drainage. Intratumoral or deep injections should be avoided, because bleeding at the injection site results in lower image quality and difficult SLN identification. Tracer injection is usually performed with small syringes with minimal dead space; alternatively, 0.1 mL of air may be drawn into the syringe behind the radiocolloid suspension to ensure complete administration. A 25 G or 27 G needle should be used, total injected activities varying from 15 to 120 MBq, depending on size and location of the primary tumor. Furthermore, the injected activity should be adjusted according to the timing of lymphoscintigraphy with respect to surgery; in particular, greater activities are required if surgery is scheduled the day after lymphoscintigraphy (2-day protocol), in order to ensure that the remaining activity exceeds 10 MBq at the time of surgery. Small volumes (0.1–0.2 mL per aliquot) should be injected at a short distance from the primary lesion, in order to avoid masking of SLN(s) possibly located in the vicinity of the injection site; small volume is recommended also to minimize spilling of the radiotracer (due to resistance to injection of, e.g., the tongue tissue) causing contamination of the field. The number of aliquots to be injected varies from two to four, depending on size and location of the lesion. The use of a local anesthetic for topical application (10% xylocaine spray) a few minutes before tracer injection is recommended for oral cavity tumors. The most frequent pitfall possibly occurring during lymphoscintigraphy is skin or mucosal radioactive contamination due to spillage of the lymphatic mapping agent either during injection or immediately thereafter. Oral contamination can be reduced by inviting the patient to use a mouthwash or to rinse the mouth before swallowing.

Key Learning Points

• The lymphatic mapping radiopharmaceutical (either 99mTc-colloid or 99mTc-tilmanocept) is generally injected intramucosally in multiple aliquots around the tumor site. • The use of hybrid radioactive and fluorescent lymphatic mapping agents (potentially combining the advantages of radioguidance with those of visual guidance) is currently under clinical investigation. • For cancers of the mouth, cautions must be adopted to minimize spilling of the radiotracer causing contamination of the field.

F. Giammarile et al.

16.4.4 Lymphoscintigraphy Preoperative lymphoscintigraphy provides images of lymphatic drainage in the tumor region. SLNs are visualized as “hot spots” along a certain pathway of lymphatic drainage. Accurate gamma camera imaging plays an important role in lymphatic mapping, since the resulting images are used to direct the surgeon to the site(s) of SLN(s). With the patient positioned as comfortably as possible on the imaging table (usually in the supine position—to reproduce positioning during surgery), acquisitions are generally performed using a large FOV gamma camera to visualize all the possible routes of lymphatic drainage. Dynamic acquisition for 20–30  min starting immediately after radiotracer injection shows the drainage pattern and helps to distinguish SLNs from second-echelon nodes. SLNs are generally identified 15–60 min after radiotracer injection as one or more hot spots to which lymphatic drainage passes and may be multiple, ipsilateral, and/or contralateral to the primary tumor, in one or more areas of the neck. Static images in the anterior and lateral views are subsequently acquired; the skin projection of SLNs can be marked with the aid of an external radioactive marker, such as a 57Co-source pen. If SLNs are not clearly visualized, static imaging can be repeated later 2–4  h postinjection or even just before surgery. Repeat radiotracer injection and imaging may be considered in case of totally absent visualization of lymph nodes; however, proceeding to neck dissection is preferred in these cases in order to avoid erroneous staging information caused by a false-negative SLN biopsy. It must be considered that accurate preoperative localization and marking of the skin projection of SLN location correlate well with the precision of the surgical procedure. Although SLNs can often be adequately localized with planar imaging alone, the use of SPECT/CT imaging facilitates in the majority of cases the procedure of SLNB, particularly when considering the complex anatomical region of the head and neck and the proximity of some primary lesions to SLNs and to other important anatomic structures. SPECT/CT is useful to identify additional SLNs (otherwise missed on planar imaging, especially when radioactive lymph nodes are adjacent to the injection site) (Fig.  16.18), as well as to exclude ambiguous imaging in case of radiotracer leakage or contamination. Furthermore, SPECT/CT imaging allows planning of the best surgical approach, thanks to the anatomical details provided by CT in hybrid images and thanks to volume rendering with 3D images; for instance, it can clarify if SLNs are located superficially underneath the skin or hidden below in deep tissues or in proximity or not to vital vascular and neural structures [61, 62]. Nevertheless, it must be emphasized that SPECT/CT imaging does not replace planar lymphoscintigraphy—but rather it complements planar imaging.

16  Image-Guided and Radioguided Surgery

377

a

b

c

d

e

Fig. 16.18  Lymphoscintigraphy obtained after perilesional, submucosal injection of 99mTc-nanocolloidal albumin in a patient with carcinoma of the tongue. (a) Planar imaging shows bilateral drainage of the lymphatic mapping agent; however, due to proximity with the site of radiocolloid injection (large area of intense activity), it is difficult to ascertain the exact number and anatomical location of the radioactive lymph nodes. (b) 3D surface volume rendering depicts an additional SLN, left from the injection site (yellow arrow). Transaxial fused SPECT/CT

f

slices (c and e) and corresponding CT slices (d and f) better localize SLNs in the submandibular area (yellow dashed circle in d) and in the lower neck (yellow dashed circles in f) (reproduced with permission from: Orsini F, Puta E, Valdés Olmos RA, Vidal-Sicart S, Giammarile F, Mariani G.  Radioguided surgery for head and neck. In: Strauss HW, Mariani G, Volterrani D, Larson SM, eds. Nuclear oncology—From pathophysiology to clinical applications. New  York, NY: Springer; 2017:1433–49)

Key Learning Points

16.4.5 Intraoperative Gamma Probe Counting/Detection

• In patients with head and neck cancers, lymphoscintigraphy should include an early dynamic acquisition, which helps to distinguish true sentinel lymph nodes from second-echelon nodes visualized along the same pathway of lymphatic drainage. • Static images in the anterior and lateral views are subsequently acquired, images being recorded even up to 2–4 h after injection if necessary when sentinel lymph nodes are not clearly visualized. • SPECT/CT imaging facilitates in the majority of cases preoperative lymphatic mapping, because of the complex anatomical region of the head and neck and the proximity of some primary lesions to SLNs and to other important anatomic structures.

A handheld γ-detecting probe is routinely used for intraoperative detection of the SLN(s) in the surgical field. The injection of blue dye at the time of surgery in addition to radioguidance with the intraoperative gamma probe is optional and may be a useful adjunct to aid SLN localization and harvesting. After SLN removal, the resection site and other cervical regions are explored with the gamma probe for any significant residual radioactivity; in this case further radioactive lymph nodes may be retrieved and harvested. An ex  vivo counting ratio of 10:1/20:1 with respect to background identifies a hot spot as a radioactive SLN to be analyzed for the presence of metastatic tumor cells. Sometimes the SLNs can be difficult to identify due to close proximity with the peritumoral injection site (the

378

so-­called shine-through effect), which can limit detection of SLNs, especially in patients with tumors of the mouth floor. Also after removing radioactive SLNs, activity remaining at the injection site can impair measurement of residual activity in the excision surgical field. For these reasons, the primary cancer should be resected before searching for the SLN(s). Deeply located SLNs can be difficult to detect because of tissue attenuation; in these cases SPECT/CT imaging considerably helps in evaluating depth of radioactive lymph nodes. A number of portable and handheld mini gamma cameras have been developed to provide direct intraoperative visualization of radioactive foci, with the purpose of improving detection of SLNs. Using these devices, the entire lymph node excision procedure in the head and neck area can be directly monitored in the surgical room. Moreover, by acquiring images of absent or residual radioactivity after excision of the SLNs, the dedicated mini gamma camera can help to assess completeness of the procedure [63, 64]. Recently the use of intraoperative gamma cameras has been combined with fluorescence cameras for synchronous SLN signal detection using the abovementioned hybrid ­lymphatic mapping radiotracers combined with indocyanine green [65]. A further technology called freehand SPECT (fhSPECT) has been introduced for navigational surgery, combining the acoustic information of a conventional gamma probe and intraoperative 3D images with real-time visualization of radiotracer distribution within the surgical field.

Key Learning Points

• The injection of blue dye at the time of surgery in addition to radioguidance with the intraoperative gamma probe may be a useful adjunct to aid sentinel lymph node localization and harvesting. • Any lymph node with an ex vivo counting ratio of 10:1/20:1 versus background should be harvested for analysis. • Intraoperative high-resolution imaging with portable, small field-of-view gamma cameras does not guide per se search for the sentinel lymph node(s), but rather it is employed for assessing completeness of radioguided sentinel lymph node excision. • The so-called “freehand SPECT” technology allows navigational surgery, combining the acoustic information of a conventional gamma probe and intraoperative 3D images with real-time visualization of radiotracer distribution within the surgical field.

F. Giammarile et al.

16.4.6 Retrieval and Analysis of  Sentinel Lymph Nodes Histopathologic analysis of SLNs is extremely effective and can detect the presence of micrometastasis (50% myometrial invasion or papillary serous, carcinosarcoma, and clear-cell cancer histology). Currently, there is no international consensus on the optimal extent of surgery for endometrial cancer staging. Thus, the SLN procedure is being explored as possible alternative option to de novo pelvic and para-aortic lymphadenectomy [78, 79]. In general, radioguided SLNB in patients with gynecological cancers involves a more complex approach than in patients with breast cancer or melanoma, due to a variety of anatomical and pathophysiological factors.

F. Giammarile et al.

Key Learning Points

• Gynecological malignancies where sentinel lymph node biopsy can be employed include vulvar cancer, cervical cancer, and endometrial cancer. • Lymph originating in the vulva normally drains to the inguino-femoral lymph nodes, proceeding from superficial to deep inguinal nodes and to pelvic lymph nodes. • The expected lymphatic drainage from the cervix is bilaterally to the parametria, to the external iliac nodes, and to the obturator lymph node, progressing thereon to the common iliac and para-aortic lymph nodes. • Lymph from the two lower thirds of the corpus uteri drains in a similar manner as lymph from the cervix, whereas lymph from the upper third drains directly to the para-aortic lymph nodes.

16.5.3 Modalities of Tracer Injection

Administration of the lymphatic mapping agent (e.g., a 99m Tc-colloid) in patients with vulvar malignancies is performed through three to four intradermal/intramucosal injections (0.1–0.2 mL each aliquot) around the primary lesion or Lymphatic drainage from the vulva normally flows to the the excision scar (Fig. 16.19). inguino-femoral lymph nodes, either on one side only or In patients with cervical cancer, lymphatic mapping is bilaterally. In particular, lymph originating from the labia performed by peritumoral or periorificial injection of flows to the superficial inguinal nodes, proceeding then to four radiopharmaceutical aliquots. In the case of previthe deep inguinal nodes and to the pelvic lymph nodes, ous conization, pericicatricial injection at the four quadwhereas some midline structures, such as the clitoris, can rants is recommended. Usually, a total volume of 2 mL is drain directly to deep lymph nodes. Tumors located in the applied. midline usually drain to both inguinal sides. One of the most controversial issues for accurate lymThe cervix is a deep midline structure whose lymph drains phatic mapping in patients with endometrial cancer conlaterally to the parametria, to the external iliac lymph nodes, cerns the best injection site/modality. Different approaches and to the obturator lymph nodes. The expected drainage is have been proposed to this purpose, such as cervical injecbilateral migration to the external iliac lymph nodes. tion, endometrial peritumoral injection assisted by hysterObturator fossa lymph nodes are the most common group, oscopy, and myometrial/subserosal injection. The cervical followed by progression of lymph to the common iliac and approach (similar to the technique adopted for cervical canpara-aortic lymph nodes as the second-echelon group. Direct cer, although with somewhat deeper punctions) is the easipara-aortic drainage has been observed in 1% of cases. est way to inject the lymphatic mapping agent. It is Parametrial lymph nodes can be very difficult to evaluate performed periorificially into the four quadrants. during lymphoscintigraphy and radioguided SLNB, due to Endometrial injection during hysteroscopy allows direct their proximity to the site of injection of the lymphatic map- injection around the tumor. If it is performed the day before ping agent. surgery, this procedure requires particular coordination Lymphatic drainage from the corpus uteri is in some among the gynecology, anesthesia, and nuclear medicine respect different from cervical drainage. In particular, lymph departments. If it is performed shortly prior to the start of from the two lower thirds of the corpus drains in a similar surgery, it is problematic to acquire lymphoscintigraphy manner as lymph from the cervix. Whereas, lymph from the and SPECT/CT. The third option, radiotracer injection into upper third of the corpus uteri drains directly to the para-­ the corpus uteri in a myometrial or subserosal location, is aortic lymph nodes [79]. usually adopted during surgery. However, it is also possible

16.5.2 Lymphatic System of Intrapelvic Female Organs

16  Image-Guided and Radioguided Surgery

381

Fig. 16.19  Modality of injection of the lymphatic mapping agent in vulvar cancer. Submucosal injection in a patient with a midline lesion in the left labia minora. Since bilateral lymphatic drainage is expected

in the majority of these patients, it is important to inject the radiocolloid around the tumor, although this can be difficult in the medial part (right panel)

to inject the tracer guided by transvaginal ultrasonography the day prior to surgery [78].

then obtained (Fig. 16.20). The lymph node that first receives a direct lymphatic channel from the tumor or shows an increase in uptake in the delayed images is usually considered as a SLN. In patients with cervical cancer (Fig.  16.21) or with endometrial cancer (Fig. 16.22), planar imaging is based on 3–5 min anterior and lateral views acquired at 30 min (early) and 60–120 min (delayed) after radiotracer injection. Early images can depict the lymphatic duct(s) and the first-­ draining lymph node(s). Delayed images may discriminate the SLN from second-echelon nodes. Planar images cannot, however, provide precise anatomical landmarks [78]. This limitation is overcome by SPECT/CT imaging obtained just after delayed imaging. In vulvar cancer, lymphatic drainage is mainly directed to Daseler’s medial inguinal region (83%), while drainage to the lateral inferior groin is minimal (0.5%) [79]. SPECT/CT is crucial to adopt the optimal approach to inguinal lymphadenectomy in patients with positive SLNB. SPECT/CT imaging is mandatory in case of cervical and endometrial malignancies, since it allows correction for tissue attenuation and may thus lead to detection of additional SLN(s). Moreover, by providing accurate anatomical localization, it plays an important role in planning surgery and shortening the operative time [80].

Key Learning Points

• In patients with vulvar cancer, the lymphatic mapping agent is administered through 3–4 intradermal/ intramucosal injections around the primary lesion or the excision scar. • In patients with cervical cancer, the lymphatic mapping agent is injected peritumorally or periorificially, at the four quadrants. • The optimal modality of injection in patients with endometrial cancer is still debated, and different approaches are being explored (e.g., cervical injection, endometrial injection during hysteroscopy).

16.5.4 Lymphoscintigraphy In the case of vulvar cancer, a dynamic acquisition is recorded immediately after tracer injection, usually for about 10 min. Early (15  min) and delayed (2  h) static planar images are

382

F. Giammarile et al.

Fig. 16.20  Lymphoscintigraphy obtained after perilesional, submucosal injection of 99mTc-nanocolloidal albumin in the same patient with vulvar cancer as depicted in Fig. 16.19. The delayed planar image (left panel) shows a predominantly left inguinal lymphatic drainage.

However, a faint uptake on the right inguinal side is observed as well (red arrow). The 3D surface volume rendering image obtained after SPECT/CT acquisition more accurately visualizes the radioactive lymph nodes (right panel)

Fig. 16.21  Lymphoscintigraphy obtained after perilesional injection of 99m Tc-nanocolloidal albumin in the patient with cervical cancer. Upper left panel: planar imaging shows bilateral lymphatic drainage with two radioactive lymph nodes in the right pelvis and one in the left pelvis. Note a hot spot near the injection site, depicted on the right-hand side. Upper right panel: 3D surface volume rendering obtained after SPECT/CT acquisition shows the anatomical distribution of the radioactive lymph nodes. Coronal fused SPECT/CT slice (lower left panel) and transaxial fused SPECT/CT

slice (lower right panel) depict more precisely the location of the radioactive lymph nodes, one of which (considered a true SLN) is located in the right parametrium (reproduced with permission from: Paredes P, VidalSicart S. Preoperative and intraoperative lymphatic mapping for radioguided sentinel node biopsy in cancers of the female reproductive system. In: Mariani G, Manca G, Orsini F, Vidal-Sicart S, Valdés Olmos RA, eds. Atlas of lymphoscintigraphy and Sentinel Node Mapping—A pictorial case-based approach. Milan: Springer 2013:249–68)

16  Image-Guided and Radioguided Surgery

Fig. 16.22  Lymphoscintigraphy obtained after cervical injection of 99m Tc-nanocolloidal albumin in the patient with endometrial cancer. The transaxial fused SPECT/CT slice (upper left panel) shows radiocolloid uptake in SLNs located in both external iliac areas; the lymph nodes are depicted in the CT component of the SPECT/CT acquisition (green circles in the lower left pane). Three-dimensional surface volume rendering obtained after SPECT/CT acquisition provides a good overview of the

Key Learning Points

• In patients with vulvar cancer, an early dynamic acquisition helps to identify the lymphatic basins at risk and also to recognize higher-echelon lymph nodes visualized after visualization of the “true” sentinel lymph node along a certain pathway of lymphatic drainage; static planar images acquired up to about 2 h postinjection are usually sufficient for adequate lymphoscintigraphic mapping in these patients. • Lymphoscintigraphy can be more complex in patients with cervical cancer or endometrial cancer, in whom SPECT/CT imaging acquired in addition to early and delayed planar imaging provides the anatomical correlates that are especially useful to guide the surgeon to retrieve the sentinel lymph node(s).

16.5.5 Intraoperative Gamma Probe/Gamma Camera Detection SLN detection using blue dyes alone in patients with vulvar or cervical cancer ranges from 75% to 95%. Although in most

383

anatomical location of the radioactive lymph nodes near major blood vessels (reproduced with permission from: Paredes P, Vidal-Sicart S. Preoperative and intraoperative lymphatic mapping for radioguided sentinel node biopsy in cancers of the female reproductive system. In: Mariani G, Manca G, Orsini F, Vidal-Sicart S, Valdés Olmos RA, eds. Atlas of lymphoscintigraphy and Sentinel Node Mapping—A pictorial case-based approach. Milan: Springer 2013:249–68)

cases the same SLNs are identified by the blue dye and by the radioactive lymphatic mapping agent, radioguided SLNB has in general better performance than optical guidance alone with blue dyes. Nevertheless, current standard practice in patients with vulvar or cervical cancer includes the simultaneous injection of both types of lymphatic mapping agent, radiopharmaceutical and blue dye, for combined radioguidance and optical guidance. The use of fluorophores (mainly indocyanine green, ICG) has recently been advocated in this scenario, with better results than with the blue dye [78, 79]. In patients with vulvar tumors, skin marks on the cutaneous projection of the SLN(s) can indicate the best location for surgical incision and subsequent use of the handheld gamma probe for exploring the surgical field. Conversely, in cervical and endometrial tumors, laparoscopy is the most frequent surgical approach; therefore, cutaneous marks are not useful as a guide. Moreover, specifically designed gamma probes must be used during laparoscopy. In particular, these probes must be inserted through a 10–12-mm-diameter trocar, must be long enough to reach the area of interest (up to a distance of 25 cm from the abdominal surface), and must have suitable maneuverability so to cover an area with a radius of 15–20 cm. Laparoscopic gamma probes have been developed taking these issues into consideration, in particular the limitation on maneuverability that may impede SLN identification. In these probes the angle of the detector rela-

384

tive to the tip varies from 0° to 45–90°, according to specific uses in different phases of the same surgical session. In clinical practice, possible interferences during laparoscopic gamma probe scanning must be kept in mind. In particular, the uterus is usually enlarged in patients with endometrial cancer, thus increasing the interference on SLN detection. Furthermore, laparoscopic gamma probe scanning along the intra-abdominal lymphatic pathways can be affected by activity accumulated in the ureter or by liver activity due to radiocolloid uptake in the reticuloendothelial system [78–82]. Due to anatomical complexity of the area, activity remaining at the injection site can mask activity coming from the SLN, as in the case of, e.g., an inguino-femoral SLN adjacent to the radiotracer injection site in a patient with vulvar cancer or parametrial SLNs in cervical/endometrial cancer. In such instance, intraoperative visual guidance (e.g., with blue dye—or with fluorescent agents) is particularly useful. Furthermore, intraoperative use of gamma probes can be supplemented with the use of portable gamma cameras or other techniques such as intraoperative freehand SPECT. The main advantages of using a portable gamma camera for SLN detection in gynecological cancers are (1) greater sensitivity for localizing parametrial lymph nodes, (2) better discrimination of interference from liver activity when resecting para-aortic lymph nodes, and (3) better ability to ascertain the completeness of SLN excision. Intraoperative freehand SPECT can provide some advantages also in this setting, as it yields virtually real-time information on precise SLN’s depth and on completeness of SLN resection. Moreover, SPECT/CT data obtained during lymphoscintigraphy with a large field-of-view gamma camera can be uploaded and included in the image display to provide anatomical landmarks and the possibility of intraoperative navigation [79, 80].

Key Learning Points

• In patients with vulvar or cervical cancer, sentinel lymph node biopsy is often performed combining preoperative lymphatic mapping and intraoperative gamma probe guidance with visual guidance using a blue dye (or more recently a fluorescent agent). • The most frequent approach to sentinel lymph node biopsy in patients with cervical or endometrial cancer is laparoscopic surgery, using specifically designed gamma probes that can be inserted in the abdominal cavity through a laparoscopic trocar and reach deep intra-abdominal sites. • Intraoperative gamma probe guidance can be somewhat hampered by interferences in counting caused

F. Giammarile et al.

by proximity with radioactivity either at the injection site or at physiologic sites of accumulation other than lymph nodes. • Visual guidance, especially with the use of a fluorescent agent, turns out to be particularly useful to overcome these interferences. • Other imaging modalities possibly employed during surgery (such as dedicated small field-of-view gamma cameras or freehand SPECT) can be useful to assess completeness of sentinel lymph node excision.

16.5.6 Retrieval and Analysis of Sentinel Lymph Nodes The optimal method for histopathological examination of SLNs in patients with gynecological cancers has not been established. Considering that routine staining with hematoxylin and eosin (H&E) may not allow the identification of micrometastasis (50% on coronary angiography plus fractional flow reserve 140 mg/dL, 2 IU of regular insulin should be injected before [18F]FDG Intravenous glucose administration can be considered; however, it is not recommended in case of SGL >120 mg/dL and in diabetic patients. 13–25 g of 50% dextrose solution should be administered depending on SGL. After 30–60 min, [18F]FDG should be injected This protocol can be used both in diabetic and nondiabetic patients, and it is based on the simultaneous infusion of insulin and glucose acting on the tissue as a metabolic challenge in order to maximize myocardial [18F]FDG uptake. After SGL estimation, insulin should be injected in declining infusion rates starting with 0.16 IU/min/kg for 4 min, followed by 0.08 IU/min/kg for 3 min and 0.04 IU/min/kg at 8 min. Co-infusion of 20% dextrose should be carried out in order to achieve a constant, euglycemic blood glucose concentration. [18F]FDG can be injected when SGL reaches the range 80–110 mg/dL. For the best image quality, insulin and dextrose infusion should be maintained for 30 min This compound induces the inhibition of the peripheral lipolysis, thus decreasing free fatty acid plasma levels. The lack of this energetic substrate shifts myocardial metabolism toward preferential glucose utilization (glucose switch). The usual protocol suggests oral 250 mg acipimox administration 90–60 min prior to [18F]FDG injection, after overnight fasting [26]

SGL serum glucose level

l­ipolytic activity of heparin that can induce a fivefold increase in the blood free fatty acid levels. Although still suboptimal, this feature actually approaches the clinical applicability. Nevertheless, repeatability of this procedure has not been thoroughly tested, particularly in PVE patients in whom a sugar-free diet has been maintained either for 12 h without any further intervention or for 48 h and combined with heparin administration. Accordingly, further studies to optimize dietary preparation for this procedure to optimize the diagnostic potential of [18F]FDG PET/CT imaging in PVE diagnosis. Dietary and/or pharmacologic preparations prior to myocardial [18F]FDG PET imaging have been proposed [76], as summarized in Tables 20.7 and 20.8.

477 Table 20.8 Methods to suppress glucose utilization by normal myocardium Method Prolonged fast High-fat/ low-­ carbohydrate diet Intravenous unfractionated heparin

Combined methods

Technique Fasting state for 12–18 h Two meals 24 h before the study, followed by an overnight fast. High fat and protein with low/no carbohydrate content meal are allowed 15–50 IU of unfractionated heparin 15 min prior to [18F]FDG administration or 500 IU 45 and 15 min prior to [18F]FDG administration (total 1000 IU). Ensure patient has no contraindications to administration of intravenous heparin including bleeding tendencies, allergies, or history of heparin-­ induced thrombocytopenia with thrombosis High-fat/low-carbohydrate diet for two meals, 1 day prior, followed by overnight fast, and intravenous regular heparin before [18F]FDG administration

Adapted from [26]

The [18F]FDG activity recommended in the joint EANM/ SNMMI guidelines on PET imaging in inflammation/infection is 2.5–5.0 MBq/kg (175–350 MBq for a 70-kg standard adult) [77]. Although antimicrobial treatment is expected to decrease the intensity of [18F]FDG accumulation, there is no evidence at this stage to routinely recommend treatment discontinuation before performing PET/CT. Whereas, corticosteroid drugs should be discontinued or at least reduced to the lowest possible dose in the 24 h preceding the scan. Image acquisition generally starts after an uptake time of 45–60 min, and the emission time per bed position depends on sensitivity of the scanner. As in oncology, the field of acquisition generally includes from the skull base to mid thighs (total body). Whole-body images including the lower limbs might be suggested to detect complications of IE such as mycotic aneurysms that may require specific treatment by embolization to prevent rupture. An additional separate bed on the cardiac region is useful to record gated images. Diagnostic CT angiography (CTA) imaging might also be performed, to maximize the diagnostic information provided by the exam. Although delayed imaging has been proposed to increase specificity in diagnosing infection of cardiovascular implants, recent data suggested that in IE delayed images are more prone to false-positive results. Also in case of [18F]FDG PET/CT, image reconstruction with and without CT-based attenuation correction is recommended to identify potential reconstruction artifacts. Techniques for metal artifact reduction are useful to minimize overcorrection, even if they do not always recover completely quality of the PET images. PET/CT images must be visually evaluated for increased [18F]FDG uptake, taking into consideration the pattern (focal, linear, diffuse), the intensity,

478

and the relationship to areas of physiologic distribution. PET information is compared with morphologic information obtained by CT and, possibly, CTA. A physiological variant that might constitute a confounding factor while reading the images is the presence of increased metabolic activity along the posterior part of the heart, where lipomatous hypertrophy of the interatrial septum may appear as a fat-containing mass with increased [18F]FDG uptake. Patients who have recently undergone cardiac surgery might present postoperative inflammation that results in non-­specific [18F]FDG uptake in the immediate postoperative period. To minimize the risk of false-positive findings, the European Society of Cardiology Guidelines recommend not utilizing [18F]FDG PET in the 3-month period following valve implantation. If the scan is performed early after surgery, the possibility of false-positive results should be taken into account. A number of additional conditions can mimic the pattern of focally increased [18F]FDG uptake that is typically observed in IE, such as active thrombi, soft atherosclerotic plaques, vasculitis, primary cardiac tumors, cardiac metastasis from a non-cardiac tumor, postsurgical inflammation, and foreign body reactions.

Key Learning Points

• Scintigraphy with radiolabelled autologous leukocytes (99mTc-HMPAO-WBC) is the cornerstone of radionuclide imaging of cardiac infections. • SPECT/CT imaging is crucial for increased specificity and for accurately localizing the foci of infection. • PET/CT with [18F]FDG for imaging cardiac infections is an especially attractive alternative option, thanks to better spatial resolution and less cumbersome labelling procedure versus 99mTc-HMPAO-­ WBC scintigraphy. • When performing [18F]FDG PET/CT in patients with suspected cardiac infection, special attention must be paid to preparation of the patient so to minimize physiological myocardial uptake of [18F] FDG.

20.3.3 Radionuclide Imaging of Prosthetic Valve Endocarditis The sensitivity of 99mTc-HMPAO-WBC SPECT/CT for prosthetic valve endocarditis has been reported at an overall 64–90%, with 36–100% specificity, and 85–100% positive and 47–81% negative predictive values [72, 74, 78]. In case of abscess formation, 99mTc-HMPAO-WBC SPECT/CT has 83–100% sensitivity, 78–87% specificity, and 43–71% p­ ositive

M. Bauckneht et al.

and 93–100% negative predictive values, even in the early post-intervention phase [79, 80]. In addition, the intensity of 99m Tc-HMPAO-WBC accumulation in the perivalvular area represents an interesting marker of local infectious activity, in that patients with a mild activity on the first exam disappearing on the second imaging evaluation seem to have a favorable outcome [79]. This observation opens the very interesting perspective for the use of molecular multimodality imaging for the assessment of antimicrobial treatment response. In a recent systematic review on the assessment of PVE, [18F]FDG PET/CT sensitivity and specificity have been reported to be 73–100% and 71–100%, respectively, with 67–100% positive and 50–100% negative predictive values, respectively [80]. Addition of [18F]FDG PET/CT to the modified Duke criteria increased sensitivity for definite IE from 52–70% to 91–97%, by reducing the number of possible PVE cases [81]. When [18F]FDG PET/CT is associated with CT angiography ([18F]FDG PET/CTA), both sensitivity and specificity for the diagnosis of IE increased to 91%, with 93% positive and 88% negative predictive values [82]. In association with the Duke criteria, [18F]FDG PET/CTA allowed reclassification of 90% of the cases initially classified as “possible” IE and provided a more conclusive diagnosis (definite/reject) in 95% of the patients. This combined multimodality procedure should be considered in all the patients where echocardiography presents significant limitations. In fact, its ability to provide relevant information on the local extent of the disease such as the presence of pseudoaneurysms, fistulas, thrombosis, and coronary involvement is significant for the subsequent clinical and surgical decision-making. In addition, by adding CTA to PET/CT in IE patients, it is possible to assess the entire chest, therefore to identify septic pulmonary infarcts and abscesses, as well as to evaluate the aorta and the coronary arteries in the perspective of surgery. Figures 20.9 and 20.10 show examples of [18F]FDG PET/CT and [18F]FDG PET/CTA contribution to correct diagnosis in patients with suspected IE after implantation of prosthetic valves.

20.3.4 Radionuclide Imaging of Native Valve Endocarditis Despite the fact that imaging interpretation might be more straightforward than in patients with PVE, the diagnostic value of [18F]FDG PET/CT has not been well determined in case of native valve endocarditis (NVE). In fact, most studies included mainly PVE or a mixed patient population with both native and prosthetic valves. The reported low sensitivity of [18F]FDG PET/CT in NVE is likely to be mainly related to the location and size of the lesions. In NVE the presence of a vegetation is the main finding, at least in the initial phase of the disease. It should also be noticed that this is a clinical

20  Radionuclide Imaging of Cardiovascular Disease

Fig. 20.9 [18F]FDG PET/CT in a 50-year-old man with an aortic biological prosthesis positioned 2  years earlier. The patient developed fever, increased ESR, and positive blood culture for Enterococcus faecalis. Echocardiography was negative. Despite suboptimal suppression

a

c

Fig. 20.10  Examples of postoperative inflammation versus infective endocarditis in prosthetic aortic valves as evaluated by [18F]FDG PET/CT and CT angiography (CTA). (a) A 38-year-old man submitted to aortic valve replacement with a bileaflet mechanical prosthesis. Fused PET/CTA transverse view of the aortic valve shows a normally inserted aortic prosthesis with mild, homogeneous valvular inflammatory reaction (SUVmax 3.1) (arrows). (b) A 56-year-old man with mechanical aortic valve infective endocarditis (valve implantation 3 years previously). PET/CTA shows intense (SUVmax 13.33) periprosthetic [18F]FDG uptake and a poorly delimited perivalvular soft tissue lesion, consistent with a periprosthetic abscess (arrowheads). (c) A 78-year-old man submitted to bioprosthetic

479

of myocardial [18F]FDG uptake, PET/CT clearly shows a focal area of increased uptake in the perivalvular region, at the medial aspect (upper panel, from left to right transaxial CT, PET emission, and fused PET/ CT images)

b

d

aortic valve replacement 1 month earlier. Fused PET/CTA thin valve leaflets and mild, homogeneous perivalvular [18F]FDG uptake (SUVmax 1.58), similar to activity in the blood pool. (d) A 79-year-old woman with early (3 months postimplantation) bioprosthetic aortic valve infective endocarditis. Fused PET/CTA shows intense (SUVmax 13), heterogeneous periprosthetic [18F]FDG uptake (arrows), thickened valve leaflets, and a multiloculated perivalvular pseudoaneurysm (asterisk) (reproduced from: Pizzi MN, Roque A, Cuéllar-Calabria H, Fernández-Hidalgo N, FerreiraGonzález I, González-Alujas MT, et al. 18F-FDG-PET/CTA of prosthetic cardiac valves and valve-tube grafts: infective versus inflammatory patterns. JACC Cardiovasc Imaging. 2016;9:1224–1227)

480

M. Bauckneht et al.

setting where echocardiography has a very high accuracy; therefore, the use of multimodality imaging is reserved to a limited number of patients, i.e., patients with severe valve calcific degeneration with suboptimal acoustic window.

ficity of 88% (95% CI, 77–94%), and AUC of 0.861 [83]. Such a finding is mainly related to the small size of the vegetations along the leads, which are often well below spatial resolution of the system. [18F]FDG PET/CT and 99mTc-HMPAO-WBC scan findings in association with Duke criteria also allowed reclassifying 20.3.5 Radionuclide Imaging most of cases initially classified as “possible” IE, distinguishof CIED Infections ing infection limited to the pocket or leads from a more severe infection affecting the whole device and identifying patients Similarly as described above for PVE, 99mTc-HMPAO-WBC requiring device extraction. Semiquantitative parameters such SPECT/CT and [18F]FDG PET/CT can be used to confirm/ as semiquantitative ratio of maximum count rate of the pocket exclude infection and characterize the extension of the infec- device over mean count rate of lung parenchyma or normalizatious process, including extracardiac workup. The diagnosis tion of SUVmax around the CIEDs to the mean hepatic blood of local infections is quite straightforward. A recent meta-­ pool ratio activity might help in differentiated mild postoperaanalysis provides pooled specificity and sensitivity in this tive residual inflammation (lasting up to 2 months after device subgroup of 93% (95% CI, 84–98%) and 98% (95% CI, implantation) versus infection. Finally, those patients with 88–100%), respectively, and AUC was 0.98 for [18F]FDG suspected infection but without [18F]FDG uptake have been PET/CT. The largest study with 99mTc-HMPAO-WBC scin- shown to have a favorable outcome following antibiotic thertigraphy reported a sensitivity of 94%, with 100% specific- apy, thus suggesting the absence of bacterial colonization of ity. The diagnostic accuracy for lead infections is lower, with CIEDs. Figure  20.11 shows an example of the use of [18F] overall pooled sensitivity of 65% (95% CI, 53–76%), speci- FDG PET/CT in a patient with CIED infection.

Fig. 20.11 [18F]FDG PET/CT in a patient previously submitted to CIED implantation, with fever lasting since several weeks with mild increased CRP and ESR.  Echocardiography was negative. Antimicrobial treatment was initiated. Increased [18F]FDG uptake around the pocket (upper row: transaxial CT, emission PET, and fused

PET/CT images from left to right). Infection involved also the intravascular portion of the electrocatheters (middle row), as well as the intracardiac portion (bottom row). Based on these imaging findings, the patient was treated with removal of the device and prolonged antimicrobial treatment

20  Radionuclide Imaging of Cardiovascular Disease

20.3.6 Radionuclide Imaging of LVAD-Related Infections The usefulness of 99mTc-HMPAO-WBC SPECT/CT and of [18F]FDG PET/CT in the diagnosis of LVAD-related infection has been shown in small patient groups under routine clinical conditions. Molecular imaging allows precise anatomic location and accurate extent of a suspected infection with sensitivity of 100% and a specificity of 94% in case of [18F]FDG PET/CT [84]. The use of the metabolic volume has been recently reported to be associated with increased diagnostic accuracy as compared to the SUVmax index in a series of 48 patients. In particular, the negative predictive value and sensitivity increased up to >95% by using the metabolic volume compared to 87.5% when using SUVmax [85].

20.3.7 Extracardiac Workup of Patients with Infective Endocarditis The most frequent extracardiac manifestations of either native valve and prosthetic valve endocarditis (reported in 30–80% of patients) are embolic stroke or septic embolization to the bone, spleen, or kidneys, although only some of these are symptomatic [86]. The majority of embolisms occur within the first 14 days after treatment initiation; nevertheless, they may appear as the initial symptom leading to the diagnosis and frequently are recurrent [87]. The localization of the emboli and their cerebral/extracerebral proportion vary according to the studies, in particular according to the frequency and modalities of imaging and the proportion of right-sided and left-sided IE.

20.3.7.1 Detection of Embolic Events The search for asymptomatic embolic events through systematic extracardiac imaging has become a very important issue especially after inclusion of the detection of asymptomatic embolic events as a minor Duke criterion in the 2015 ESC criteria. The panel of imaging modalities used routinely to evaluate patients with extracardiac infective processes includes dental radiography, abdominal ultrasound, CT scan of the brain, whole-body CT, or MRI scan. CT scan (including cerebral) has long been considered the main imaging technique for the diagnosis of embolic events in IE patients; MRI is a valuable alternative in case of cerebral embolism, having the advantage of a higher sensitivity in detecting recent ischemic lesions and small ischemic or hemorrhagic lesions without the need of administering an iodinated contrast medium. Both [18F]FDG PET/CT and 99m Tc-HMPAO-­ WBC SPECT/CT enable whole-body exploration and therefore the possibility of revealing with a single imaging procedure both cardiac and extracardiac sites of infection.

481

Early detection of embolic events has been reported using [18F]FDG PET/CT with a high sensitivity (87–100%) and specificity (80%) [80] at a reasonable cost-effectiveness ratio, especially in patients with Gram-positive bacteremia [88]. Extra-cerebral peripheral localizations of IE were found in 24–74% among the definite IE population; most of these peripheral localizations were silent (50–71%) and revealed by [18F]FDG PET/CT. In a case-control study, [18F] FDG PET/CT detected peripheral localizations in 57.4% of IE patients, representing the only initially positive imaging technique in about half of the patients with embolic events [89]. Detection of metastatic infection by [18F]FDG PET/CT led to change of treatment in up to 35% of patients [90], with a twofold reduction in the number of relapses [89]. [18F]FDG PET/CT is very accurate in organs with low physiological uptake, therefore not applicable in ruling out the presence of brain embolism [91], where CT/MRI are more sensitive and specific. 99mTc-HMPAO-WBC SPECT/CT shares with PET/ CT the possibility of acquiring whole-body images. [73]. Also in the case of CIED infections, accurate evaluation of the whole-body imaging might detect septic embolisms and identify the possible infection portal of entry, impacting on the subsequent therapeutic management and reducing the risk of relapse. Indeed, in CIED infection the detection of lung embolisms, considered as a major criterion of the Duke score, has shown to increase the diagnostic sensitivity.

20.3.7.2 I dentification of the Infection Portal of Entry The extracardiac diagnostic workup in patients with IE is important not only for the identification of sites of embolism and metastatic infections but also for the identification of the infection portal of entry. Locating the portal of entry lowers the risk of recurrence in patients with IE.  In a recent study, systematic search for the portal of entry identified the site of primary infection in 74% of patients, mainly cutaneous (40%), followed by oral or dental (29%) and gastrointestinal (23%) [92]. [18F]FDG PET/CT reveals the source of infection in most of the patients, including cases where the portal of entry was a neoplasm (colonic cancer) [82]. Several studies support the association between bacteremia or IE due to Streptococcus gallolyticus and Streptococcus infantarius and gastrointestinal disease, mostly represented by colon cancer, adenomatous polyps, diverticulosis, and biliary lesions [93]. Multiple portals of entry are also possible. Once the portal of entry has been identified, risk modification can be performed.

20.3.8 Clinical Background on Vascular Prosthetic Infections Vascular prosthetic infection (VPI) is the most serious complication following surgical or endovascular implantation.

482

Despite a relative low incidence (between 0.5% and 5%) [94], it represents one of the most challenging conditions with very high morbidity and mortality rates (around 50% and 25–75%, respectively). It is more common in the inguinal region (about 13%) followed by aorto-bifemoral bypass and femoropopliteal bypass. Despite the shift in trends of aortic repair in the past two decades with increased proportion of endovascular repairs, the expected reduction in the rates of VPI did not occur. Infection might arise as a consequence of vascular graft or stent-graft perioperative contamination or during bacteremia with consequent seeding of the implant material. Mechanical erosion of the graft/stent-graft into an adjacent structure (esophagus, bronchial system, and duodenum) is a very rare but challenging complication. Depending on the timing of clinical presentation, prosthetic graft infection can be classified as early versus late or very late infections, the threshold for late infection being 4–6  months after the primary surgery or endovascular intervention. Late infection is more indolent and does not present with signs of septicemia; one of the most suggesting signs is the absence of graft incorporation with surrounding tissues and the presence of perigraft fluid (and gas particles) containing large amounts of leukocytes. The majority of cases of VPI are due to Staphylococcus aureus, Escherichia coli, and Staphylococcus epidermidis, whereas Klebsiella, Pseudomonas, Enterobacter, and Proteus account for most of the remaining portion [95]. Diagnosis of VPI is difficult, since no single diagnostic procedure has 100% of accuracy; therefore, a combination of physical examination, laboratory tests, and several imaging techniques is mandatory. In fact, patients may report a variety of clinically equivocal complaints. Furthermore, blood chemistry parameters can only show moderately elevated WBC counts and ESR and/or CRP values, a rather common, non-specific finding. Once a vascular graft infection is suspected, prompt and accurate detection is required for the correct choice of treatment. Success of surgical intervention is closely dependent on early diagnosis [96]. Ultrasonography with color flow Doppler is often a first-­ line imaging procedure, but in case of aortic graft, the predictive value is limited both by air content in the intestinal lumen and, sometimes, by abundant subcutaneous fat. CT angiography is the technique of choice for both confirmation of the infection and the detection of complications, with 94% sensitivity and 85% specificity. While the presence of fluid and air surrounding the aortic graft is a normal finding in the early postoperative period, finding gas in the periprosthetic tissue on the CT scan should be considered abnormal beyond 6–8  weeks after surgery. Despite several advantages (high specificity, guidance for needle aspiration and microbiological analysis, speed of execution), the main limitation of CT imaging is its low sensitivity in detecting early postsurgical infections and low-grade infections [97]. Magnetic resonance imaging (MRI) enables to distinguish between perigraft fluid and perigraft fibrosis, thanks to

M. Bauckneht et al.

signal intensity differences between T1- and T2-weighted images. However, MRI shares the same limitations as CT imaging in the early postoperative period because of non-­ specific signal abnormalities in the vicinity of the perigraft [98]. Nevertheless, MRI accuracy increases to 90–95% when it is performed 3–4 months after surgery.

20.3.8.1 R  adionuclide Imaging of Vascular Prosthetic Infections Nuclear medicine techniques have generally been reserved for cases with equivocal conventional imaging findings or for patients managed by high-expertise multidisciplinary groups including nuclear medicine physicians. Molecular multimodality imaging has demonstrated high accuracy in detecting graft infection in patients with aortic graft and without specific signs of infection (low-grade phases). Overall, 99mTc-HMPAO-WBC imaging has sensitivity ranging from 82% to 100%, with 75–100% specificity. The extensive use of SPECT/CT allows accurate characterization of abnormal foci of 99mTc-HMPAO-WBC accumulation, particularly by visualizing extension of the lesion and by confirming or rejecting graft involvement even in the presence of postsurgical distortions and in complex anatomical sites. In this regard, the use of SPECT/CT is associated with a significant reduction in the false-positive results of planar imaging (i.e., abdominal non-specific accumulation). High specificity is maintained even when scintigraphy is performed during the first month after surgery [99] and also in case of late ­low-­grade VPI [100]. Figures 20.12 and 20.13 show examples of the use of 99mTc-HMPAO-WBC imaging (planar and SPECT/CT) in patients with VPI infections. [18F]FDG PET/CT has emerged as a valuable tool for the evaluation of patients with suspected VPI.  By considering the presence of focal [18F]FDG uptake around the prosthesis as a diagnostic criterion, sensitivity reaches about 93% and specificity 70–91%, with positive and negative predictive values of 82–88% and 88–96%, respectively [101]. Figures 20.14 and 20.15 show examples of the use of these criteria in the diagnosis of VPI infections. A semiquantitative approach, by means of SUVmax and/ or the tissue-to-background ratio, can be also used. More accurate quantification methods are also possible such as, e.g., textural analysis to evaluate [18F]FDG uptake heterogeneity in aortic VPI.  The short-run high gray-level emphasis demonstrated higher values for the infected compared to the uninfected prosthetic grafts. The shortrun high gray-level emphasis was most efficient in identifying aortic graft infection within the suspected group, whereas for the same task, the performances of SUVmax, tissue-to-background ratio, and visual grading scale measurements were all limited [102]. A particular case of cardiovascular infections arises after correction of aorta defects with a special surgical approach, the Bentall procedure. This procedure consists in positioning

20  Radionuclide Imaging of Cardiovascular Disease

a

a1

a2

483

c

c1

30 min

30 min

2h

2h

b

c2

d b1

d1

b2

d2

Fig. 20.12  99mTc-HMPAO-WBC scintigraphy in a patient with suspected late infection of an aortobisiliac vascular prosthesis presenting with cellulitis and erysipelas of the left lower leg and fever. Swab culture isolated Enterobacter sakazakii and Escherichia coli. Anterior (a1 and c1) and posterior (a2 and c2) planar images at 30 min and 2 h after infusion of 99mTc-HMPAO-WBCs at baseline (a) and during follow-up (c). Coronal (b1 and d1) and transaxial (b2 and d2) slices: CT (left) and corresponding fused SPECT/CT sections (right) at baseline (b) and during follow-up (d). The baseline planar and SPECT/CT images show increased 99mTc-HMPAO-WBC accumulation corresponding to the vas-

cular graft. After treatment with antimicrobial therapy, the repeat 99mTc-­ HMPAO-­ WBC scintigraphy showed persistence of uptake at the vascular prosthesis, although less intense than at baseline. Thereafter the patient underwent debridement of the prosthesis and in situ replacement with autologous femoral veins; microbiology showed the presence of Candida albicans (reproduced from: Erba PA, Leo G, Sollini M, Tascini C, Boni R, Berchiolli RN, et al. Radiolabelled leucocyte scintigraphy versus conventional radiological imaging for the management of late, low-grade vascular prosthesis infections. Eur J Nucl Med Mol Imaging. 2014;41:357–368)

a composite aortic graft (combining a vascular tube graft with an attached mechanical or biologic valve) to replace the proximal ascending aorta and the aortic valve. The main indications for performing Bentall surgery are the presence of aortic regurgitation, Marfan’s syndrome, aortic dissection, and aortic aneurysm. Infections after Bentall surgery are reported in about 3% of the cases [103]; despite such relatively low incidence, the risk of death is high in patients affected by this

condition. There is a predominance of Staphylococcus aureus infections (35%), with a recent 20% increase in methicillinresistant Staphylococcus aureus strains [104]. The goal of diagnostic imaging in patients with infection after the Bentall procedure is to demonstrate and distinguish the presence of aortic valve-root involvement or of infection localized to the vascular aortic graft. These two conditions might also coexist and involve the surrounding structures

484

Fig. 20.13  99mTc-HMPAO-WBC scintigraphy in a patient with suspected late infection of a right iliofemoral and femoropopliteal vascular prosthesis. Planar anterior and posterior images at 30 min and at 4 h shown in upper left panel, demonstrating increased 99mTc-HMPAO-­ WBC accumulation in the right inguinal region. Right panels show axial and coronal sections at different levels (from left to right: SPECT, CT, fused SPECT/CT images) localizing the focal accumulation of radiolabelled leucocytes to the vascular graft in both the iliac and the

M. Bauckneht et al.

femoral portions. Microbiology showed that the infection was caused by polymicrobial agents. Lower left panel: 3D surface rendering SPECT/CT images better demonstrating 99mTc-HMPAO-WBC accumulation all along the vascular graft (reproduced from: Erba PA, Leo G, Sollini M, Tascini C, Boni R, Berchiolli RN, et al. Radiolabelled leucocyte scintigraphy versus conventional radiological imaging for the management of late, low-grade vascular prosthesis infections. Eur J Nucl Med Mol Imaging. 2014;41:357–368)

Fig. 20.14 [18F]FDG PET/CT images show increased uptake at the site of an aortic aneurysm (MIP in left panel; right panel from left to right, CT, emission PET, and fused PET/CT images in the coronal, transaxial, and sagittal views)

20  Radionuclide Imaging of Cardiovascular Disease

485

Fig. 20.15 [18F]FDG PET/CT images show increased tracer uptake at the thoracic aortic prosthesis after Bentall procedure (MIP in left; right panels, coronal emission PET, CT, and fused PET/CT images)

such as the mediastinal soft tissue and the sternum. Identification of patients with infection limited to the vascu• However, the limited spatial resolution hampers lar portion of the thoracic aortic grafts (VPI, about 2%) is its diagnostic sensitivity, even when SPECT is very important since the ideal treatment, replacement of the combined with MDCT in the hybrid SPECT/CT graft, carries a high mortality, especially in cases of long-­ approach. lasting infections or severe comorbidities [105]. No specific • PET/CT imaging displays the high avidity of guidelines for the management of aortic valve-root-vascular inflammatory infiltrates for [18F]FDG. prosthesis infections are available, the standard • When using [18F]FDG PET/CT for imaging patients ­recommendations for infectious endocarditis and prosthetic with suspected infective endocarditis, special attengraft infections being usually followed, including echocartion must be paid to time interval from surgery, diography and ce-CT and/or MRI in short interval. TEE patient preparation, image reconstruction, and evalplays a key role in the assessment of IE, but is not always uation criteria in order to maximize its diagnostic conclusive due to the numerous artifacts related to the prespotential for infective endocarditis. ence of the prosthesis. In most of the published reports, the diagnosis of infection in composite aortic grafts requires a combination of echocardiography, CT, and PET/CT [106] (see Fig.  20.15). Also in the presence of infection after 20.4 Clinical Indications, Technique, Bentall procedure in relation to possible comorbidities, the and Principles of Noninfectious, risk of distant sites of infections should not be underestiNon-atherosclerotic Cardiovascular mated. Prompt extracardiac workout will allow identification Diseases of both embolic events or concomitant source of infection/ 20.4.1 Clinical Background on Myocarditis inflammation.

Key Learning Points



99m Tc-HMPAO-WBC SPECT is probably the radionuclide imaging technique with the longest experience in patients with suspected infective endocarditis. • This technique is highly specific both in endocarditis of native structures and in prosthetic valve endocarditis (PVE).

Myocarditis, or inflammation of the heart muscle, may present as an acute, subacute, or chronic illness. A large proportion of individuals with this condition may be asymptomatic. The causes of myocarditis include a variety of infectious or systemic diseases, toxins, and drugs. Viruses, especially enteroviruses, are the most important causes of myocarditis in developed countries. The enterovirus genome has been identified in the myocardium of patients with myocarditis or with dilated cardiomyopathy. Until the introduction of

486

e­ ndomyocardial biopsy in 1962, the diagnosis of myocarditis ­during life was often presumptive and not always correct, mainly in young patients with heart failure preceded by a febrile illness. Most cases were diagnosed as late as at autopsy. Prospective and retrospective studies have identified myocardial inflammation in 1–9% of routine postmortem examinations. Myocarditis is a major cause of sudden, unexpected death (up to 20% of cases) in adults younger than 40 years. According to the 1995 World Health Organization (WHO)/International Society and Federation of Cardiology Task Force on the Definition and Classification of Cardiomyopathy, inflammatory cardiomyopathy is now included as a subtype of specific cardiomyopathies, defined as myocarditis in association with cardiac dysfunction. Infectious, autoimmune, and idiopathic forms of inflammatory heart disease have been recognized. In 1986, the Dallas criteria for the histologic diagnosis of myocarditis were defined, based on the identification of infiltrating lymphocytes and of myocytolysis [107]. Patients with lymphocytic infiltration, but without myocytolysis, were classified as having borderline or ongoing myocarditis. However, less than 10% of the patients with suspected myocarditis had positive biopsies when assessed by the Dallas criteria, thus raising the issue of low sensitivity and high inter-observer variability. A second clinico-pathological classification system was proposed in 1991 but has not been widely accepted [108]. Although there have not been major advances in the identification of the etiology of myocarditis in recent years, new molecular techniques such as polymerase chain reaction and genomic hybridization have allowed confirmation of the etiology in some cases, which would have otherwise remained undiagnosed. The serum level of creatine kinase-MB (CK-­ MB) has high specificity but limited sensitivity for the diagnosis of myocarditis; on the other hand, serum troponin-I is more often elevated than CK-MB in patients with myocarditis. Echocardiography performed in the setting of acute myocarditis may demonstrate either normal heart function or global or regional left ventricular hypokinesis, with an overall low sensitivity [109]. MRI allows discrimination of myocarditis from myocardial infarction by depicting scattered areas of hyper-­enhancement with a nonvascular distribution (in mid-myocardial or subepicardial locations) in patients with myocarditis. These areas of hyper-enhancement correspond to inflammation and cell necrosis and are most commonly seen in the inferior and inferolateral myocardial segments [110].

20.4.1.1 Radionuclide Imaging of Myocarditis Diffusely increased metabolic activity in the left ventricular myocardium with mild heterogeneity suggesting the occurrence of myocarditis has been observed at [18F]FDG PET/CT

M. Bauckneht et al.

in a patient with chronic active EBV infection presenting with fever, dyspnea on exertion, general malaise, and hepatosplenomegaly; the patient subsequently developed heart failure and myocarditis was confirmed by endomyocardial biopsy. This condition must be distinguished from congestive heart failure, right ventricular strain, and hypertrophy due to elevated pulmonary artery pressure, which can also lead to increased [18F]FDG uptake—but in the right ventricular myocardium [111]. Tuberculous and viral myocarditis should be considered for the differential diagnosis in patients with heterogeneous [18F]FDG uptake at PET/CT imaging performed for diagnosing infiltrative cardiomyopathy. Unlike tuberculosis and sarcoidosis, viral myocarditis does not typically manifest with perfusion defects. Given the similar appearance of myocardial [18F]FDG uptake in sarcoidosis and tuberculosis by PET imaging, a detailed medical history and histologic correlation are crucial for differentiating tuberculous myocarditis and sarcoidosis [112].

Key Learning Points

• The condition of myocarditis (a major cause of sudden, unexpected death in up to 20% in adults aged 20-fold over the last 25  years, reaching 0.31  in 100,000 adults between 2008 and 2012. The clinical manifestations of CS relate primarily to the location and inflammatory effects of the granulomas. The presence of granulomas in the ventricular myocardium may lead to abnormal automaticity and re-entrant tachyarrhythmias manifesting as palpitations or syncope. Involvement of the conduction system may lead to bradyarrhythmias and syncope. Congestive heart failure may result from widespread sarcoidosis of the myocardium, with progressive decline in left ventricular ejection fraction (EF) and death. The revised Japanese Ministry of Health and Welfare (JMHW) guidelines for the diagnosis of CS are currently used in many studies as a reference. A more recent publication by the Heart Rhythm Society (HRS), in collaboration with other organizations including the World Association for Sarcoidosis and Other Granulomatous Disorders (WASOG), published an expert consensus recommendation on criteria for the diagnosis of CS [118].

20.4.3.1 R  adionuclide Imaging of Cardiac Sarcoidosis Due to the multifocal, patchy pattern of myocardial sarcoid involvement, the sensitivity of endomyocardial biopsy is as low as 20% [119]. Whereas, even patchy active inflammation can be detected as patchy areas with increased [18F]FDG uptake on the PET/CT scan, thus providing important information not only for diagnostic and/or prognostic purposes but also to guide endomyocardial biopsy [120]. The prognostic information provided by [18F]FDG PET/CT is enhanced when combined with perfusion imaging; in particular, patients with concomitant abnormalities on the 82Rb PET scan and on the [18F]FDG scan had a hazard ratio of 3.9 for ventricular tachycardia and death [121]. Imaging modalities recommended in the documents mentioned above include [18F]FDG PET or PET/CT, 67Ga-citrate scintigraphy, and cardiac MRI. Although 67Ga-citrate scintigraphy is included in the major diagnostic criteria for CS, its sensitivity for detecting CS is reported to be as low as 36.4%, which is significantly lower than that of [18F]FDG PET.  Therefore 67Ga-citrate scintigraphy has largely been replaced with [18F]FDG PET/CT for diagnosing CS. In fact, [18F]FDG PET/CT has several practical and technical advantages over 67Ga-citrate scintigraphy, including favorable tracer kinetics, lower radiation exposure, and better quality images. Furthermore, [18F]FDG PET/CT is more accurate than 67Ga-citrate and allows better evaluation of extrapulmonary involvement, including CS. Hence, [18F]FDG

M. Bauckneht et al.

PET/CT imaging is currently most commonly used for imaging myocardial inflammation in clinical practice. Nevertheless, [18F]FDG PET/CT has not been officially included in univocally accepted diagnostic guidelines. This is due to its relatively high sensitivity but low specificity and high variability, which limits the use of [18F]FDG PET/CT for strictly diagnosing CS. The main reason for low specificity and high variability of this imaging technique is physiological [18F]FDG uptake in the normal myocardium. Despite these limitations [18F] FDG PET/CT imaging is most commonly used for imaging myocardial inflammation in clinical practice. In sarcoidosis, the uptake of [18F]FDG is related to inflammatory cell infiltrates. Since [18F]FDG accumulates in active lesions where inflammatory cells utilize glucose as an energy source, active sarcoid lesions in various organs are visualized by increased [18F]FDG uptake on PET/CT imaging. Currently used strategies for myocardial suppression include prolonged fasting, dietary modifications, and a heparin load before imaging, as reported in a recent meta-analysis by Tang et al. [122]. This preparation (especially heparin administration) will also improve accuracy of the procedure. [18F]FDG ­imaging starts after an uptake period of 60–90  minutes. Following CT for attenuation correction purpose, non-gated cardiac [18F]FDG imaging is acquired first. Thereafter, whole-body images are obtained along with a CT for ­attenuation correction. [18F]FDG PET is often combined with a cardiac perfusion scan. Comparison of the perfusion images with the [18F]FDG PET images will allow for differentiation of scar tissue from normal myocardium or change of active sarcoid granuloma to normal myocardium or progression to scar tissue in response to therapy. It is also important to exclude epicardial coronary artery disease in these patients (with either invasive or CT-based coronary angiography), since the patterns of perfusion/metabolism mismatch in patients with left ventricular systolic dysfunction may reflect either hibernating myocardium or myocardial inflammation according to the clinical context. CS can affect any part of the heart, the most frequently involved area being the ventricular septum. Several visual score have been used for evaluating the intensity of [18F]FDG uptake, with or without the combination with myocardial perfusion with either [13N]ammonia or 82Rb PET imaging. [18F]FDG PET/CT has 89% sensitivity and 78% specificity for CS detection [123]. In addition, patients with CS present a significantly higher SUVmax than non-CS patients. With a cutoff value of 4.0 for myocardial SUVmax, sensitivity was 97.3% and specificity 83.6% for diagnosing CS. The specificity of quantitative analysis was greater than that of visual analysis. Moreover, SUVmax on [18F]FDG PET/CT was the only independent predictor among clinical and imaging variables for diagnosing CS. Therefore, [18F]FDG PET/CT using quantitative analysis is expected to facilitate correct diagnosis of CS.  Several studies compared late enhancement in CMR and [18F]FDG uptake in PET/CT, showing only mild to

20  Radionuclide Imaging of Cardiovascular Disease

489

Fig. 20.17 [18F]FDG PET/CT in a young man with sudden arrhythmic episodes. Sarcoidosis was diagnosed by biopsy of a mediastinal lymph node guided by the PET/CT images. The images show increased [18F]

FDG uptake in multiple mediastinal lymph nodes, as well as cardiac involvement by sarcoidosis (left panel MIP, right panel transaxial emission, CT, and fused PET/CT at different levels)

moderate correlation. This finding is due to the fact that, while delayed enhancement represents cardiac damage and scarring, [18F]FDG uptake represents active inflammation. When CMR and [18F]FDG PET were compared with the Japanese Ministry of Health and Welfare guidelines, CMR had a higher specificity but a lower sensitivity. Figure 20.17 shows an example of CS evaluated with [18F]FDG PET. [18F]FDG PET can also be used to assess response to therapy [124]. Following corticosteroid therapy, [18F]FDG uptake in cardiac lesions declines. Similarly, in a retrospective study of 23 patients with cardiac sarcoidosis treated with immunosuppressive therapies guided by serial PET scans, reduction in the intensity and extent of inflammation, as quantified by [18F]FDG uptake, was associated with increasing LVEF values.

20.4.4 Clinical Background on Cardiac Amyloidosis

Key Learning Points

• Non-infective active inflammation (e.g., cardiac sarcoidosis) can be detected as patchy uptake of [18F] FDG on PET/CT, linked to the inflammatory cells infiltrate accumulation within myocardial walls. • Moreover, [18F]FDG PET/CT imaging has been shown to be able to guide endomyocardial biopsy and to predict response to therapy. • Physiologic myocardial [18F]FDG uptake requires dedicated patient preparation to avoid the pitfall of inadequately suppressed basal uptake.

Cardiac amyloidosis (CA) may be genetic/familial (ATTR) or non-genetic/non-familial (AL/prealbumin, senile); the genetic variant usually involves deposition of the transthyretin/prealbumin protein (and is therefore defined as the “ATTR” form), while the non-familial variant usually involves deposition of light-chain amyloid (and is therefore defined as the “AL” form). Imaging plays a major role in assessing patients with cardiac amyloidosis, according to the following clinical scenarios: • Differential diagnosis between hypertrophic and restrictive cardiomyopathy. • Evaluation of cardiac involvement in patients with either form of amyloidosis. • Staging of the disease and monitoring of response to therapy. Echocardiography is a first-line technique in cardiac amyloidosis and is recommended for the diagnosis, functional and anatomical characterization, differential diagnosis, and prognostic stratification of these patients. Nevertheless, it provides relatively late non-specific information on thickening of amyloidotic myocardium. Gadolinium-based contrast MR (CMR) offers particular advantages in patients with suspected cardiac amyloidosis, because of its high spatial resolution and its ability to characterize myocardial tissue. CMR is often used after CA is

490

M. Bauckneht et al.

s­ uspected by echocardiography to confirm or refute the diagnosis and in experienced hands represents a powerful tool with important diagnostic, with reasonably high sensitivity [125] and prognostic implications. Late gadolinium enhancement (LGE) is typically observed in a circumferential subendocardial pattern or with bilateral subendocardial LGE of the septum with a dark mid-wall (zebra pattern); combination of this finding with altered gadolinium blood pool kinetics is highly suggestive for cardiac amyloidosis. Patterns of focal LGE may also be found but are less specific for cardiac amyloidosis. Myocardial non-contrast T1 values are longer in cardiac amyloidosis than in controls, a finding with higher sensitivity for detecting early subclinical cardiac involvement than LGE. It should be noted that neither echocardiography nor CMR enable to distinguish between the two variants of cardiac amyloidosis.

several phosphate derivatives labelled with 99mTc, resulting in the current clinical use of the following tracers: 99mTc-­2,3-­ dicarboxypropane-­1,1-diphosphonate (99mTc-DPD) mainly in Europe and Asia and 99mTc-pyrophosphate (99mTc-PYP) in the United States. Their main advantage is avid uptake by ATTR amyloidosis and minimal uptake with the AL variant, thus providing one of the best noninvasive ways to differentiate these two subtypes of cardiac amyloidosis [126]. In AL amyloidosis, cardiac uptake is found in less than half of patients and is generally less intense (likely due to the lower concentration of calcium-containing products in AL amyloid). Furthermore, AL patients generally exhibit no muscular 99m Tc-DPD or 99mTc-HDP uptake, while visceral uptake (liver, spleen) is more common (see Fig. 20.18). 99mTc-DPD has also been shown to provide independent prognostic value. Scintigraphy with either 99mTc-DPD or 99mTc-PYP scintigraphy shows early signs of amyloid involvement of myo20.4.4.1 Radionuclide Imaging of Cardiac cardial tissue and might be helpful in distinguishing between Amyloidosis ATTR and AL amyloidosis [126]. The most reasonable Radiolabelled bone-seeking phosphates were incidentally explanation for accumulation of 99mTc-labelled bisphosphonoted to localize in amyloid deposits using 99mTc-­ nates in the amyloidotic myocardium is related to their affindiphosphonate. This observation led to the development of ity for the high calcium content of transthyretin amyloid. a

99mTc-PYP

b

99mTc-DPD

Fig. 20.18  Example of cardiac SPECT in a patient with ATTR amyloidosis. 99mTc-PYP (a) and 99mTc-DPD (b) (courtesy of Selene Capitanio, MD, Nuclear Medicine Unit, University Hospital of Trieste, Trieste, Italy)

20  Radionuclide Imaging of Cardiovascular Disease

The intensity of 99mTc-DPD uptake in ATTR patients is negatively correlated with the LV ejection fraction assessed by echocardiography, being a predictor of major adverse cardiac outcomes. Moreover, the concomitance of 99mTc-PYP uptake and of myocardial perfusion defects suggests the opportunity to combine the 99mTc-PYP scan with SPECT or PET perfusion imaging, for enhanced diagnostic accuracy. After intravenous administration of 740  MBq of 99mTc-­ DPD/HDP or 99mTc-PYP, a whole-body scan is performed 3 h or 1 h later (anterior and posterior projections). If there is active uptake in the heart, chest SPECT is performed (Fig.  20.18). The analysis is performed by semiquantitative visual scoring of cardiac uptake as compared to bone uptake (scores from 0 to 3) and by computing the ratio, after correction for background activity, of the mean counts in the heart region divided by the mean counts in the contralateral chest (H/CL ratio). Hence, planar 99mTc-DPD/HDP scintigraphy helps to phenotype cardiac amyloidosis, particularly for distinguishing of ATTR from AL amyloidosis. This property is shared by PET/CT with 18 F-fluoride, which can distinguish ATTR from AL cardiac amyloidosis [127] with similar diagnostic accuracy but greater spatial resolution. In this regard, it can be speculated that PET/ MR with 18F-fluoride may combine optimal diagnostic performance of the two imaging modalities. Finally, novel PET tracers originally developed for brain beta-amyloid imaging, such as the 11C-labelled Pittsburgh compound B ([11C]PiB) [128] or 18F-florbetapir [129, 130], appear promising for imaging cardiac amyloidosis (particularly AL amyloidosis) and are currently under clinical investigation.

Key Learning Points

• Scintigraphy with either 99mTc-DPD or 99mTc-PYP shows early signs of amyloid involvement of myocardial tissue and might be helpful for distinguishing ATTR from AL amyloidosis. • Excellent agreement has been shown between echocardiography findings and 99mTc-DPD uptake in ATTR patients, the intensity of 99mTc-DPD uptake being negatively correlated with the LV ejection fraction. • 99mTc-DPD uptake in the myocardium predicts major adverse cardiac outcomes. • Although PET/CT has been intensively investigated, its clinical role is still limited. • While [18F]FDG PET/CT imaging does not accurately display cardiac involvement in patients with systemic amyloidosis, several non-[18F]FDG tracers show promising results. • Since either modality is potentially able to provide complementary information, PET/MRI might become the hybrid imaging method of choice for patients with cardiac amyloidosis.

491

20.5 Heart Failure Cardiac pump dysfunction is the final endpoint of a wide range of cardiac disorders including ischemic heart disease, replacement fibrosis, infiltration, inflammation, and myocardial iron deposition. Establishing the specific etiology of heart failure (HF) is crucial because it is associated with prognosis and it guides specific therapies. Multimodality imaging in the setting of HF aims to join the different physical bases of image formation to create an image set with complementary (or even synergistic) information to demonstrate the occurrence of cardiac dysfunction, to display the underlying structural heart disease, and finally to determine the HF mechanisms (HF with reduced or preserved systolic function) [131]. In patients with ischemic HF, identifying residual viable myocardium has a major impact on prognosis and therapy. Current guidelines recommend myocardial revascularization in patients with ejection fraction ≤35% only in the presence of viable myocardium. Detection of residual myocardial glucose uptake with [18F]FDG PET/CT is a widely accepted approach to identify these patients with viable (increased [18F]FDG uptake) myocardium, therefore with potential for recovery of contractile function after revascularization. On the other hand, a matching flow-metabolic defect indicates scar formation. The substrate and hormonal levels in the blood must switch metabolism in the myocardium to glucose over fatty acids. To achieve this, a glucose load (25–100 g) is administered orally to induce an endogenous insulin response and the consequent overexpression of GLUT-4 on cardiomyocytes. Intravenous loading may also be used. Supplemental insulin administration can be also considered, as needed. In case of diabetic patients, a major challenge occurs, either because they have limited endogenous insulin production or because their cells respond less to insulin stimulation. Potential alternative protocols in this clinical scenario are insulin administration along with close monitoring of blood glucose or to delay image acquisition from 2 to 3 h after [18F] FDG injection (see also Tables 20.7 and 20.8). In a pooled meta-analysis of 24 studies in 756 patients, the weighed mean sensitivity and specificity of [18F]FDG PET/CT for myocardial viability were 92% and 63%, respectively [132]. A wide series of radiotracers have also been proposed to investigate myocardial innervation of the failing heart [133]. Currently, 123I-metaiodobenzylguanidine (123I-MIBG), a radioiodinated norepinephrine analogue, is the most clinically used for myocardial SPECT imaging of sympathetic function. 123I-MIBG is administered i.v. in the amount of 185  MBq over 1–2  min, after thyroid blockade by oral administration of 400 mg of potassium perchlorate. Ten-min planar images of the thorax in standard anterior view (with matrix 128 × 128) are recorded at 30 min (early image) and 3–4 h (late image) after tracer administration. For early and

492

M. Bauckneht et al.

late planar images, the heart-to-mediastinum (H/M) ratio is computed by dividing the mean counts per pixel within the myocardium by the mean counts per pixel within the mediastinum. The washout rate (WR) is calculated using the following equation:

WR = ( early H / M - late H / M ) / early H / M ´ 100.

Various quantitation methods for measuring myocardial uptake have been proposed, although the simplest index is the heart-to-mediastinum ratio (HMR), which is a crude but practical parameter. However, the practical simplicity of the HMR does not necessarily result in its consistency, as various factors are known to cause variations in this score; in particular, choice of the collimator introduces considerable variations [134, 135]. Since the HMR is determined by an averaged count ratio of heart and mediastinum, the location, size, and shape of the region of interest (ROI) can all induce variability. Although in clinical practice such variations between different centers exist, reproducibility of the HMR is generally considered to be good for inter- and intra-patient comparisons in the same center [136]. However, some ­variations still exist, depending on ROI setting methods; predefined or semiautomatic methods for ROI definition are preferable [137, 138]. There is no general consensus even regarding the prognostic threshold of 1.6 that has been proposed for the HMR. Many studies have demonstrated that cardiac uptake of 123 I-MIBG is reduced in HF patients, thus suggesting a prognostic power for myocardial innervation imaging [139]. However, widespread application of 123I-MIBG imaging is still limited, partly due to suboptimal target-tobackground ratios and to the use of planar imaging for quantitation of myocardial activity relative to background activity in the mediastinum or for the estimation of myocardial washout.

Key Learning Points

• In ischemic HF, identification of residual viable myocardium profoundly impacts prognosis and guides therapy. • Assessment of residual myocardial viability with [18F]FDG PET/CT identifies ischemic myocardium that is dysfunctional but viable and, therefore, has potential for recovery of contractile function after revascularization. • Different radiotracers have been proposed to explore myocardial innervation of the failing heart. • Currently, 123I-MIBG is the most widely used tracer for myocardial planar and/or SPECT imaging of sympathetic function.

• Cardiac uptake of 123I-MIBG is reduced in HF patients, indicating the prognostic power of myocardial innervation imaging. • For early and late planar images, the heart-to-mediastinum (H/M) ratio is computed by dividing the mean counts per pixel within the myocardium by the mean counts per pixel within the mediastinum. • The washout rate (WR) is calculated using the following formula: (early H/M  –  Late H/M)/early H/M × 100.

20.6 Large Vessels Vasculitis Large vessel vasculitis consists of a group of conditions causing inflammation of the aorta and its branches. It is an uncommon but well-known cause of non-specific symptoms that constitute a relevant diagnostic challenge. Large vessel vasculitis includes giant cell arteritis (GCA), Takayasu’s arteritis, and isolated aortitis; other associated conditions are Behçet’s disease, Cogan’s syndrome, systemic lupus erythematosus, spondyloarthritis, and sarcoidosis. Other types of vasculitis, such as the ANCA-positive vasculitis and polyarteritis nodosa, may occasionally involve the large vessels walls. [18F]FDG PET, CT coronary angiography (CTCA), and MRI demonstrate that extracranial involvement in GCA is quite frequent, occurring in 30–74% of patients. In patients with typical clinical presentation, the diagnosis of GCA is not difficult, and temporal artery biopsy (TAB) is considered to be the “gold standard” for diagnostic confirmation, although hampered by low sensitivity (with 15–40% false-­ negative rate). Instead, diagnosis may be challenging when symptoms are non-specific, given the wide range of clinical manifestations of GCA. Due to the inflammatory nature of the disease, [18F]FDG PET/CT can detect vascular involvement in GCA patients. Other imaging techniques (i.e., Doppler ultrasonography, CT, and MRI) have been proposed for the assessment of vascular inflammation in GCA patients. However, while they are able to demonstrate anatomic changes in the affected vessels (wall thickening, dilatation and aneurysms, and enhancement of perivascular connective tissue) if the inflammatory process is advanced, they are not sensitive enough to diagnose early inflammatory changes, which are potentially reversible following therapy. Furthermore, patient follow-up and assessment of response to medical treatment are not easy to perform on the basis of morphological information alone [140]. The main limitation of [18F] FDG PET/CT as a reliable diagnostic tool is the lack of a standardized definition of vascular inflammation based on the intensity of [18F]FDG uptake.

20  Radionuclide Imaging of Cardiovascular Disease

Qualitative and semiquantitative methods for assessing the presence and for grading the severity of GCA-related vascular inflammation on [18F]FDG PET/CT imaging have been proposed, qualitative analysis of [18F]FDG uptake being currently the most widely adopted approach. Qualitative analysis has been used both to perform dichotomous assessment (confirm or exclude the presence of vascular inflammation) [141] and to grade the severity of vascular involvement according to ordinal scales, [18F]FDG uptake in the vessel wall being visually analyzed or compared with that of a reference structure. Among the visual grading systems, the vessel-to-liver ratio is the most frequently adopted. The advantage of qualitative methods, in addition to being more immediate and rapid than semiquantitative methods, lies in their high specificity, with sensitivity ranging from 56% to 77%. Moreover, qualitative methods are characterized by high inter-observer agreement (90%) and intra-observer reproducibility (93.3%). Nevertheless, the diagnostic discrimination between vasculitis and other vascular diseases (e.g., atherosclerosis) is still challenging. As to the semiquantitative methods, several authors have proposed arterial SUVmax as a diagnostic factor [140], while others have normalized arterial SUVmax to background activity, represented by the mean uptake value of a selected reference structure. When comparing the SUVmax-based approach without normalization with simple visual qualitative analysis, high sensitivity and low specificity has been reported by using a cutoff value. Nevertheless, SUVmax cutoff values obtained without normalization appear to be population-­specific, being therefore difficult to apply to the general population. In a recent meta-analysis, Puppo et al. found that semiquantitative methods normalized to background activity seem to outperform qualitative approaches and semiquantitative methods without normalization [140]. They also noticed that the power of discrimination between GCA and control groups also depends on the anatomical structure chosen as the background. An emerging semiquantitative approach is the arterial-to-background ratio proposed by Moosig et  al. [142], which outperformed the other methods where liver and lung are used as reference structures for background.

Key Learning Points

• Due to the inflammatory nature of the disease, [18F] FDG PET/CT imaging is a feasible noninvasive method to detect vascular involvement in patients with large vessel vasculitis. • The main limitation of [18F]FDG PET/CT as a reliable diagnostic tool is the lack of a standardized definition of vascular inflammation based on the intensity of [18F]FDG uptake.

493

• Several qualitative and semiquantitative methods for assessing the presence and for grading the severity of vasculitis-related inflammation on [18F]FDG PET/CT imaging have been proposed. • For quantitative analysis, the SUV-based approach has not been sufficiently validated for evaluating inflammation and infection. • SUV-based methods should be used with caution in the clinical practice.

References 1. Rossi A, Uitterdijk A, Dijkshoorn M, Klotz E, Dharampal A, van Straten M, et al. Quantification of myocardial blood flow by adenosine-stress CT perfusion imaging in pigs during various degrees of stenosis correlates well with coronary artery blood flow and fractional flow reserve. Eur Heart J Cardiovasc Imaging. 2013;14(4):331–8. 2. Marini C, Seitun S, Zawaideh C, Bauckneht M, Morelli MC, Ameri P, et  al. Comparison of coronary flow reserve estimated by dynamic radionuclide SPECT and multi-detector X-ray CT. J Nucl Cardiol. 2017;24:1712–21. 3. Glagov S, Weisenberg E, Zarins CK, Stankunavicius R, Kolettis GJ.  Compensatory enlargement of human atherosclerotic coronary arteries. N Engl J Med. 1987;316(22):1371–5. 4. Libby P, Pasterkamp G. Requiem for the 'vulnerable plaque. Eur Heart J. 2015;36(43):2984–7. 5. Hackett D, Davies G, Maseri A. Pre-existing coronary stenoses in patients with first myocardial infarction are not necessarily severe. Eur Heart J. 1988;9(12):1317–23. 6. Sambuceti G, Marini C, Morbelli S, Paoli G, Derchi M, Pomposelli E. Witnessing ischemia or proofing coronary atherosclerosis: two different windows on the same or on different pathways precipitating cardiovascular events? J Nucl Cardiol. 2009;16(3):447–55. 7. Gould KL, Lipscomb K, Hamilton GW.  Physiologic basis for assessing critical coronary stenosis. Instantaneous flow response and regional distribution during coronary hyperemia as measures of coronary flow reserve. Am J Cardiol. 1974;33(1):87–94. 8. Sambuceti G, Marzilli M, Mari A, Marini C, Schluter M, Testa R, et  al. Coronary microcirculatory vasoconstriction is heterogeneously distributed in acutely ischemic myocardium. Am J Physiol Heart Circ Physiol. 2005;288(5):H2298–305. 9. Saracco A, Bauckneht M, Verna E, Ghiringhelli S, Repetto R, Sambuceti G, Provasoli S, Storace M. A mathematical model for the vessel recruitment in coronary microcirculation in the absence of active autoregulation. Microvasc Res. 2016;104:38–45. 10. Sambuceti G, Marzilli M, Fedele S, Marini C, L’Abbate A.  Paradoxical increase in microvascular resistance during tachycardia downstream from a severe stenosis in patients with coronary artery disease: reversal by angioplasty. Circulation. 2001;103(19):2352–60. 11. Marini C, Acampa W, Bauckneht M, Daniele S, Capitanio S, Cantoni V, et al. Added prognostic value of ischaemic threshold in radionuclide myocardial perfusion imaging: a common-sense integration of exercise tolerance and ischaemia severity. Eur J Nucl Med Mol Imaging. 2015;42(5):750–60. 12. Sambuceti G, Giorgetti A, Corsiglia L, Marini C, Schneider-­ Eicke J, Brunelli C, et al. Perfusion-contraction mismatch during inotropic stimulation in hibernating myocardium. J Nucl Med. 1998;39(3):396–402.

494 13. Ward RP, Al-Mallah MH, Grossman GB, Hansen CL, Hendel RC, Kerwin TC, et al. American Society of Nuclear Cardiology review of the ACCF/ASNC appropriateness criteria for single-photon emission computed tomography myocardial perfusion imaging (SPECT MPI). J Nucl Cardiol. 2007;14(6):e26–38. 14. Hesse B, Lindhardt TB, Acampa W, Anagnostopoulos C, Ballinger J, Bax JJ, et  al. EANM/ESC guidelines for radionuclide imaging of cardiac function. Eur J Nucl Med Mol Imaging. 2008;35(4):851–85. 15. Hendel RC, Berman DS, Di Carli MF, Heidenreich PA, Henkin RE, Pellikka PA, et  al. ACCF/ASNC/ACR/AHA/ASE/SCCT/ SCMR/SNM 2009 appropriate use criteria for cardiac radionuclide imaging: a report of the American College of Cardiology Foundation Appropriate Use Criteria Task Force, the American Society of Nuclear Cardiology, the American College of Radiology, the American Heart Association, the American Society of Echocardiography, the Society of Cardiovascular Computed Tomography, the Society for Cardiovascular Magnetic Resonance, and the Society of Nuclear Medicine. J Am Coll Cardiol. 2009;53(23):2201–29. 16. Verberne HJ, Acampa W, Anagnostopoulos C, Ballinger J, Bengel F, De Bondt P, et al. EANM procedural guidelines for radionuclide myocardial perfusion imaging with SPECT and SPECT/CT: 2015 revision. Eur J Nucl Med Mol Imaging. 2015;42(12):1929–40. 17. Henzlova MJ, Duvall WL, Einstein AJ, Travin MI, Verberne HJ.  ASNC imaging guidelines for SPECT nuclear cardiology procedures: stress, protocols, and tracers. J Nucl Cardiol. 2016;23(3):606–39. 18. Hesse B, Tagil K, Cuocolo A, Anagnostopoulos C, Bardies M, Bax J, et  al. EANM/ESC procedural guidelines for myocardial perfusion imaging in nuclear cardiology. Eur J Nucl Med Mol Imaging. 2005;32(7):855–97. 19. Taillefer R. Should early post-stress imaging be performed on a routine clinical basis for myocardial perfusion studies? J Nucl Cardiol. 2014;21(6):1177–80. 20. Mut F, Giubbini R, Vitola J, Lusa L, Sobic-Saranovic D, Peix A, et al. Detection of post-exercise stunning by early gated SPECT myocardial perfusion imaging: results from the IAEA multi-­ center study. J Nucl Cardiol. 2014;21(6):1168–76. 21. Kaufmann BA, Pfisterer ME, Viswanathan S, Muller-Brand J, Zellweger MJ.  Stunning and left ventricular function--how long is the ventricle knocked out? Left ventricular function correlated with ischemic burden and recovery time after stress. Int J Cardiol. 2006;112(2):223–8. 22. Emmett L, Iwanochko RM, Freeman MR, Barolet A, Lee DS, Husain M. Reversible regional wall motion abnormalities on exercise technetium-99m-gated cardiac single photon emission computed tomography predict high-grade angiographic stenoses. J Am Coll Cardiol. 2002;39(6):991–8. 23. Johnson LL, Verdesca SA, Aude WY, Xavier RC, Nott LT, Campanella MW, et  al. Postischemic stunning can affect left ventricular ejection fraction and regional wall motion on post-stress gated sestamibi tomograms. J Am Coll Cardiol. 1997;30(7):1641–8. 24. Heller GV, Mann A, Hendel R.  Nuclear cardiology: technical applications. New York: McGraw-Hill; 2009. 25. Yoshinaga K, Klein R, Tamaki N.  Generator-produced rubidium­82 positron emission tomography myocardial perfusion imagingfrom basic aspects to clinical applications. J Cardiol. 2010;55(2): 163–73. 26. Dilsizian V, Bacharach SL, Beanlands RS, Bergmann SR, Delbeke D, Dorbala S, et al. ASNC imaging guidelines/SNMMI procedure standard for positron emission tomography (PET) nuclear cardiology procedures. J Nucl Cardiol. 2016;23(5):1187–226. 27. Maddahi J, Packard RR.  Cardiac PET perfusion tracers: current status and future directions. Semin Nucl Med. 2014;44(5):333–43.

M. Bauckneht et al. 28. Di Carli MF, Dorbala S, Meserve J, El Fakhri G, Sitek A, Moore SC. Clinical myocardial perfusion PET/CT. J Nucl Med. 2007;4:783–93. 29. Huisman MC, Higuchi T, Reder S, Nekolla SG, Poethko T, Wester HJ, et al. Initial characterization of an 18F-labeled myocardial perfusion tracer. J Nucl Med. 2008;49(4):630–6. 30. Yu M, Guaraldi MT, Mistry M, Kagan M, McDonald JL, Drew K, et al. BMS-747158-02: a novel PET myocardial perfusion imaging agent. J Nucl Cardiol. 2007;14(6):789–98. 31. Nekolla SG, Reder S, Saraste A, Higuchi T, Dzewas G, Preissel A, et al. Evaluation of the novel myocardial perfusion positron-­ emission tomography tracer 18F-BMS-747158-02: comparison to 13 N-ammonia and validation with microspheres in a pig model. Circulation. 2009;119(17):2333–42. 32. Sherif HM, Nekolla SG, Saraste A, Reder S, Yu M, Robinson S, et  al. Simplified quantification of myocardial flow reserve with flurpiridaz F-18: validation with microspheres in a pig model. J Nucl Med. 2011;52(4):617–24. 33. Klein R, deKemp RA. Cardiac PET imaging: principles and new developments. In: Khalil M, editor. Basic science of PET imaging. Cham: Springer; 2017. p. 451–83. 34. Canepa M, Bezante G, Vianello P, Ameri P, Milaneschi Y, Aste M, et al. Diagnostic value of ischemia severity at myocardial perfusion imaging in elderly persons with suspected coronary disease. J Cardiovasc Med (Hagerstown). 2016;17(10):719–28. 35. Slomka P, Xu Y, Berman D, Germano G.  Quantitative analysis of perfusion studies: strengths and pitfalls. J Nucl Cardiol. 2012;19(2):338–46. 36. Garcia EV, Faber TL, Cooke CD, Folks RD, Chen J, Santana C. The increasing role of quantification in clinical nuclear cardiology: the Emory approach. J Nucl Cardiol. 2007;14(4):420–32. 37. Germano G, Kavanagh PB, Slomka PJ, Van Kriekinge SD, Pollard G, Berman DS. Quantitation in gated perfusion SPECT imaging: the cedars-Sinai approach. J Nucl Cardiol. 2007;14(4):433–54. 38. Ficaro EP, Lee BC, Kritzman JN, Corbett JR.  Corridor4DM: the Michigan method for quantitative nuclear cardiology. J Nucl Cardiol. 2007;14(4):455–65. 39. Liu YH.  Quantification of nuclear cardiac images: the Yale approach. J Nucl Cardiol. 2007;14(4):483–91. 40. Rossen JD, Quillen JE, Lopez AG, Stenberg RG, Talman CL, Winniford MD.  Comparison of coronary vasodilation with intravenous dipyridamole and adenosine. J Am Coll Cardiol. 1991;18(2):485–91. 41. Mekkaoui C, Jadbabaie F, Dione DP, Meoli DF, Purushothaman K, Belardinelli L, et al. Effects of adenosine and a selective A2A adenosine receptor agonist on hemodynamic and thallium-201 and technetium-99m-sestaMIBI biodistribution and kinetics. JACC Cardiovasc Imaging. 2009;2(10):1198–208. 42. Bauckneht M, Piva R, Ferrarazzo G, Ceriani V, Buschiazzo A, Morbelli S, et  al. Non-invasive measurement of coronary flow reserve: uniqueness of radionuclide methods and alternative techniques. Q J Nucl Med Mol Imaging. 2016;60(4):324–37. 43. Sampson UK, Dorbala S, Limaye A, Kwong R, Di Carli MF.  Diagnostic accuracy of rubidium-82 myocardial perfusion imaging with hybrid positron emission tomography/computed tomography in the detection of coronary artery disease. J Am Coll Cardiol. 2007;49(10):1052–8. 44. Dorbala S, Vangala D, Sampson U, Limaye A, Kwong R, Di Carli MF. Value of vasodilator left ventricular ejection fraction reserve in evaluating the magnitude of myocardium at risk and the extent of angiographic coronary artery disease: a 82Rb PET/CT study. J Nucl Med. 2007;48(3):349–58. 45. Kajander S, Joutsiniemi E, Saraste M, Pietila M, Ukkonen H, Saraste A, et al. Cardiac positron emission tomography/computed tomography imaging accurately detects anatomically and functionally significant coronary artery disease. Circulation. 2010;122(6):603–13.

20  Radionuclide Imaging of Cardiovascular Disease 46. Johnson NP, Gould KL.  Physiological basis for angina and ST-segment change PET-verified thresholds of quantitative stress myocardial perfusion and coronary flow reserve. JACC Cardiovasc Imaging. 2011;4(9):990–8. 47. Packard RR, Huang SC, Dahlbom M, Czernin J, Maddahi J. Absolute quantitation of myocardial blood flow in human subjects with or without myocardial ischemia using dynamic flurpiridaz F-18 PET. J Nucl Med. 2014;55(9):1438–44. 48. Ko SM, Hwang HK, Kim SM, Cho IH. Multi-modality imaging for the assessment of myocardial perfusion with emphasis on stress perfusion CT and MR imaging. Int J Cardiovasc Imaging. 2015;31(Suppl 1):1–21. 49. Alexanderson-Rosas E, Guinto-Nishimura GY, Cruz-Mendoza JR, Oropeza-Aguilar M, De La Fuente-Mancera JC, Barrero-­ Mier AF, et al. Current and future trends in multimodality imaging of coronary artery disease. Expert Rev Cardiovasc Ther. 2015;13(6):715–31. 50. Osler W. The Gulstonian lectures, on malignant endocarditis. Br Med J. 1885;1(1264):577–9. 51. Thuny F, Giorgi R, Habachi R, Ansaldi S, Le Dolley Y, Casalta JP, et al. Excess mortality and morbidity in patients surviving infective endocarditis. Am Heart J. 2012;164(1):94–101. 52. Gelijns AC, Moskowitz AJ, Acker MA, et al. Management practices and major infections after cardiac surgery. J Am Coll Cardiol. 2014;64:372–81. 53. Donlan RM.  Biofilms and device-associated infections. Emerg Infect Dis. 2001;7:277–81. 54. von Eiff C, Jansen B, Kohnen W, Becker K. Infections associated with medical devices. Drugs. 2005;65:179–214. 55. Baddour LM, Wilson WR, Bayer AS, Fowler VG Jr, Tleyjeh IM, Rybak MJ, et al. Infective endocarditis in adults: diagnosis, antimicrobial therapy, and management of complications: a scientific statement for healthcare professionals from the American Heart Association. Circulation. 2015;132(15):1435–86. 56. Jensen AG, Wachmann CH, Poulsen KB, Espersen F, Scheibel J, Skinhøj P, et al. Risk factors for hospital-acquired Staphylococcus aureus bacteremia. Arch Intern Med. 1999;159(13):1437–44. 57. Jensen AG, Wachmann CH, Espersen F, et al. Treatment and outcome of Staphylococcus aureus bacteremia: a prospective study of 278 cases. Arch Intern Med. 2002;162:25–32. 58. Perez de Isla L, Zamorano J, Lennie V, et al. Negative blood culture infective endocarditis in the elderly: long-term follow-up. Gerontology. 2007;53:245–9. 59. Li JS, Sexton DJ, Mick N, Nettles R, Fowler VG Jr, Ryan T, et al. Proposed modifications to the Duke criteria for the diagnosis of infective endocarditis. Clin Infect Dis. 2000;30(4):633–8. 60. Habib G, Lancellotti P, Antunes MJ, Bongiorni MG, Casalta JP, Del Zotti F, et  al. 2015 ESC guidelines for the management of infective endocarditis: The Task Force for the Management of Infective Endocarditis of the European Society of Cardiology (ESC). Endorsed by: European Association for Cardio-Thoracic Surgery (EACTS), the European Association of Nuclear Medicine (EANM). Eur Heart J. 2015;36(44):3075–128. 61. Neelankavil JP, Thompson A, Mahajan A. Managing cardiovascular implantable devices during perioperative care. The Anesthesia Patient Safety Foundation [Internet]. 2013. Available from: https:// www.apsf.org/newsletters/html/2013/fall/01_cieds.htm. 62. Podoleanu C, Deharo JC.  Management of cardiac implantable electronic device infection. Arrhythm Electrophysiol Rev. 2014;3:184–9. 63. Voigt A, Shalaby A, Saba S. Rising rates of cardiac rhythm management device infections in the United States: 1996 through 2003. J Am Coll Cardiol. 2006;48:590–1. 64. Margey R, McCann H, Blake G, et al. Contemporary management of and outcomes from cardiac device related infections. Europace. 2010;12:64–70.

495 65. Camus C, Leport C, Raffi F, et  al. Sustained bacteremia in 26 patients with a permanent endocardial pacemaker: assessment of wire removal. Clin Infect Dis. 1993;17:46–55. 66. Tarakji KG, Chan EJ, Cantillon DJ, et al. Cardiovascular implantable electronic device infections: presentation, management, and patient outcomes. Heart Rhythm. 2010;7:1043–7. 67. Holman WL, Naftel DC, Eckert CE, et  al. Durability of left ventricular assist devices: interagency registry for mechanically assisted circulatory support (INTERMACS). J Thorac Cardiovasc Surg. 2006–2011;2013(146):437–41. 68. Xie A, Phan K, Yan TD. Durability of continuous-flow left ventricular assist devices: a systematic review. Ann Cardiothorac Surg. 2014;3:547–56. 69. Wickline SA, Fischer KC. Can infections be imaged in implanted devices? ASAIO J. 2000;46:80–1. 70. Siméon S, Flécher E, Revest M, et  al. Left ventricular assist device-related infections: a multicentric study. Clin Microbiol Infect. 2017;23:748–51. 71. Koval CE, Rakita R.  AST Infectious Disease Community of Practice Ventricular assist device related infections and solid organ transplantation. Am J Transplant. 2013;13:348–54. 72. Erba PA, Conti U, Lazzeri E, et al. Added value of 99mTc-HMPAO-­ labeled leukocyte SPECT/CT in the characterization and management of patients with infectious endocarditis. J Nucl Med. 2012;53:1235–43. 73. Erba PA, Lancellotti P, Vilacosta I, et  al. Recommendation on nuclear and multi-modality imaging in IE and CIED infections. Eur J Nucl Med Mol Imaging. 2018;45(10):1795–815. 74. Rouzet F, Chequer R, Benali K, et  al. Respective performance of 18F-FDG PET and radiolabeled leucocyte scintigraphy for the diagnosis of prosthetic valve endocarditis. J Nucl Med. 2014;55:1980–5. 75. Scholtens AM, Verberne HJ, Budde RP, Lam MG.  Additional heparin preadministration improves cardiac glucose metabolism suppression over low-carbohydrate diet alone in 18F-FDG PET imaging. J Nucl Med. 2016;57(4):568–73. 76. Dilsizian V. Transition from SPECT to PET myocardial perfusion imaging: a desirable change in nuclear cardiology to approach perfection. J Nucl Cardiol. 2016;23(3):337–8. 77. Jamar F, Buscombe J, Chiti A, Christian PE, Delbeke D, Donohoe KJ, et al. EANM/SNMMI guideline for 18F-FDG use in inflammation and infection. J Nucl Med. 2013;54:647–58. 78. Jimenez-Ballve A, Perez-Castejon MJ, Delgado-Bolton RC, Sanchez-Enrique C, Vilacosta I, Vivas D, et al. Assessment of the diagnostic accuracy of 18F-FDG PET/CT in prosthetic infective endocarditis and cardiac implantable electronic device infection: comparison of different interpretation criteria. Eur J Nucl Med Mol Imaging. 2016;43(13):2401–12. 79. Hyafil F, Rouzet F, Lepage L, et al. Role of radiolabelled leucocyte scintigraphy in patients with a suspicion of prosthetic valve endocarditis and inconclusive echocardiography. Eur Heart J Cardiovasc Imaging. 2013;14:586–94. 80. Gomes A, Glaudemans AW, Touw DJ, et al. Diagnostic value of imaging in infective endocarditis: a systematic review. Lancet Infect Dis. 2017;17:e1–e14. 81. Saby L, Laas O, Habib G, et  al. Positron emission tomography/computed tomography for diagnosis of prosthetic valve 18 endocarditis: increased valvular F-fluorodeoxyglucose uptake as a novel major criterion. J Am Coll Cardiol. 2013;61: 2374–82. 82. Pizzi MN, Roque A, Fernández-Hidalgo N, et  al. Improving the diagnosis of infective endocarditis in prosthetic valves and Intracardiac devices with 18F-Fluordeoxyglucose positron emission tomography/ computed tomography angiography: initial results at an infective endocarditis referral center. Circulation. 2015;132:1113–26.

496 83. Vilacosta I, Graupner C, San Roman JA, et al. Risk of embolization after institution of antibiotic therapy for infective endocarditis. J Am Coll Cardiol. 2002;39:1489–95. 84. Dell’Aquila AM, Mastrobuoni S, Alles S, et al. Contributory role of fluorine 18-Fluorodeoxyglucose positron emission tomography/ computed tomography in the diagnosis and clinical Management of Infections in patients supported with a continuous-flow left ventricular assist device. Ann Thorac Surg. 2016;101:87–94. 85. Avramovic N, Dell’Aquila AM, Weckesser M, et  al. Metabolic volume performs better than SUVmax in the detection of left ventricular assist device driveline infection. Eur J Nucl Med Mol Imaging. 2017;44:1870–7. 86. Di Salvo G, Habib G, Pergola V, et  al. Echocardiography predicts embolic events in infective endocarditis. J Am Coll Cardiol. 2001;37:1069–107. 87. Juneau D, Golfam M, Hazra S, Zuckier LS, Garas S, Redpath C, Bernick J, et al. Positron emission tomography and single-photon emission computed tomography imaging in the diagnosis of cardiac implantable electronic device infection: a systematic review and meta-analysis. Circ Cardiovasc Imaging. 2017;10(4). pii: e005772. https://doi.org/10.1161/CIRCIMAGING.116.005772. 88. Vos FJ, Bleeker-Rovers CP, Sturm PD, et  al. 18F-FDG PET/CT for detection of metastatic infection in gram-positive bacteremia. J Nucl Med. 2010;51:1234–40. 89. Orvin K, Goldberg E, Bernstine H, et al. The role of FDG-PET/ CT imaging in early detection of extra-cardiac complications of infective endocarditis. Clin Microbiol Infect. 2015;21:69–76. 90. Kestler M, Munoz P, Rodriguez-Creixems M, et al. Role of 18F-­ FDG PET in patients with infectious endocarditis. J Nucl Med. 2014;55:1093–8. 91. Ozcan C, Asmar A, Ggill S, et  al. The value of FDG-PET/CT in the diagnostic work-up of extracardiac infectious manifestations in infectious endocarditis. Int J Cardiovasc Imaging. 2013;29:1629–37. 92. Delahaye F, M’Hammedi A, Guerpillon B, et al. Systematic search for present and potential portals of entry for infective endocarditis. J Am Coll Cardiol. 2016;67:151–8. 93. McCoy CW, Mason JM.  Enterococcal endocarditis associated with carcinoma of the sigmoid; report of a case. J Med Assoc State Ala. 1951;21:162–6. 94. Berger P, Vaartjes I, Moll FL, et al. Cumulative incidence of graft infection after primary prosthetic aortic reconstruction in the endovascular era. Eur J Vasc Endovasc Surg. 2015;49:581–5. 95. Hicks RC, Greenhalgh RM.  The pathogenesis of vascular graft infection. Eur J Vasc Endovasc Surg. 1997;14:5–9. 96. Kilic A, Arnaoutakis DJ, Reifsnyder T, Black JH III, et  al. Management of infected vascular grafts. Vasc Med. 2016;21:53–60. 97. Low RN, Wall SD, Jeffrey RB Jr, et al. Aortoenteric fistula and perigraft infection: evaluation with CT. Radiology. 1999;175:157–62. 98. Orton DF, LeVeen RF, Saigh JA, et  al. Aortic prosthetic graft infections: radiologic manifestations and implications for management. Radiographics. 2000;20:977–93. 99. Liberatore M, Misuraca M, Calandri E, et  al. White blood cell scintigraphy in the diagnosis of infection of endovascular prostheses within the first month after implantation. Med Sci Monit. 2006;12:5–9. 100. Erba PA, Leo G, Sollini M, et al. Radiolabelled leucocyte scintigraphy versus conventional radiological imaging for the management of late, low-grade vascular prosthesis infections. Eur J Nucl Med Mol Imaging. 2014;41:357–68. 101. Spacek M, Belohlavek O, Votrubova J, Sebesta P, Stadler P. Diagnostics of “non-acute” vascular prosthesis infection using 18 F-FDG PET/CT: our experience with 96 prostheses. Eur J Nucl Med Mol Imaging. 2009;36:850–8. 102. Saleem BR, Beukinga RJ, Boellaard R, et al. Textural features of 18 F-fluorodeoxyglucose positron emission tomography scanning

M. Bauckneht et al. in diagnosing aortic prosthetic graft infection. Eur J Nucl Med Mol Imaging. 2017;44:886–94. 103. Tossios P, Karatzopoulos A, Tsagakis K, et al. Successful surgical in situ treatment of prosthetic graft infection by staged procedure after Bentall operation and total aortic arch replacement. Springerplus. 2014;3:172. https://doi.org/10.1186/2193-1801-3-172. 104. Legout L, D’Elia PV, Sarraz-Bournet, et al. Diagnosis and management of prosthetic vascular graft infections. Méd Mal Infect. 2012;42:102–9. 105. Hargrove WC III, Edmunds LH Jr. Management of infected thoracic aortic prosthetic grafts. Ann Thorac Surg. 1984;37:72–7. 106. Takano T, Terasaki T, Wada Y, et al. Treatment of prosthetic graft infection after thoracic aorta replacement. Ann Thorac Cardiovasc Surg. 2014;20:304–9. 107. Aretz HT, Billingham ME, Edwards WD, et  al. Myocarditis: a histopathologic definition and classification. Am J Cardiovasc Pathol. 1987;1:3–14. 108. Lieberman EB, Hutchins GM, Herskowitz A, et  al. Clinico-­ pathologic description of myocarditis. J Am Coll Cardiol. 1991;18:1617–26. 109. Magnani JW, Dec GW.  Myocarditis: current trends in diagnosis and treatment. Circulation. 2006;113:876–90. 110. Laissy JP, Hyafil F, Feldman LJ, et  al. Differentiating acute myocardial infarction from myocarditis: diagnostic value of early- and delayed-perfusion cardiac MR imaging. Radiology. 2005;237:75–82. 111. Takano H, Nakagawa K, Ishio N, et  al. Active myocarditis in a patient with chronic active Epstein–Barr virus infection. Int J Cardiol. 2008;130:e11–3. 112. Oikawa M, Kagaya Y, Otani H, et al. Increased [18F]fluorodeoxyglucose accumulation in right ventricular free wall in patients with pulmonary hypertension and the effect of epoprostenol. J Am Coll Cardiol. 2005;45:1849–55. 113. O’Leary SM, Williams PL, Williams MP, et al. Imaging the pericardium: appearances on ECG-gated 64-detector row cardiac computed tomography. Br J Radiol. 2010;83:194–205. 114. Wang ZJ, Reddy GP, Gotway MB, et al. CT and MR imaging of pericardial disease. Radiographics. 2003;23(Spec No):S167–80. 115. Alter P, Figiel JH, Rupp TP, et al. MR, CT, and PET imaging in pericardial disease. Heart Fail Rev. 2013;18:289–306. 116. Dong A, Dong H, Wang Y, et al. 18F-FDG PET/CT in differentiating acute tuberculous from idiopathic pericarditis: preliminary study. Clin Nucl Med. 2013;38:e160–5. 117. Couturier B, Huyge V, Soyfoo MS.  Pericarditis revealing large vessel vasculitis. ISRN Rheumatol. 2011;2011:648703. https:// doi.org/10.5402/2011/648703. 118. Birnie DH, Sauer WH, Bogum F, Cooper JM, Culver DA, Duvernoy CS, et al. HRS expert consensus statement on the diagnosis and management of arrhythmias associated with cardiac sarcoidosis. Heart Rhythm. 2014;11:1304–23. 119. Cooper LT, Baughman KL, Feldman AM, Frustaci A, Jessup M, Kuhl U, et  al. The role of endomyocardial biopsy in the management of cardiovascular disease: a scientific statement from the American Heart Association, the American College of Cardiology, and the European Society of Cardiology Endorsed by the Heart Failure Society of America and the Heart Failure Association of the European Society of Cardiology. Eur Heart J. 2007;28(24):3076–93. 120. Simonen P, Lehtonen J, Kandolin R, Schildt J, Marjasuo S, Miettinen H, et al. F-18-fluorodeoxyglucose positron emission tomographyguided sampling of mediastinal lymph nodes in the diagnosis of cardiac sarcoidosis. Am J Cardiol. 2015;116(10):1581–5. 121. Blankstein R, Osborne M, Naya M, Waller A, Kim CK, Murthy VL, et al. Cardiac positron emission tomography enhances prognostic assessments of patients with suspected cardiac sarcoidosis. J Am Coll Cardiol. 2014;63(4):329–36.

20  Radionuclide Imaging of Cardiovascular Disease 122. Tang R, Wang JT, Wang L, Le K, Huang Y, Hickey AJ, Emmett L.  Impact of patient preparation on the diagnostic performance of 18F-FDG PET in cardiac sarcoidosis: a systematic review and meta-analysis. Clin Nucl Med. 2016;41:e327–39. 123. Youssef G, Leung E, Mylonas I, Nery P, Williams K, Wisenberg G, et  al. The use of 18F-FDG PET in the diagnosis of cardiac sarcoidosis: a systematic review and metaanalysis including the Ontario experience. J Nucl Med. 2012;53(2):241–8. 124. Schneider S, Batrice A, Rischpler C, Eiber M, Ibrahim T, Nekolla SG. Utility of multimodal cardiac imaging with PET/MRI in cardiac sarcoidosis: implications for diagnosis, monitoring and treatment. Eur Heart J. 2014;35(5):312. 125. Ruberg FL, Appelbaum E, Davidoff R, Ozonoff A, Kissinger KV, Harrigan C, et al. Diagnostic and prognostic utility of cardiovascular magnetic resonance imaging in light-chain cardiac amyloidosis. Am J Cardiol. 2009;103(4):544–9. 126. Perugini E, Guidalotti PL, Salvi F, Cooke RM, Pettinato C, Riva L, et  al. Noninvasive etiologic diagnosis of cardiac amyloidosis using 99mTc-3,3-diphosphono-1,2-propanodicarboxylic acid scintigraphy. J Am Coll Cardiol. 2005;46(6):1076–84. 127. Trivieri MG, Dweck MR, Abgral R, Robson PM, Karakatsanis NA, Lala A, et al. 18F-sodium fluoride PET/MR for the assessment of cardiac amyloidosis. J Am Coll Cardiol. 2016;68(24):2712–4. 128. Antoni G, Lubberink M, Estrada S, Axelsson J, Carlson K, Lindsjo L, et al. In vivo visualization of amyloid deposits in the heart with 11 C-PIB and PET. J Nucl Med. 2013;54(2):213–20. 129. Dorbala S, Vangala D, Semer J, Strader C, Bruyere JR Jr, Di Carli MF, et al. Imaging cardiac amyloidosis: a pilot study using 18 F-florbetapir positron emission tomography. Eur J Nucl Med Mol Imaging. 2014;41(9):1652–62. 130. Law WP, Wang WY, Moore PT, Mollee PN, Ng AC. Cardiac amyloid imaging with 18F-Florbetaben PET: a pilot study. J Nucl Med. 2016;57(11):1733–9. 131. Capitanio S, Marini C, Bauckneht M, Sambuceti G.  Nuclear cardiology in heart failure. Curr Cardiovasc Imaging Rep. 2014;7(3):9256. 132. Schinkel AF, Bax JJ, Poldermans D, Elhendy A, Ferrari R, Rahimtoola SH. Hibernating myocardium: diagnosis and patient outcomes. Curr Probl Cardiol. 2007;32(7):375–410. 133. Bauckneht M, Sambuceti G, Pomposelli E, Fiz F, Marini C. Pathophysiological basis of myocardial innervation imaging in heart failure. Clin Transl Imaging. 2015;3(5):347–55.

497 134. Verberne HJ, Feenstra C, de Jong WM, Somsen GA, van EckSmit BL, Busemann Sokole E.  Influence of collimator choice and simulated clinical conditions on 123I-MIBG heart/mediastinum ratios: a phantom study. Eur J Nucl Med Mol Imaging. 2005;32(9):1100–7. 135. Verschure DO, de Wit TC, Bongers V, Hagen PJ, Sonneck Koenne C, D’Aron J, et  al. 123I-MIBG heart-to-mediastinum ratio is influenced by high-energy photon penetration of collimator septa from liver and lung activity. Nucl Med Commun. 2015;36(3):279–85. 136. Veltman CE, Boogers MJ, Meinardi JE, Al Younis I, Dibbets-­ Schneider P, Van der Wall EE, et  al. Reproducibility of planar 123 I-meta-iodobenzylguanidine (MIBG) myocardial scintigraphy in patients with heart failure. Eur J Nucl Med Mol Imaging. 2012;39(10):1599–608. 137. Klene C, Jungen C, Okuda K, Kobayashi Y, Helberg A, Mester J, et al. Influence of ROI definition on the heart-to-mediastinum ratio in planar 123I-MIBG imaging. J Nucl Cardiol. 2018;25(1):208–16. 138. Okuda K, Nakajima K, Hosoya T, Ishikawa T, Konishi T, Matsubara K, et  al. Semi-automated algorithm for calculating heart-to-mediastinum ratio in cardiac Iodine-123 MIBG imaging. J Nucl Cardiol. 2011;18(1):82–9. 139. Jacobson AF, Senior R, Cerqueira MD, Wong ND, Thomas GS, Lopez VA, et  al. Myocardial iodine-123 meta-­ iodobenzylguanidine imaging and cardiac events in heart failure. Results of the prospective ADMIRE-HF (AdreView Myocardial Imaging for Risk Evaluation in Heart Failure) study. J Am Coll Cardiol. 2010;55(20):2212–21. 140. Puppo C, Massollo M, Paparo F, Camellino D, Piccardo A, Shoushtari Zadeh Naseri M, et al. Giant cell arteritis: a systematic review of the qualitative and semiquantitative methods to assess vasculitis with 18F-fluorodeoxyglucose positron emission tomography. Biomed Res Int. 2014;2014:574248. 141. Hooisma GA, Balink H, Houtman PM, Slart RH, Lensen KD. Parameters related to a positive test result for FDG PET(/CT) for large vessel vasculitis: a multicenter retrospective study. Clin Rheumatol. 2012;31(5):861–71. 142. Moosig F, Czech N, Mehl C, Henze E, Zeuner RA, Kneba M, et  al. Correlation between 18-fluorodeoxyglucose accumulation in large vessels and serological markers of inflammation in polymyalgia rheumatica: a quantitative PET study. Ann Rheum Dis. 2004;63(7):870–3.

Radionuclide Imaging of Benign Pulmonary Diseases

21

Federica Guidoccio, Edoardo Airò, and Giuliano Mariani

Contents 21.1   General Background on the Role of Radionuclide Imaging in Benign Pulmonary Diseases

 500

21.2   Anatomy of the Lungs

 500

21.3   Radiopharmaceuticals Used in Benign Pulmonary Diseases

 502

21.4   Lung Perfusion Imaging

 502

21.5   Lung Ventilation Scintigraphy with Radioaerosol

 503

21.6   Lung Ventilation Scintigraphy with Inert Radioactive Gases 21.6.1  Ventilation Imaging with 133Xe 21.6.2  Ventilation Imaging with 81mKr

 503  503  503

21.7   Sequence of Imaging

 506

21.8   SPECT/Low-Dose CT (SPECT/CT)

 506

21.9    Ga-Citrate Scintigraphy

 506

21.10   Clinical Indications

 507

21.11   Acute Pulmonary Embolism

 507

21.12   Follow-Up of Patients After Acute Pulmonary Embolism

 514

21.13   Obstructive Lung Disease

 515

21.14   Quantification of Differential Pulmonary Function Before Pulmonary Surgery

 516

21.15   Interstitial Lung Disease

 516

References

 519

67

Learning Objectives

• Understand the basic notions on pathophysiology and clinical presentation of the most common benign pulmonary diseases that can be managed with the help of radionuclide imaging. F. Guidoccio (*) · G. Mariani Regional Center of Nuclear Medicine, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy E. Airò Fondazione CNR/Regione Toscana Gabriele Monastero, Pisa, Italy

• Acquire information on anatomy and physiology of the lungs as they relate to the distribution of diagnostic radiopharmaceuticals that can be utilized for imaging in patients with benign pulmonary diseases. • Learn the protocols utilized for lung perfusion scintigraphy and lung ventilation scintigraphy. • Learn the protocols utilized for 67Ga-citrate scintigraphy. • Understand the clinical indications for employing radionuclide imaging in patients with benign pulmonary diseases, including acute pulmonary embolism, obstructive lung disease, and interstitial lung disease.

© Springer Nature Switzerland AG 2019 D. Volterrani et al. (eds.), Nuclear Medicine Textbook, https://doi.org/10.1007/978-3-319-95564-3_21

499

500

21.1 G  eneral Background on the Role of Radionuclide Imaging in Benign Pulmonary Diseases

F. Guidoccio et al.

death in 13% of autopsies [5]. In the absence of risk factors, PE is rare in children under 15 years of age (5 mm or as an oesophageal mass that causes luminal obstruction (and proximal dilation), with soft tissue and fat stranding that generate a feathery appearance, showing a weak and diffuse [18F]FDG uptake related to the perioesophageal fat invasion. The presence of tracheobronchial, aortic, and bone invasion is important because it determines the suitability of resection (T4a tumours are resectable, whereas T4b are not). Tracheobronchial invasion can be suspected when there is a discrete indentation on the posterior tracheal wall or displacement of the trachea/bronchus by the tumour. Aortic invasion is suggested when the contact area between the tumour and aorta is >90% or if there is obliteration of the triangular fat space between the oesophagus, aorta, and spine adjacent to the primary tumour. Detection of pathologic lymph nodes by CT depends primarily on size criteria (>1 cm in short-axis diameter or 5 mm in supraclavicular location), but normal-sized lymph nodes can be invaded, which can be discovered by an increase of [18F]FDG uptake.

Table 32.3  Oesophageal cancer location classification OC location Cervical

Upper thorax Middle thorax Lower thorax

Position of the upper edge of the tumour Between the cricopharyngeus muscle and the level of the sternal notch; 15–20 cm from the incisors at oesophagoscopy Between the sternal notch and the azygos arch; 20–25 cm from the incisors Between the azygos arch and the level of the inferior pulmonary vein; 25–30 cm from the incisors At lower oesophageal sphincter; 30–40 cm from the incisors

32  Hybrid Imaging for Malignant Conditions of the Gastrointestinal Tract

Distant metastases have been reported at initial presentation in 20–30% of OC, especially in the liver (35%), lungs (20%), bones (9%), and adrenal glands (5%) and rarely in the peritoneum and brain. On CT, liver metastases usually appear as hypoattenuating lesions that are best visualized during the portal venous phase of liver enhancement. Pulmonary metastases are usually of round shape, smooth-­bordered, and noncalcified on CT. [18F]FDG PET/CT has the advantage of allowing total-body coverage, and its primary role is to depict distant metastases, with superior capability as compared to CT, especially in the liver and bone. Moreover, [18F]FDG PET/CT can depict metastases in unexpected locations, including the brain, skeletal muscles (Fig. 32.1), subcutaneous tissues, thyroid gland, and pancreas.

Key Learning Points

• [18F]FDG PET/CT is not sufficiently accurate for T staging as compared to endoscopic US but has a better diagnostic performance to detect distant metastases as compared to CT. • When recurrence is suspected, [18F]FDG PET/CT can help to differentiate it from fibrosis and oedema, but due to high false-positive findings, histologic confirmation is required. • [18F]FDG PET/CT can be useful for radiotherapy planning since it can identify nodal infiltration initially not considered to be included in the radiation field, thus potentially improving treatment results.

32.3 Gastric Cancer GC is the sixth most common tumour worldwide, with an incidence of 7% of all tumour types [13]. It is more common in males (ratio 2:1), with the highest rates in Eastern Asia, Eastern Europe, and South America and lower rates in North America and Western Europe, where a gradual decline has been registered. More recent declines in high-prevalence countries have also been reported and differ from the relative increase in tumours of the gastroesophageal junction. GC is the fourth most common cancer-related cause of death in the world (8.8% of all annual cancer deaths) [13]. Distal or antral GC is associated with risk factors such as H. pylori infection, alcohol consumption, and diets with high salt, large quantities of processed meat, and/or low fruit and vegetable intake. Proximal GC (cardia) is associated with obesity. Tumours of the gastroesophageal junction are associated with reflux and Barrett’s oesophagus and predominate in non-Asian countries. Some GC is familial (~10% of cases). There is an inherited genetic predisposition in 1–3%

845

associated with hereditary nonpolyposis colorectal cancer (CRC), familial adenomatous polyposis CRC, hereditary diffuse GC, gastric adenocarcinoma and proximal polyposis of the stomach, and Peutz-Jeghers syndrome. Adenocarcinomas constitute 90% of GC.  Less common GC such as GI stromal tumours (GISTs), lymphomas, and neuroendocrine tumours are covered elsewhere. GC most frequently presents with progressive upper abdominal discomfort, commonly with weight loss, anorexia, early satiety, nausea, vomiting (particularly when lesions are located in the pylorus (or with widespread disease)), dysphagia (when lesions are located in the cardia), and GI bleeding. If a diagnosis of GC is suspected, it should be confirmed by gastroscopy and biopsy, reporting histology according to the WHO criteria. GC spreads primarily to the local lymph nodes, liver, and peritoneum. Systemic spread is not common; metastases go first to the liver and then to the lungs, adrenal glands, kidneys, bones, and brain. Initial staging and risk assessment should include ceCT of the thorax and abdomen ± pelvis to detect local/distant lymphadenopathy and metastatic disease or ascites. Laparoscopy is recommended for patients with resectable GC.

32.3.1 Indications of [18F]FDG PET/CT [18F]FDG PET/CT has a detection rate of only ~ 55% for the primary tumour and therefore has no role in the diagnosis and T staging of GC. The lesions are often less metabolically active [14–16]. [18F]FDG PET/CT may improve GC staging by detecting involved lymph nodes or metastatic disease as it has a significantly higher accuracy in preoperative staging (68%) than CT (53%) [17, 18]. The National Comprehensive Cancer Network (NCCN) recommends PET/CT for initial staging when there is no evidence of metastases (but it may not be appropriate for T1 tumours) and for post-treatment assessment in unresectable disease or non-surgical candidates following primary treatment if they have renal failure or allergy to intravenous (IV) iodinated contrast medium [18]. [18F]FDG PET/CT is also useful for predicting response to preoperative chemotherapy [19], as well as for the evaluation of recurrence when ceCT is non-diagnostic [20, 21]. However, it is not sensitive for detecting peritoneal involvement, which requires laparoscopy and biopsy [18].

32.3.2 [18F]FDG PET/CT Interpretation and Reporting Focal or diffuse thickening of the gastric wall detected on CT is an important but nonspecific sign of gastric disease. With

846

adequate distension of the stomach with air or contrast agent, wall thickening >5 mm can be considered abnormal. The primary tumour may appear as a focal, nodular, or irregular thickening of the gastric wall or as a polypoid intraluminal mass of soft tissue attenuation. Extension of the tumour into the perigastric fat (very common when wall thickness >2  cm) blurs the serosal surface, and strands and nodules of tumour are seen in the adjacent fat. Malignant lymph nodes are frequently of round shape and have short-axis diameter >6 mm in perigastric region, with central necrosis and heterogeneous or high enhancement. However, the sensitivity of CT for lymph node staging is variable (62.5–91.9%) [22], and global consensus is lacking on specific diagnostic criteria. Nodes near the celiac axis and in the gastrohepatic ligaments and those showing high FDG uptake are most likely to be involved. Local recurrence appears as focal wall thickening at the anastomosis or in the remaining stomach. Nodal recurrence is most common along the course of the hepatic artery or in the para-aortic region. Peritoneal recurrence is seen in the cul-de-sac, on parietal peritoneal surfaces, or on the surface of the bowel.

Key Learning Points

• [18F]FDG PET/CT has no role in the diagnosis and T staging of GC since it is associated with a low detection rate. • [18F]FDG PET/CT is recommended for initial staging when there is no evidence of metastatic disease and for post-treatment assessment in unresectable disease or non-surgical candidates following primary treatment if they have renal insufficiency or allergy to IV contrast. • [18F]FDG PET/CT is not sensitive to detect peritoneal disease.

32.4 Pancreatic Cancer (Adenocarcinoma) The incidence of PC is steadily increasing in most countries. PC is the seventh most common cause of cancer death, with overall 5-year survival of 5.5%. PC presents as a locally advanced disease in 30%, and 50% and may only be treated with palliative chemotherapy. Among the patients who survive surgical resection, the 5-year survival rate remains low (~15–40%) [23]. Surgeons often underestimate the extent of disease and resectability; therefore, improvement of current imaging modalities to better diagnose and stage PC may help avoid futile surgeries.

J. Duch and A. Flotats

PC may present with obstructive jaundice, weight loss, and abdominal or mid-back pain. Imaging has led to increased detection of pancreatic abnormalities, with ~33% of pancreatic incidental findings finally resulting in PC [24]. CeCT is the best-validated and most widely available modality for diagnosis and initial management [25].

32.4.1 Indications for [18F]FDG PET/CT For staging, [18F]FDG/ceCT or PET/MR is helpful to detect both local extent of disease and identify metastasis, particularly in patients with high risk of metastases (borderline resectable disease, markedly elevated carbohydrate antigen 19-9 (CA 19-9), large primary tumours, large regional lymph nodes, or very symptomatic patients). [18F] FDG PET/CT can also be useful to distinguish recurrence from fibrosis after surgical or radiotherapy treatment [26, 27]. The use of [18F]FDG PET/CT following ceCT showed increased sensitivity for the detection of metastatic disease when compared with ceCT and [18F]FDG PET/CT alone in a retrospective study. The sensitivity of detecting metastases for PET/CT alone, ceCT, and the combination of both were 61%, 57%, and 87%, respectively, and additionally, the clinical management of 11% of patients with invasive pancreatic cancer was changed as a result of the [18F]FDG PET/CT [28].

32.4.2 [18F]FDG PET/CT Interpretation and Reporting PC usually appears as a hypoattenuating mass within pancreatic parenchyma on CT.  However, in ~10% of cases, the lesion is isoattenuating, especially small tumours (≤2 cm), thus making diagnosis difficult. Indirect signs of obstruction of pancreatic and common bile duct such as upstream ­pancreatic duct dilation or the double-duct sign (i.e. the presence of simultaneous dilatation of the common bile and pancreatic ducts) are helpful for diagnosis [29].

Key Learning Points

• [18F]FDG PET CT should be considered when there are equivocal findings on ceCT or MR. • [18F]FDG PET/CT has higher sensitivity than CT for the detection of metastases. • [18F]FDG PET/CT can be useful to distinguish recurrence from fibrosis after surgical or radiotherapy treatment. • [18F]FDG PET/CT improves the evaluation of recurrent PC, especially in the setting of elevated CA 19-9 and negative or equivocal CT findings.

32  Hybrid Imaging for Malignant Conditions of the Gastrointestinal Tract

32.5 Colorectal Cancer CRC is the third most common tumour in men and second in women, with an incidence of 10% of all tumour types worldwide [1, 30]. Incidence is higher in males (ratio 1.2:1). Risk factors for CRC include a family history of colon cancer in a first-degree relative, a history of colorectal adenomas or ovarian cancer, personal history of chronic ulcerative colitis or Crohn’s colitis, and diets high in red or processed meat and low in fibre. These patients are often obese and are not physically active. Several genetic factors are associated with CRC including Lynch syndrome and familial adenomatous polyposis [31]. Chronic nonsteroidal anti-inflammatory treatment, especially aspirin, is associated with reduced incidence. CRC is the fifth most common cancer-related cause of death in the world (8.5% of all annual cancer deaths) [13]. The CRC-related 5-year survival rate approaches 60% [32]. Mortality has declined in many Western countries, mainly due to cancer screening programmes, removal of colorectal adenomas, early detection of cancerous lesions, and availability of more effective therapies, mostly for early-stage disease. Left-sided CRC presents most commonly with rectal bleeding, altered bowel habits, and abdominal or back pain. Cecal and ascending CRC usually presents with symptoms of anaemia, occult blood in stool, or weight loss. Complications of CRC include bowel obstruction, volvulus, perforation, and fistulae. Spread of CRC includes direct extension with penetration of the colon wall, lymphatic drainage to regional nodes, haematogenous drainage through portal veins to the liver or through systemic paravertebral venous to the lungs (rectosigmoid CRC), and intraperitoneal seeding. CRC metastasizes in ~50% of patients (at initial diagnosis in ~25% of patients). Metastatic CRC (mCRC) contributes to the high mortality rate of this malignancy [32]. The clinical outcome for mCRC has improved mainly due to earlier detection of metastatic disease and improvement in the efficacy of systemic therapies. Staging of CRC is important in determining therapy and prognosis. Treatment of CRC is determined primarily by TNM staging (Table 32.4). The main procedure for diagnosis of CRC is biopsy during endoscopy by either sigmoidoscopy (as >35% of tumours are located in the rectosigmoid) or, preferably, total colonoscopy. Abdominopelvic ceCT is indicated for subsequent staging. MR is best for the detection and characterization of small lesions in liver and peritoneum.

32.5.1 Indications of [18F]FDG PET/CT Routine use of [18F]FDG PET/CT is not recommended for initial staging, since it does not modify the treatment approach in most patients [33]. However, it can be performed

847

Table 32.4  Colorectal cancer TNM definitions (AJCC 2017) Primary tumour (T) TX Primary tumour cannot be assessed T0 No evidence of primary tumour Tis Carcinoma in situ, intramucosal carcinoma (involvement of the lamina propria with no extension through the muscularis mucosae) T1 Tumour invades the submucosa (through the muscularis mucosae but not into the muscularis propria) T2 Tumour invades the muscularis propria T3 Tumour invades through the muscularis propria into pericolorectal tissues T4 Tumour invades the visceral peritoneum or invades or adheres to adjacent organ or structure  T4a Tumour invades through the visceral peritoneum (including gross perforation of the bowel through tumour and continuous invasion of tumour through serial of inflammation to the surface of the visceral peritoneum)  T4b Tumour directly invades or adheres to adjacent organs or structures Regional lymph nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Infiltration of 1–3 regional lymph nodes (tumour in lymph node measuring ≥0.2 mm) or any number of tumour deposits are present, and all identifiable lymph nodes are negative  N1a Infiltration of 1 regional lymph node  N1b Infiltration of 2–3 regional lymph nodes  N1c No regional nodal infiltration, but there are tumour deposits in the subserosa, mesentery, or non-­ peritonealized pericolic or perirectal/mesorectal tissue metastasis N2 Infiltration of ≥4 regional lymph nodes  N2a Infiltration of 4–6 regional lymph nodes  N2b Infiltration of ≥7 regional lymph nodes Distant metastasis (M) M0 No distant metastasis by imaging, etc.; no evidence of tumour in distant sites or organs (this category is not assigned by pathologists) M1 Metastasis to ≥1 distant site or organ or peritoneum  M1a Metastasis to 1 site or organ without peritoneal metastasis  M1b Metastasis to ≥2 sites or organs without peritoneal metastasis  M1c Metastasis to the peritoneal surface, alone or with other site or organ metastases

when other imaging modalities are equivocal regarding the presence of distant metastases. In case of elevated tumour markers (carcinoembryonic antigen, CEA) without knowledge of the location of relapse or in the presence of equivocal or suspicious lesions on ceCT, [18F] FDG PET/CT is indicated for the detection of local recurrence at the site of the initial colorectal surgery and especially for the detection of metastatic disease (Figs. 32.2 and 32.3) [34–36]. A stepwise imaging approach is recommended to evaluate patients with suspected CRC, starting with a contrast-enhanced CT of the chest, abdomen, and pelvis. If the findings on CT are not diagnostic, an additional evaluation with MR is helpful to

848

J. Duch and A. Flotats

Fig. 32.2  Maximum intensity projection [18F]FDG PET image (on the left), with corresponding axial slices of CT, PET, and fused PET/CT localizing recurrence of colorectal cancer as small but hypermetabolic liver metastases

a

b

c

d

Fig. 32.3  Maximum intensity projection [18F]FDG PET image (a), with corresponding axial slices of CT (b), PET (c), and fused PET/CT (d) localizing recurrence of colorectal cancer as a solitary right pulmonary metastatic nodule with increased [18F]FDG uptake

32  Hybrid Imaging for Malignant Conditions of the Gastrointestinal Tract Fig. 32.4  Maximum intensity projection [18F]FDG PET image (a), with corresponding axial slices of CT (b), PET (c), and fused PET/CT (d) showing an incidental hypermetabolic right colon polyp, which resulted in malignancy after colonoscopy and biopsy

a

849

b

c

d

evaluate the liver, peritoneum, and pelvis, while PET/CT is helpful to detect extrahepatic disease (Fig. 32.3). The role of [18F]FDG PET/CT for the assessment of response after chemoradiation is under investigation, but not superior to CT except for rectal cancer. The extent of subsequent rectal surgery should not be modified yet based on this [37].

32.5.2 [18F]FDG PET/CT Interpretation and Reporting CRC may be a colon polyp seen incidentally as hypermetabolic well-defined oval or round intraluminal projection. Therefore, incidental focal colonic [18F]FDG uptake may require colonoscopy to rule out (pre)malignancy (Fig. 32.4). CRC can also appear as a larger hypermetabolic intraluminal mass with nodular contours and irregular mucosal surfaces that may narrow the lumen of the colon. Central mass

hypometabolism and low attenuation due to haemorrhage or necrosis can be seen. Air within the tumour indicates ulceration. Loss of fat planes between the tumour and adjacent structures suggests local invasion. CRC recurrences are uncommon, occurring in 6.4% of patients within 5 years of surgery. The risk of developing recurrence is highest within the first year after surgery [38]. Recurrences occur at the site of the original tumour in ~50% of the cases (location of the original tumour may be identified by high-attenuation metallic bowel clips in the anastomosis), whereas the remainder recur at distant sites, especially in the liver (Fig. 32.2). Presacral soft tissue densities seen in patients with abdominoperineal resection may represent a recurrence or fibrosis. While recurrent tumour shows increased [18F]FDG uptake and tends to be nodular and convex anteriorly, fibrosis tends to be more uniform and concave anteriorly, without hypermetabolism (unless inflamed).

850

Key Learning Points

• PET/CT is not routinely indicated for initial staging but can be performed in addition to MR in the case of resectable metastatic disease (initial workup or recurrence) to exclude the presence of other metastatic sites. • Consider [18F]FDG PET/CT in front suspicion of recurrence in the case of serial CEA elevation, after normal colonoscopy and CT for colon cancer, and as first-line imaging in the case of rectal cancer. • Colonoscopy is recommended in the case of incidental focal colonic [18F]FDG uptake.

32.6 Anal Cancer Epidermoid anal cancer or squamous cell carcinoma of the anus (SCCA) is an infrequent tumour (2–4% of CRC and anal tumours). The incidence of SCCA is higher in females, and it is increasing due to the strong association with human papillomavirus (HPV) infection, which is the contributing agent in 80–85% of patients [39]. Factors increasing the risk of HPV infection (anal intercourse and a high lifetime number of sexual partners) and/or modulating host response and the persistence of HPV infection (human immunodeficiency virus (HIV), immune suppression, a history of other HPV-­ related cancers, autoimmune disorders, social deprivation, and smoking) have also influence on the epidemiology of SCCA. SCCA prognosis has remained stable in the last two decades, and 5-year survival rates are ~70%. SCCA commonly presents with bleeding, which may be associated with a mass, non-healing ulcer, pain, itching, discharge, faecal incontinence, and fistulae. The diagnosis is made on biopsy-proven histology. At diagnosis, lymph node involvement is present in 30–40% of patients, whereas systemic spread is uncommon (5–8%). Imaging should include pelvic MR or endo-anal US. [18F]FDG PET/CT has a high sensitivity in identifying involved lymph nodes, causing a change in staging in ~20% of patients, and changes patient management in ~3–5% of cases. Treatment is mostly determined by TNM staging (Table 32.5), usually with locoregional scope owing to the low rate of distant metastases and aimed to achieve cure with locoregional control and preservation of anal function.

J. Duch and A. Flotats Table 32.5  Anal cancer TNM definitions (AJCC 2017) Primary tumour (T) TX Primary tumour cannot be assessed T0 No evidence of primary tumour Tis Carcinoma in situ (Bowen disease, high-grade squamous intraepithelial lesion [HSIL], AIN II–III T1 Tumour ≤2 cm in greatest dimension T2 Tumour >2 cm but ≤5 cm in greatest dimension T3 Tumour >5 cm in greatest dimension T4 Tumour of any size invades adjacent organ(s) (e.g. vagina, urethra, bladder); direct invasion of the rectal wall, perirectal skin, subcutaneous tissue, or the sphincter muscle(s) is not classified as T4 Regional lymph nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Metastasis in perirectal lymph node(s) N2 Metastasis in unilateral internal iliac and/or inguinal lymph node(s) N3 Metastasis in perirectal and inguinal lymph nodes and/or bilateral internal iliac and/or inguinal lymph nodes Distant metastasis (M) M0 No distant metastasis M1 Distant metastasis AIN Anal intraepithelial neoplasia

After completion of chemoradiation therapy, SCCA tends to regress slowly. Clinical rectal and inguinal examinations are the mainstay of determining complete response. Pelvic MR or CT is also necessary. To date, few [18F]FDG PET/CT studies have assessed treatment response, and the timing of assessment is controversial. The benefit of PET/CT is to detect residual subclinical pelvic or extrapelvic/para-aortic node involvement.

32.6.2 [18F]FDG PET/CT Interpretation and Reporting When searching for nodal involvement, one should bear in mind that drainage of anal canal lesions differs from that of anal margin tumours. Proximal lesions drain to perirectal nodes along the inferior mesenteric artery. Immediately above the dentate line of the anal canal, drainage is to internal pudendal nodes and to the internal iliac system. Below the dentate line of the anal canal and perianal skin, drainage is to the inguinal, femoral, and external iliac nodes.

32.6.1 Indications of [18F]FDG PET/CT Key Learning Point

[18F]FDG PET/CT has a high sensitivity for the detection of involved lymph nodes, thus influencing radiation therapy planning, and is recommended in the diagnostic workup of SCCA.

• [18F]FDG PET/CT has a high sensitivity in identifying involved lymph nodes in SCCA and is recommended for treatment planning.

32  Hybrid Imaging for Malignant Conditions of the Gastrointestinal Tract

32.7 Hepatobiliary Tract Cancer (Cholangiocarcinoma)

nation and 5-year overall survival prognosis. Abdomen ceMR cholangiopancreatography and ceCT are the primary modalities for staging.

CC refers to both intrahepatic and extrahepatic tumours of the bile tract, including gallbladder and ampulla of Vater. Although it represents only 3% of all GI malignancies [40], prognosis is generally poor. CC is associated with risk factors such as chronic inflammation, especially primary sclerosing cholangitis, but also with cirrhosis, hepatitis B and C, and fatty liver disease. The incidence of CC increases with age except those cases related to primary sclerosing cholangitis. Symptoms are those derived from bile duct obstruction including jaundice, abdominal pain, cola-coloured urine, and skin pruritus. Symptoms are common in extrahepatic CC, but intrahepatic CC and, particularly, gallbladder cancer are often diagnosed incidentally during a laparoscopic cholecystectomy performed for cholelithiasis. The staging system of CC is separated depending on its location on the bile duct. Each location has a T stage desig-

a

851

32.7.1 Indications of [18F]FDG PET/CT The small tumour size and the growth pattern of CC (small nests of cells embedded in a fibrous stroma) may compromise the accuracy of [18F]FDG PET/CT to detect low volume deposits in regional lymph node metastases, resulting in limited sensitivity but high specificity (Fig. 32.5) [41]. [18F]FDG PET/CT should be considered in front of equivocal findings on CT or MR, especially if a positive finding would change management [42, 43], but its routine use in the preoperative setting has not been yet established in prospective trials. Several studies have shown [18F]FDG PET/CT to be superior to CT for detecting distant metastases [44, 45]. The role of [18F]FDG PET/CT in monitoring response to therapy is controversial with no ideal imaging modality yet

b

c

d

e

f

g

Fig. 32.5  Cholangiocarcinoma with hepatic hilar lymph node infiltration. Maximum intensity projection [18F]FDG PET image (a) and ceCT axial slices of liver arterial phase (b), portal venous phase (c), and equi-

librium phase (d), with the corresponding axial slice of fused PET/CT (e) demonstrate the primary tumour. Additionally, axial slices of CT (f) and fused PET/CT (g) show locoregional lymph node infiltration

852

J. Duch and A. Flotats

available. In addition, more trials are needed to establish the utility of [18F]FDG PET/CT in identifying tumour recurrence [40] and for radiotherapy planning.

32.7.2 [ F]FDG PET/CT Interpretation and Reporting 18

CC is typically homogeneously low in attenuation on unenhanced CT and demonstrates heterogeneous minor peripheral enhancement with gradually central enhancement on ceCT (Fig. 32.5). Duct ectasia with or without a visible mass can be seen in intraductal tumours. It is important to keep in mind that lymph node involvement at the porta does not necessarily contraindicate surgical resection but metastases to more distant nodes such as in the celiac, retro-pancreatic, and para-aortic chains do [46].

Key Learning Points

• [18F]FDG PET/CT has limited sensitivity but high specificity in the detection of regional lymph node metastases in CC. • [18F]FDG PET/CT is an emerging indication in defining recurrent disease after treatment, especially in gallbladder cancer.

32.8 Hepatocellular Cancer HCC is the third most common cause of cancer death [47]. HCC is associated with viral hepatitis and cirrhosis, and strategies to reduce its high mortality rely on screening patients with these risk factors by US followed by CT or MR and liver biopsy. Surgical resection and liver transplant remain the best options for cure, but they are frequently not possible due to tumour size, multifocal disease, extrahepatic involvement, or insufficient hepatic reserve. Clinical presentation may include jaundice, anorexia, weight loss, upper abdominal pain, hepatomegaly, and ascites. HCC diagnosis involves multiphasic ceMR or multiphasic ceCT liver protocol. Different staging systems have been proposed. The Barcelona Clinic Liver Cancer (BCLC) system is the reference staging system in which radiologic tumour extent is only an element of classification. The system identifies those patients with early HCC who may benefit from curative therapies (≤3 lesions of 6 h regular or short-acting insulin   • >4 h rapid-acting insulin In diabetics on continuous insulin infusion, stop the insulin pump ≥4 h prior to [18F]FDG administration Weigh the patient upon arrival in the PET unit on a calibrated weighing scale Check blood glucose (for clinical studies 13 mmol/L (1.5 mg/dL) and/or glomerular filtration 70 Gy (up to 200–300 Gy) can thus be selectively delivered through locoregional administration of agents emitting β− particles (labeled with either 131I, 90Y, 188Re, or 166 Ho), while the non-affected liver (where blood supply derives from the portal vein) receives low radiation levels, with minimal risk of inducing RILD [13–17].

33.3 Radioactive Lipiodol I-Lipiodol (a suspension of lipidic particles) injected into the hepatic artery of patients with HCC follows the preferential blood flow toward the tumor, where the radiolabeled micellae are retained by pinocytosis both in the tumor cells and in the endothelium arterial vessels feeding the tumor [18]. This route of administration is performed under angiographic monitoring and results in >75% of the injected 131I-lipiodol being retained in the liver, the remainder distributing mainly to the lungs. Thyroidal uptake of free radioiodine released after treatment must be prevented with adequate medication. Either a fixed activity of 131I-lipiodol (2.4 GBq, or 65 mCi) can be administered or a patient-specific dosimetry estimate can guide definition of the activity to be injected. Tumor/ non-tumor ratios >5 are generally achieved, with >10% of the injected radioactivity remaining in the tumor with an effective half-life >6 days [14, 15, 19]. Because of such long half-life of 131I-lipiodol in the tumor, legislation regarding radioprotection of the general population requires in some countries hospitalization for about 1 week. 131

33  Radionuclide Therapy for Tumors of the Liver and Biliary Tract

There are usually no serious adverse side effects caused by treatment with 131I-lipiodol, interstitial pneumopathy due to trapping and retention of the radiolabeled particle suspension being reported as the main risk of such therapy [20]. The objective response rates of trans-arterial therapy with 131 I-lipiodol are reported in 40–50% range [21–25]. In HCC patients with portal thrombosis, 71% survival was reported at 3  months versus 10% with the best supportive care; the median overall survivals were 26  weeks and 10  weeks, respectively [24]. The benefits of treatment with 131I-lipiodol have been proved in several respects, including median time to recurrence; overall survival at 3 and 5 years; recurrence rate; 5-, 7-, and 10-year disease-free survival; and overall survival [26, 27]. Compared with TACE, embolization with131I-lipiodol ensued similar therapeutic efficacy associated however with better tolerance [23]. An alternative approach to therapy with 131I-lipiodol is constituted by lipiodol labeled with 188Re using 4-hexadecyl 2,2,9,9-tetramethyl-4,7-diaza-1,10-decanethiol (HDD) as the chelating agent [28] in patients with inoperable large and/or multifocal HCCs. 188Re has a shorter half-life than 131I (16.9 h versus 8  days), emission of β− particles with high energy (Emax = 2.1 MeV), and emission of 155-keV γ-rays suitable for gamma camera imaging (for the purpose of dosimetry estimates). 188Re can be obtained through the 188W/188Re generator, the parent radionuclide having a physical half-life of 69 days, which is particularly suitable for distribution logistics. 188 Re-HDD-lipiodol represents an excellent alternative to 131 I-lipiodol, because of higher tumor-killing efficacy ­combined with lower toxicity [29]. In patients with inoperable HCC, therapy with 188Re-HDD-lipiodol resulted in favorable safety profile and good clinical response, as shown also by objective responses [30].

Key Learning Points

• While blood to normal hepatic parenchyma is supplied for >70% by the portal system, blood to malignant tissue is supplied mostly by the arterial system. • High radiation doses to the tumor lesions can therefore be selectively delivered through intra-arterial administration of radioactive agents emitting β− particles. • At the same time, following this administration route, the non-affected liver (where blood supply derives from the portal vein) receives low radiation levels—therefore with minimal risk of causing radiation-­induced liver disease. • 131I-Lipiodol and 188Re-HDD-lipiodol (suspensions of lipidic particles carrying 131I or 188Re, respectively) have been used in patients with inoperable hepatocellular carcinoma.

861

33.4 C  ommercially Available Radiolabeled Microspheres for TARE/SIRT 33.4.1  90  Y-Microspheres TARE/SIRT with 90Y-microspheres has shown definite clinical benefits in patients with inoperable primary or metastatic liver tumors [31]. The pure β− emitter 90Y decays to stable 90 Zr with a physical half-life of 64.2 h. The average energy of β− emission is 0.936 MeV, with a mean tissue penetration of 2.5 mm and a maximum tissue range of 10 mm. The 90Y-microspheres commercially available are made of glass (TheraSphere®, MDS Nordion, Ottawa, Ontario, Canada) or of resin (SIR-Spheres®, Sirtex Medical, Sydney, Australia), respectively [32]. The TheraSphere® particles are 20–30 μm in diameter and carry 2500 Bq of 90Y per particle (high specific activity); about 1.2 million microspheres are injected intra-arterially for a single treatment, corresponding to a total administered activity of about 3 GBq (81 mCi). The SIR-Spheres® particles are 20–60 μm in diameter and carrying 50 Bq of 90Y per particle (low specific activity); 40–80 million microspheres are injected for a single treatment to achieve a similar total administered activity of 3 GBq [32].

33.4.2  166  Ho-Microspheres This recently approved formulation known as QuiremSpheres® (distributed in Europe by Quirem Medical BV, Deventer, The Netherlands) consists of a poly(llactic acid) acid matrix containing 166Ho (166Ho-PLLA-­ microspheres); this radionuclide emits β− particles with two main energy peaks (1.74 MeV and 1.85 MeV, with intensity 48.7% and 50.0%, respectively) and decays to stable 166Er with a physical half-life of 26.8 h. The maximum range in tissues of the β− particles emitted by 166Ho is 8.7 mm, with an average range of 2.5 mm. A standard administered therapeutic activity of 166 Ho-PLLA-microspheres contains about 30 million microspheres having a mean diameter of 30 μm (with 97% of the particles being included in the 15–60 μm size range). In addition to its predominant decay with emission of β− particles, 166Ho also emits low-energy γ-rays (81 keV), which allow for scintigraphic visualization posttreatment. This feature is very useful for the following three main reasons: • Prior to actual TARE/SIRT therapy, a small scout activity of 166 Ho-PLLA-MS can be administered for predicting distribution of the treatment dose, although the manufacturer recommends use of a pretreatment scout scintigraphy with 99m Tc-MAA—similarly as recommended for the 90 Y-microspheres; thus, a potential advantage over 90 Y-microspheres is not yet fully exploited in the clinical routine.

862

• Quantitative analysis of the data based on the pre-therapy scout scan with 166Ho-microspheres enable to assess/predict the radiation dose delivered both to the tumor lesion(s) and to normal liver parenchyma. • Trivalent holmium is the element with the highest paramagnetic properties among all known elements (106 μB); therefore, it can in principle be visualized also using MR imaging, with potential clinical applications that are currently being explored.

F. Guidoccio et al.

or bilobar. Ascites indicates poor hepatic reserve or peritoneal metastasis, therefore poor prognosis. For final decision on treatment with 90Y-microsphere TARE/SIRT, all functional and anatomic parameters should be carefully considered within a multidisciplinary team involving interventional radiologists, surgical oncologists, medical oncologists, nuclear medicine physicians, radiation oncologists, medical physicists, and radiation safety experts.

33.5.1  Assessment of Arterial Anatomy Key Learning Points

• Three types of radiolabeled microspheres for radioembolization therapy are commercially available: 90 Y-microspheres made of glass, 90Y-microspheres made of resin, and 166Ho-microspheres made of a poly(l-lactic acid) matrix, respectively. • There are small variations in size of the microspheres among the three preparations, all of them allowing efficient embolization of the precapillary vessels. • Each type of preparation has specific features concerning the amount of radioactivity carried by each microsphere and the number of microspheres administered per treatment. • Whereas 90Y is an almost pure β− emitter (there is in fact an extremely small proportion of decay through internal pair production, with 32 events per million decays), 166Ho also emits low-energy γ-rays that allow for scintigraphic visualization with a standard gamma camera. • Holmium has very high paramagnetic properties; the potential clinical applications linked to this feature are being explored.

As an essential requisite for the therapeutic procedure, pretreatment angiography serves to assess the vascular anatomy of the liver, patency of the portal vein, and presence of artero-­ portal shunting and/or shunting to extrahepatic territories (the most important being the liver-to-lung shunt). Blood flow carrying the radiolabeled microspheres outside the liver vasculature is prevented by prophylactic embolization of the responsible vessels identified during angiography, such as the gastroduodenal artery and the right gastric artery [5]. This safe and effective procedure minimizes the risks of hepato-­enteric flow. Mesenteric angiography ensures that blood supply to the lesions has been adequately identified, since incomplete/ inaccurate definition of the pattern of blood supply may lead to incomplete/ineffective targeting of the tumor lesion.

33.5.2  Pretreatment Imaging with 99mTc-MAA

During pretreatment angiography, 99mTc-macroaggregated serum albumin (99mTc-MAA), or alternatively 99mTc-HSA-­ microspheres, is injected into the hepatic artery to confirm homing of the radiolabeled particles in the tumor lesion(s) and to assess for the presence of shunting to extrahepatic circulation, either in the splanchnic bed or in pulmonary 33.5 Patient Selection bed. Scintigraphy of the chest and upper abdomen by either planar or SPECT/CT imaging (the optimal imaging techBefore treatment, functional status of the liver is evaluated nique) is performed for this purpose (Figs. 33.1, 33.2, and by routine blood chemistry workup. Patients with an ECOG 33.3) [33]. The optimal time window for imaging after performance status >2 are not considered ideal candidates 99mTc-­ MAA injection is within 1  h post-administration, for TARE/SIRT. Treatment with 90Y-microspheres is contra- since intrahepatic degradation of MAA is relatively fast indicated by inadequate liver function, with total bilirubin and radioactivity (smaller fragments and/or free 99mTcO4−) >2.0  mg/dL and serum albumin 70 Gy. The risk of liver failure rises sharply for radiation doses to the liver parenchyma >35 Gy.

866

F. Guidoccio et al.

highest tolerable dose to the lungs is 30  Gy for a single administration and less than 50 Gy as the cumulative dose for multiple treatments [36]. The geometric mean of the lung and liver counts in the anterior and posterior views, respectively, is used to calculate the LSF by ROI analysis performed on the planar scans acquired after 99mTc-MAA administration, according to the following equation: LSF ( % ) =

Lung counts ×100 Lung counts + Liver counts

Any LSF value 20%. For LSF values between 10% and 20%, administered activity is adjusted as follows: for LSF values between 10% and 15%, activity is reduced by 20%, whereas for LSF values included between 15% and 20%, activity is reduced by 40% (see Fig. 33.4). Different models have been developed to calculate the 90Y activity allowing delivery of the highest dose to the tumor

LSF = 6.4%

while sparing normal liver tissue. The MIRD-based partition model considers three different compartments for radiation dose estimates: the tumor, the non-tumoral liver, and the lungs [37]. The partition model assumes that the LSF and distribution of 99mTc-MAAs in the tumor and non-tumor liver compartments (expressed as T/N ratio) reliably predict distribution of the 90Y-microspheres that are then administered during another interventional radiology procedure performed a few days later by the same radiologist, who tries to replicate exactly the same position of the intra-arterial catheter as during 99mTc-MAA administration. This approach is adopted for TARE/SIRT with the glass 90Y-microspheres (TheraSphere®) and yields safe and reproducible estimates of the expected toxicity and clinical outcomes. For treatment with the resin 90Y-microspheres (SIR-­ Spheres®), the activity to be administered can be estimated according to the empiric method or to the body surface area (BSA) method. For the empiric method, the fraction of liver occupied by the tumor relative to the overall liver volume is estimated by either CT or MRI.  For a tumor/liver fraction 0.5. On the other hand, a tumor/liver fraction >0.7 constitutes an absolute contraindication to treatment. The following equation is used to calculate the 90 Y-microsphere activity (A) to be administered (BSA being expressed in sq m):



A ( GBq ) = ( BSA − 0.2 ) +

Tumor volume Total liver volume

Although these two methods are routinely used, both of them frequently overestimate the activity to be administered [38, 39], and the MIRD-based partition model is recommended also when treating patients with the resin 90 Y-microspheres. Concerning instead pretreatment dosimetry estimates when using 166Ho-PLLA-microspheres, the following equation is recommended by the manufacturer to calculate the activity to be administered (AHo166) in order not to exceed a certain radiation burden to the whole liver (LD, expressed in Gy): A ( MBq ) = LD ( Gy ) × LW ( kg ) × 63 ( MBq / J ) Ho166 where LW is weight of the liver as assessed by either CT or MRI. By optimizing the formula to a maximum liver dose of 60 Gy, the following equation is obtained: A ( MBq ) = LD ( Gy ) × 3781( MBq / kg ) × LW ( kg ) Ho166 From all the above equations, it follows that accurate evaluation of the liver and tumor mass (generally based on anatomic imaging) and of 99mTc-MAA distribution (based on scintigraphic imaging) is required for patient-specific dosimetry, although the predictive value of 99mTc-MAA scintigraphy as to the distribution of 90Y-microspheres in the liver is still debated [40–42]. There are in fact some parameters that potentially induce discrepancies between intrahepatic 99mTc-­MAA distribution and 90Y-microsphere distribution, such as interval differences in catheter position during injection in the two separate interventional radiology sessions, physiologic changes in hepatic blood flow, and differences in size and shape of the 99m Tc-MAA particles and the 90Y-microspheres [41, 42]. Thus, some investigators consider as suboptimal the ability of 99mTc-MAA to predict radiation dosimetry expected from 90Y-microsphere administration; however, a definite dose-response correlation has been shown by most of the retrospective studies so far reported using this technique, with threshold values for objective tumor response varying from 120 to 205 Gy (and even up to 500 Gy) [43–45]. Nevertheless, no univocal threshold value has been identified as yet regarding objective tumor response. Dosimetry-based methods utilized to calculate the activity to be administered for TARE/SIRT are described in detail in Chap. 11 of this book.

867

Key Learning Points

• Pretreatment radiation dosimetry estimates are crucial in order to minimize possible adverse side effects of radioembolization due to excessive irradiation of nontarget tissues (either normal liver parenchyma, the pulmonary parenchyma, or other sites in the splanchnic arterial bed). • Geometric mean values derived from planar imaging are normally utilized to calculate the liver-to-­ lung shunt fraction (LSF). • No restrictions to the planned administered activity are applied for LSF values 20% constitute an absolute contraindication to radioembolization therapy. • Administered activity is reduced by 20% for LSF values between 10% and 15%. • Administered activity is reduced by 40% for LSF values between 15% and 20%. • Different models have been developed to calculate the 90Y activity allowing delivery of the highest dose to the tumor while sparing normal liver tissue, the MIRD-based partition model and the model based on body surface area for glass microspheres and for resin microspheres, respectively. • A specific equation based on threshold radiation dose to the whole liver and on liver mass is utilized when planning radioembolization therapy with 166 Ho-microspheres. • The predictive value of 99mTc-MAA scintigraphy as to the actual distribution of therapeutic radiolabeled microspheres in the liver is still debated. • A definite dose-response correlation has been shown by most of the retrospective studies so far reported.

33.7 90Y-Microsphere Administration Under fluoroscopic guidance during transcutaneous arterial catheterization (and trying to replicate as closely as possible the same catheter positioning as during pretreatment administration of 99mTc-MAA), the suspension of 90Y-microspheres is injected into the artery feeding the target lesion(s). When treatment is directed to one liver lobe, the procedure is defined as “selective,” whereas the term “super-selective” is used for treatment directed to specific liver segment(s). There are different procedures for delivering the therapeutic amount of 90Y-microspheres, either manually using a dedicated delivery system operating in a step-by-step manner, which is useful to avoid early full embolization of the arterial bed that might prevent successful infusion of the total intended amount of radioactivity. The procedure is usually performed

868

F. Guidoccio et al.

alternating injections of iodinated contrast medium and sterile water/glucose solution (when administering the resin 90 Y-microspheres) or saline solution (when administering the glass 90Y-microspheres). Continuous fluoroscopy monitoring is employed to ensure the occurrence of no stasis during infusion and to confirm that flow of the infused material is similar to that observed during the prior angiographic workup.

33.8 Posttreatment Scan The pattern of 90Y-microsphere deposition is assessed by high-quality imaging early after treatment, to confirm that no significant radioactivity accumulation in gastrointestinal tract has occurred and to evaluate the radiation-absorbed dose delivered to the tumor. Although a post-therapy planar and SPECT/CT scan based on the bremsstrahlung emission generated by the high-­energy β− particles of 90Y has been described to confirm correct deposition of the radiolabeled microspheres in the tumor lesions [46], images obtained in this manner are usually of very poor quality and do not allow an accurate quantification of microsphere distribution, especially when treating small lesions.

99mTc-MAA

SPECT/CT

Fig. 33.5  Correspondence between intrahepatic distribution of pretreatment 99mTc-MAA scintigraphy (SPECT/CT) after injection into the right hepatic artery (left) and posttreatment 90Y-PET/CT (right). For SPECT/CT, the MIP image is displayed in upper right panel, the axial CT image in upper left panel, the fused axial SPECT/CT image in lower left panel, and the 3D surface volumetric rendering in lower right panel. Similar displays for PET/CT: MIP image in upper right panel, axial CT

Instead, PET imaging based on the emission of β+ particles by 90Y provides a better way of assessing distribution of the radiolabeled microspheres [47]. In this regard, in addition to the largely predominant emission of β− particles, 90Y decays also through internal pair production that generates 511 keV annihilation photons, even if in the extremely small fraction of 32 events per million. When performing radioembolization of liver tumors with 90Y-microspheres, a very high concentration of the radionuclide is reached in a relatively small volume at the administration site; thus, even the extremely small fraction of 90Y decays that occurs through internal pair production is sufficient to acquire clinically useful PET images for validation and dosimetric purposes [48–51] (Fig.  33.5). In this manner, PET imaging allows detection of possible extrahepatic distribution of the 90 Y-microspheres, and the absorbed dose delivered during the radioembolization procedure can reliably be estimated [47, 48]. Depending on local regulation of radioprotection, treatment with 90Y-microspheres can be performed as an outpatient procedure, thus making it possible to discharge the patient from the hospital on the same day of treatment or on the next day.

90Y-PET/CT

image in upper left panel, fused PET/CT image in lower left panel, and 3D surface volumetric rendering in lower right panel (reproduced with permission from: Boni G, Guidoccio F, Volterrani D, Mariani G.  Radionuclide therapy of tumor of the liver and biliary tract. In: Strauss HW, Mariani G, Volterrani D, Larson SM, Eds. Nuclear Oncology – From Pathophysiology to Clinical Applications. New York: Springer; 2017:1337–60)

33  Radionuclide Therapy for Tumors of the Liver and Biliary Tract

Key Learning Points

• The suspension of therapeutic radiolabeled microspheres is injected through selective (directed to one liver lobe) or super-selective (directed to one or more liver segments) arterial catheterization. • Posttreatment images to verify distribution of the therapeutic 90Y-microspheres can be obtained by acquiring the bremsstrahlung emission generated by the high-energy β− particles or preferably by PET acquisition based on internal pair production. • When using 166Ho-microspheres for radioembolization, posttreatment imaging can be acquired using a standard gamma camera based on the 81 keV γ-rays emitted during decay.

33.9 P  atient Follow-Up and Assessment of Response to Therapy Clinical and radiological monitoring is useful to assess tumor response to radioembolization, while routine blood chemistry is useful to monitor possible toxicity due to treatment; when treating patients with HCC, changes in the serum leva

869

els of the tumor-associated marker alpha-fetoprotein (AFP) constitute a basis to assess evolution of the disease. In general, contrast-enhanced CT is performed at 1  month posttreatment; subsequently non-contrast-enhanced CT scans are repeated every 3 months to assess response to treatment. In addition to the general RECIST criteria, tumor response to therapy can be assessed by specific criteria for liver malignancies defined by the World Health Organization (WHO) and by the European Association for the Study of the Liver (EASL); such criteria are based on size parameters and on necrosis parameters in the target lesions, respectively [13, 52]. Metabolic assessment with [18F]FDG PET/CT provides highly useful prognostic information to assess response to TARE/SIRT with 90Y-microspheres [53–56]. Similarly as observed for other malignancies in different clinical settings, functional metabolic imaging with PET performs better than conventional criteria based on morphology, such as RECIST, for early assessment of tumor response to treatment, as convincingly demonstrated in patients with either HCC [57, 58], intrahepatic cholangiocarcinoma [56] (Fig. 33.6), metastatic CRC [59, 60] (Figs.  33.7 and 33.8), liver metastases from breast cancer [61], or metastatic neuroendocrine malignancies (the latter using 68Ga-DOTANOC as the PET tracer) [62] (Fig. 33.9).

b

Responder Non-responder censored censored

1.0

c

d

Cumulative survival

0.5

0.6

0.4

0.2

0.0

0

50

100

150

200

Weeks

Fig. 33.6  Left panel: response to trans-arterial therapy with 90 Y-microspheres in a patient with intrahepatic cholangiocarcinoma. (a) Axial fused pre-TARE [18F]FDG-PET/CT section; (b) corresponding section of diagnostic CT; (c) axial fused post-TARE [18F]FDG-PET/CT section; (d) corresponding section of diagnostic CT. SUVmax declined by 70% 3 months after TARE, and serum CA 19-9 declined from 85.2 to 49.2 U/mL; the patient was still alive 12 months after TARE without evidence of progression within the liver. Right panel: Kaplan-Meier

survival curves as a function of ΔSUV2SD. Patients responding to TARE (blue line) had significantly longer survival (P  30 Gy with a single infusion (or even 50  Gy for repeat treatment). The inability to prevent deposition of the radiolabeled microspheres in the gastrointestinal tract is an additional absolute contraindication to TARE/SIRT. Concerning instead relative contraindications, the choice of performing TARE/SIRT should be carefully evaluated in patients with reduced pulmonary function, poor functional liver reserve, serum creatinine >2.0  mg/dL, and platelet count 2.0 mg/dL, and platelet count 7, PSA >20 ng/mL, clinical stage T2c–3a), the likelihood of lymph node and bone metastases is particularly high. Several studies demonstrate the superiority of 68 Ga-PSMA-­ligand PET/CT as compared to CT, magnetic resonance imaging (MRI), or bone scintigraphy for detection of metastases for initial staging after primary diagnosis [24, 25]. Nevertheless, the 68Ga-PSMA-ligand scan cannot replace pelvic MRI for local tumor delineation (as needed by the surgeon). It is still debated whether additional functional imaging with bone-seeking agents (e.g., bone scintigraphy, 18 F-NaF PET/CT) has a relevant added value after performing a 68Ga-PSMA-ligand PET/CT, e.g., in patients with PSMAnegative tumors or densely sclerotic bone lesions (Fig. 35.11).

35.1.6.6 E  merging Clinical Applications • Staging before and during PSMA-directed radiotherapy (mainly in metastatic castration-resistant prostate cancer)

suspicious pelvic lymph nodes (T3aN0M1). 68Ga-PSMA-ligand PET/ CT detects a metastatic left pelvic lymph node and excludes liver metastasis (T3aN1M0)

912

Imaging before PSMA-directed therapy (e.g., radioligand therapy) is crucial to determine the presence and intensity of expression of the target molecule [26–28], as low PSMA expression in target lesions contraindicates radioligand therapy. • Targeted biopsy after previous negative biopsy in patients with high suspicion of prostate cancer Preliminary data indicate that 68Ga-PSMA-ligand PET/CT is of value for guidance of repeat biopsy in patients with high suspicion of prostate cancer and prior negative biopsies. In fact, the scan provides useful information to localize the primary prostate cancer [29], therefore to direct biopsies in the prostate cancer surveillance population who undergo repeat biopsies. • Monitoring of systemic treatment in metastatic prostate cancer Conventional monitoring systems such as RECIST 1.1 are limited by the high prevalence of non-measurable lesions, e.g., lymph node and bone metastases. Bone scintigraphy is limited by a potential flare phenomenon. 68 Ga-PSMA-ligand PET/CT has the potential to monitor systemic disease, although it has not yet been definitely demonstrated that this imaging modality overcomes the limitations of other modalities.

35.1.6.7 Emerging Therapeutic Applications Prostate cancer is an ideal target for the development of targeted radionuclide therapy because of the frequent occurrence of multifocal disseminated disease. In particular, recurrence of a prostate carcinoma initially classified as confined to the prostate gland and frequent presentation of the cancer in an advanced stage preclude curative surgery and radiation therapy. Furthermore, chemotherapy does not have long-term efficacy in patients with metastatic castration-­resistant prostate cancer. Radiolabeled molecules with high binding affinity for PSMA constitute the basis for targeted radionuclide therapy, with the aim of delivering sufficient radiation doses to tumor lesions while sparing the adjacent normal tissues. Lutetium-177 (177Lu) has gained popularity as the therapeutic radionuclide of choice due to its desirable physical properties; 177Lu is a β− emitter (78% abundance of 490 keV) with a maximum tissue penetration of about 2  mm; this somewhat short β− range of 177Lu provides better irradiation of small tumors. 177Lu is a reactor-produced radiometal that emits gamma rays at 208 and 113  keV with 11% and 6% abundance, respectively. The gamma emission from 177Lu allows for external gamma camera imaging and consequently the collection of information pertaining to tumor localization and dosimetry. Furthermore, 177Lu has a relatively long physical half-life of 6.73 days. It is these physical properties that allow for the delivery of high activities of 177Lu-PSMA ligand to prostate cancer cells. 177 Lu-PSMA ligand is an easily administered targeted therapy with no significant symptoms at the time of injec-

A. Al-Ibraheem et al.

tion; the main safety issues are standard radiation safety precautions that are inherent in all intravenously injected, renally excreted radionuclide therapies. Injected activities have ranged from 3 to 8 GBq per single administration with up to six injections given to men, generally at a minimum 6-week intervals; dose calculations depend on a combination of disease burden, patient weight, and renal function. 177Lu-PSMA ligand is administered by a slow intravenous injection (30–60  s) in a volume of 5 mL (diluted with 0.9% sterile sodium chloride solution), followed by a flush of sterile 0.9% sodium chloride. It is recommended that the patients are hydrated pre- and postadministration of 177Lu-PSMA ligand with 1–1.5 L of water and encouraged to void as frequently as possible. 177 Lu PSMA is excreted in the urine in the first 48  h following injection, and instructions need to be given to patients, staff, and families on managing potential radioactive spills. Given the rapid renal excretion of 177Lu-PSMA ligand, patients are required to remain in the nuclear medicine department for 2–4 h for observation and for measured radiation levels to decrease. There are currently a limited number of published trials in the use of 177Lu-PSMA ligand for the treatment of metastatic castrate-resistant prostate cancer (mCRPC) [30]. Although these are generally small, retrospective trials, they have shown almost uniformly that most men treated with 177 Lu-PSMA ligand have a significant response to treatment. 177 Lu-PSMA ligand is showing promising responses in men with mCRPC and almost certainly has an important future role in the treatment of prostate cancer. Preliminary publications suggest it has a low toxicity profile and appears generally well tolerated in men with end-stage metastatic disease. Prospective randomized trials are needed to determine its impact on survival and to rigorously assess its clinical benefit compared to other treatments of prostate cancer, including chemotherapy, external beam radiotherapy, and androgen blockade.

35.1.7 Perspectives for 18F-FACBC PET/CT The US FDA has recently approved a synthetic l-leucine analogue (trans-1-amino-3-18F-fluorocyclobutane-1-­carboxylic acid or 18F-FACBC) for PET imaging in patients with prostatic cancer. Uptake in prostatic cancer cells of this agent (also denominated 18F-fluciclovine) is mediated by neutral amino acid sodium-independent systems, through transporters associated with proliferation and angiogenesis [31–33]. High physiologic uptake of 18F-FACBC is noted in the liver, red bone marrow, lung, and pancreas [34–36]. 18F-FACBC shows a somewhat favorable biodistribution versus radiolabeled choline [37] (Fig. 35.12). In fact, this agent undergoes minimal excretion through the kidneys and urinary tract, thus resulting in improved detection of small lesions in the prostatic

35  Hybrid Imaging for Male Malignancies Fig. 35.12  MIP images from PET/CT with 18F-FACBC (left) and with [11C]choline (right), showing comparative biodistribution patterns of the two tracers (reproduced from: Nanni C, Schiavina R, Boschi S, Ambrosini V, Pettinato C, Brunocilla E, et al. Comparison of 18F-FACBC and 11C-choline PET/CT in patients with radically treated prostate cancer and biochemical relapse: preliminary results. Eur J Nucl Med Mol Imaging. 2013;40(Suppl 1):S11–7)

913

SUV = 19,5

SUV = 9,7

SUV = 0

bed and pelvis. Furthermore, PET images can be acquired as early as 4–5 min post-­administration, due to its very fast blood clearance and very rapid uptake in prostatic cancer cells. Sensitivity of PET with 18F-FACBC in patients with prostate cancer is directly correlated to PSA concentrations and inversely correlated to PSA doubling time; however, the uptake of 18F-FACBC is similar in prostatic cancer as in benign prostatic hyperplasia, thus resulting in low specificity [38]. Nevertheless, the combination of 18F-FACBC PET/CT with MRI appears to be optimal for accurate localization of cancer [37], either in the prostatic bed or at distant metastatic sites [39]. Studies comparing 18F-FACBC PET/CT with [11C]choline PET/CT found that the detection rate of tumor lesions with 18 F-FACBC PET/CT was superior to that of [11C]choline PET/CT, regardless of the PSA levels [40, 41] (Fig. 35.13). Further, larger-scale studies are necessary to confirm the potential clinical usefulness of PET/CT with 18F-FACBC for the primary diagnosis of prostatic cancer, as well as for ­staging and restaging prostate cancer patients after potentially curative treatment.

Key Learning Points

• Bone scintigraphy with 99mTc-bisphosphonates is still considered the cornerstone for detection of bone metastases in prostate cancer patients due to wide availability, good sensitivity, low radiation dose, and low cost. The specificity of bone scintigraphy can be dramatically improved by using SPECT or, better, SPECT/CT. • Clinical indications and interpretation criteria for 18 F-sodium fluoride (18F-NaF) PET/CT are similar to those for bone scintigraphy with 99mTc-­

SUV = 0













bisphosphonates, with the advantage of better sensitivity and specificity. The role of [18F]FDG PET/CT is limited to the most aggressive prostate cancers, with higher Gleason scores and biochemical failure or PSA relapse. PET/CT with [11C]CHO or 18F-FCH is most useful for detecting prostate cancer recurrence in patients with biochemical recurrence and, to a lesser extent, for the detection of prostatic recurrence, lymph node, and skeletal lesions during staging, especially in patients with high-risk prostate cancer. 68 Ga-PSMA-ligand PET/CT is one of the most promising imaging procedures for primary staging in high-risk disease and for monitoring the efficacy of systemic treatment. It is currently considered the best available radionuclide imaging modality for the evaluation of patients with biochemical failure. In patients with a positive 68Ga-PSMA-ligand PET/ CT scan, targeted radionuclide therapy with 177 Lu-PSMA ligand ensues promising responses in the treatment of metastatic castration-resistant prostate cancer. Promising results have been obtained with the synthetic l-leucine analogue (trans-1-amino-3-18F-fluorocyclobutane-1-carboxylic acid or 18F-FACBC) for PET/CT imaging in patients with prostatic cancer. Although characterized by low specificity because of 18F-FACBC uptake also in benign prostatic hyperplasia, this agent displays biodistribution and uptake characteristics that constitute potential advantages versus imaging with radiolabeled choline-based PET tracers.

914

A. Al-Ibraheem et al.

Fig. 35.13  Axial PET and fused PET/CT images in patients with corresponding positive findings (indicated by white arrows) on the [11C] choline scans (left panels) and on the 18F-FACBC scans (right panels), respectively. TBR target-to-background ratio (reproduced from: Nanni

C, Schiavina R, Boschi S, Ambrosini V, Pettinato C, Brunocilla E, et al. Comparison of 18F-FACBC and 11C-choline PET/CT in patients with radically treated prostate cancer and biochemical relapse: preliminary results. Eur J Nucl Med Mol Imaging. 2013;40(Suppl 1):S11–7)

35.2 Testicular Cancer

Seminomas, which represent approximately 50% of all germ cell tumors, are very radiosensitive. The cure rate for patients with seminoma (all stages combined) exceeds 90%. NSGCT tend to metastasize early to the retroperitoneal lymph nodes and to the lungs. Tumors with histologic mixtures of seminoma and non-seminomatous components are treated as NSGCT.  Initial work-up is testicular ultrasound followed by measurements of serum tumor-associated biomarkers, such as alpha-fetoprotein (AFP). MRI is especially useful as a problem solver in equivocal cases. Trans-scrotal needle biopsy or orchiectomy is contraindicated because these procedures can potentially lead to nonperitoneal lymphatic dissemination of the tumor cells. CT is the primary

35.2.1 Clinical Background Testicular cancer is a relatively rare malignancy comprising 1% of all the diagnosed cancer cases. Although rare, it is the most frequent solid non-hematological cancer diagnosed in young adult men, and its incidence is increasing. The most common presentation is a painless, hard, testicular lump. Histologically, testicular tumors can be of germ cell origin or non-germ cell origin. Germ cell tumors (GCT) are the most common testicular cancers, further divided into seminomas and non-seminomatous germ cell tumors (NSGCT).

35  Hybrid Imaging for Male Malignancies

imaging modality for assessing lymph node status and for evaluation of distant metastasis. Since bone metastases are rare in testicular cancers, bone scintigraphy is not indicated. TNM staging is conventionally used for testicular cancer. Accurate initial staging is very important to classify patients into low- or high-risk groups, because management differs between the two groups. Accurate restaging post-chemotherapy is also important to stratify patients for further management plans.

915

management plan for surveillance versus adjuvant chemotherapy [43]. Eight of the 11 patients with equivocal CT findings had a true-negative [18F]FDG PET/CT scan, while 3 patients had a true-positive [18F]FDG PET/CT scan. These findings are consistent with previous reports leading to the conclusion that [18F]FDG PET/CT is most useful in patients with equivocal CT scans [44, 45]. The five high-risk patients with a negative CT scan had negative [18F]FDG PET/CT scan, but two of them subsequently relapsed. The authors concluded that [18F]FDG PET/CT is helpful in primary staging when the primary staging CT 35.2.2 [18F]FDG PET/CT in Testicular Cancer scan is equivocal, but it is not sensitive enough to predict occult metastases in high-risk patients with a negative CT 35.2.2.1 Indications scan [43]. • Diagnosis and staging • Elevated serum tumor markers There is at present no sufficient evidence justifying to [18F]FDG PET/CT plays a clinically relevant role in 18 recommend the use of [ F]FDG PET/CT for primary GCT patients who exhibit elevated serum levels of tumor diagnosis or staging of testicular cancer, although the markers associated with a negative CT scan, since in a results of the scan are frequently accurate and useful for large percentage of these patients PET imaging detects planning the most adequate treatment strategy. In a recent sites of the disease [46]. In a series of 47 [18F]FDG PET/ 18 meta-­analysis of 16 articles including an overall 957 [ F] CT scans obtained for assessing residual masses (18 with FDG PET scans in 807 patients, pooled sensitivity and raised serum tumor markers) and 23 scans for investigatspecificity of diagnostic [18F]FDG PET/CT were 75% and ing patients with raised serum markers associated with 87%, respectively [42]. This sensitivity, with a negative normal CT findings, it was found that all but one of the likelihood ratio of 0.31, is not sufficient to confidently [18F]FDG PET/CT scans identified the site of disease exclude testicular malignancy and metastases. False[46]. In this series, however, the NPV was as low as 50%, negative studies are mostly due to the partial volume thus indicating that a negative scan did not reliably preeffect of small lesions and microscopic tumor foci in semdict the absence of disease. On the other hand, subsequent inoma and NSGCT; in addition, teratoma components of [18F]FDG PET/CT imaging in those patients with negaNSGCT do not demonstrate significant metabolic activity tive scans was frequently the first imaging modality that on [18F]FDG PET/CT [42]. Furthermore, false-positive identified the site of disease. Therefore, in the presence of findings have occasionally been reported, due to infecraised serum tumor marker levels and negative imaging tive/inflammatory processes or to inadequate timing post-­ findings, including a negative [18F]FDG PET/CT, the chemotherapy [42]. most appropriate follow-up imaging may be to repeat • Equivocal findings on the CT scan [18F]FDG PET/CT [46]. As a functional/metabolic imaging procedure, [18F] • Post-chemotherapy residual seminoma GCT masses FDG PET/CT offers in principle several advantages over Post-chemotherapy residual masses and metastatic semCT. Since detection of metastatic lymph node disease at inoma occur in approximately 25% of the cases [47]. [18F] CT is based on the size criterion (>1 cm diameter of short FDG PET/CT is indicated for evaluation of post-­ axis) and morphology (round lymph nodes >0.8 cm), this chemotherapy residual masses and metastatic seminoma. imaging modality lacks the desired accuracy for characThis is currently the only indication for [18F]FDG PET/CT in testicular cancer endorsed by NCCN guidelines, particuterizing lymph nodes, and sensitivity can be low as 36%. larly when the residual mass exceeds 3  cm in maximal In a recent study by Cook et al., [18F]FDG PET/CT was used at initial staging of 16 patients (10 with seminoma, 5 diameter. The prospective SEMPET study shows that [18F] with NCGCTs, 1 with mixed GCT). Eleven patients FDG PET/CT can confidently distinguish residual disease underwent [18F]FDG PET/CT to clarify equivocal findfrom posttreatment fibrosis [48]. Several other studies have ings on the staging CT scan, usually consisting of para-­ shown the superiority of [18F]FDG PET/CT compared with aortic lymph nodes that were in the expected drainage CT in predicting viable tumor in seminoma residual masses from the primary tumor (such as the retroperitoneal after chemotherapy, with accuracy >90% [49, 50]. basin), but were less than the 1  cm cutoff size value A retrospective validation of the SEMPET trial includrequired to be classified as metastasis on CT; in addition, ing 127 patients yielded sensitivity, specificity, negative five high-risk patients with a negative primary staging CT predictive value, and positive predictive values of [18F] 18 FDG PET/CT at 50%, 77%, 91%, and 25%, respectively scan underwent [ F]FDG PET/CT to aid decisions on

916

A. Al-Ibraheem et al. Column A

Column B

Fig. 35.14  Trans-axial sections at the abdominal level of a patient with metastatic seminoma submitted to repeat [18F]FDG PET/CT scans for assessment of response to chemotherapy (CT images in top row, PET images in middle row, fused PET/CT images in bottom row). Column A: baseline scan demonstrating a bulky hypermetabolic retroperitoneal

[51]. Correct timing of PET scans is critical and may substantially reduce false-positive results. [18F]FDG PET/CT should be performed not earlier than 6 weeks after the end (day 21) of the last chemotherapy cycle [52]. In order to reduce the post-chemotherapy false-positive cases, it is currently recommended to repeat the [18F] FDG PET/CT scan after at least 6 additional weeks. Repeating [18F]FDG PET/CT is a helpful strategy particularly if residual disease is considered clinically unlikely; it is also helpful to guide biopsy in those patients who otherwise can also be falsely classified as negative. If the repeat scan is not performed, the residual mass must then be biopsied [52, 53] (Fig. 35.14).

Column C

metastatic lymphadenopathy. Column B: post-chemotherapy scan demonstrating residual lymph node without significant [18F]FDG uptake, indicating complete response to treatment. Column C: follow-up scan at 12 months confirming the absence of any significant metabolic activity, indicating long-lasting remission of the disease

• Post-chemotherapy residual NSGCT masses [18F]FDG PET/CT is not routinely indicated for post-­ chemotherapy restaging in NSGCT patients [52]. Residual masses after completion of first-line chemotherapy are found in approximately 40% of patients, even after normalization of serum tumor markers [54]. Histology of the resected lesions reveals necrosis in 40%, vital carcinoma in 20%, and mature teratoma in 40% of cases [55]. Mature teratomas are completely chemoresistant; in addition, they carry the risk of subsequent malignant transformation, and in that case they usually become [18F]FDG-avid. Therefore, surgical dissection is indicated for all residual NSGCT masses larger than 1 cm [52].

35  Hybrid Imaging for Male Malignancies

In a prospective trial including 121 stage IIC or III NSGCT patients, [18F]FDG PET/CT was performed after completion of chemotherapy, and the results were correlated with histopathology, CT scan, and serum tumor markers [56]. [18F]FDG PET/CT predicted tumor viability in only 56% of the cases, with approximately the same accuracy as CT (55%) or tumor markers (56%). This prospective, histology-controlled study demonstrated that [18F]FDG PET/CT does not yield a clear additional clinical benefit with respect to the standard diagnostic procedures (CT and serum tumor markers) for predicting tumor viability in residual masses [56]. On the other hand, [18F]FDG PET/CT can be used to localize post-chemotherapy hypermetabolic tissue, therefore serving as a guide for biopsy or surgical intervention. • Active surveillance There is still no evidence to support the use of [18F] FDG PET/CT in the follow-up of testicular cancers during active surveillance.

35.2.2.2 Contraindications None. 35.2.2.3 Procedure After a fast for at least 4  h and checking blood glucose level to be less than 200 mg/dL, the patient is injected with 2–4 MBq/kg of [18F]FDG intravenously. After the 60–90 min uptake period, the patient is positioned with the arms elevated and supported above the head to avoid beam-hardening artifacts in the abdominal and pelvic regions, as well as artifacts caused by truncation of the measured FOV. If the patient cannot tolerate this position, one arm can be kept above the head with the other positioned alongside the body, or both arms can be positioned alongside and close to the body. First, a CT topogram is used to define the coaxial imaging range. Axial anatomical scan coverage range from the base of the skull to the mid-thigh is sufficient. A standard diagnostic CT scan with intravenous contrast agent may, if appropriate, be performed. 35.2.2.4 Side Effects None. 35.2.2.5 Data Processing The PET emission data must be corrected for geometrical response and detector efficiency (normalization), system dead time, random coincidences, scatter, and attenuation. All the corrections necessary to obtain quantitative image data should be applied during the reconstruction process. When available, time-of-flight information should be used during reconstruction. Resolution modeling during reconstruction or other new reconstruction or image processing

917

methods may be applied. The CT attenuation-corrected scan data are reconstructed by filtered back projection or iterative algorithms available and reformatted in coronal, sagittal, and trans-axial slices.

35.2.2.6 Interpretation Images must be evaluated using software capable to display fused PET and CT data and use an SUV scale. PET images should be displayed with and without attenuation correction to identify artifacts possibly caused by contrast agents, metal implants, and/or patient motion. The presence or absence of abnormal [18F]FDG accumulation on the PET images, in combination with level of uptake and morphologic correlation, should be evaluated. The absence of tracer accumulation in anatomical abnormalities seen on the CT scan or other imaging may be particularly significant. It is important to review these images on a workstation that has the capacity to triangulate findings in axial, coronal, and sagittal planes. The PET scan is deemed positive if there is increased [18F]FDG uptake above the normal surrounding tissue in an area where metastases could be found. Any lesions identified on the PET component are then correlated with the CT images, reviewing soft tissue, lung, and bone windows as appropriate to the location of the abnormality (Fig. 35.15). As a final step, the CT images are sequentially reviewed using soft tissue, lung, and bone windows to identify structural abnormalities not previously identified on PET imaging. If the [18F]FDG PET/CT scan is negative in patients with seminomas and residual mass larger than 3  cm, no further treatment is needed. On the other hand, caution must be paid to NSGCT patients with residual masses detected on the CT scan; lesions less than 1.0 cm may still harbor residual disease and therefore must be interpreted with caution. [18F] FDG PET/CT has a limited NPV in NSGCT patients with residual masses, but still might be useful for guiding multimodality treatment [57]. 35.2.2.7 Reporting The PET/CT report should include the study identification, clinical information, procedure description, description of the findings, and brief impression. When appropriate, the report should correlate PET/CT findings with those of other diagnostic tests, interpret them in that context, and consider them in relation to the clinical data. About 20% of patients with seminoma tumors have metastases at the time of diagnosis, most often lymph node metastases in the retroperitoneum and/or the supraclavicular region. NSGCT often displays rapid spread, and about 50% of patients have metastases at the time of diagnosis [58]. The PET scan is deemed positive if there is increased [18F] FDG uptake above the normal surrounding tissue in an area where metastases could be found. For both seminoma and non-seminoma testicular cancer, metastasis occurs primarily

918 Fig. 35.15  MIP and axial images of [18F]FDG PET/CT scans performed in a young man with history of metastatic seminoma. In the baseline scan (upper panels), there is a large left mass with markedly increased tracer uptake (arrows in upper panels); the mass extends from the lower border of T12 to the lower border of L4, measures 8 × 7 × 12 cm, and has SUVmax 20 at the level of renal hilum. The [18F]FDG PET/CT performed after chemotherapy (lover panels) demonstrates excellent response to treatment, with a residual hypometabolic mass (arrow in lower panel) (reproduced from: Strauss HW, Mariani G, Volterrani D, Larson SM, eds. Nuclear Oncology – From Pathophysiology to Clinical Applications, 2nd Edition. New York: Springer, 2017)

A. Al-Ibraheem et al.

35  Hybrid Imaging for Male Malignancies

919

to the retroperitoneal lymph nodes and subsequently further along the thoracic duct. Metastasis can also occur through the hematogenous route, especially in patients with NSGCT [59].

or subtropical regions of Latin America, Asia, and Africa. Incidence of penile cancer is also affected by race and ethnicity, with the highest incidence (1.01 per 100,000) in North American white Hispanics [61]. 35.2.2.8 Future Perspectives Risk factors include phimosis, chronic inflammatory con• Initial reports indicate a predictive and prognostic role of ditions such as balanoposthitis and lichen sclerosus et atro[18F]FDG PET/CT for early interim metabolic assessment phicus, sexual history (multiple partners, early age at first in patients with metastatic seminoma who are receiving intercourse), history of condylomata, and smoking. Human chemotherapy [60]. These results imply the possibility to papillomavirus DNA has been identified in 70–100% of personalize the number of chemotherapy cycles. intraepithelial neoplasms and in 40–50% of invasive penile • [18F]FDG PET/CT has an established role in seminoma cancers. The virus plays an important role in oncogenesis for evaluating whether further treatment is necessary after through interaction with oncogenes and tumor suppressor chemotherapy for residual masses >3  cm. However, its genes [61]. impact on long-term disease outcome and cost-­ Ultrasonography (US) is useful for evaluating suspected effectiveness still needs to be established. penile masses and is recommended as the first-line imaging • PET probe-guided surgery might be useful in surgical modality for primary penile cancer deemed to invade the oncology for recurrent testicular cancer in difficult-to-­ corpora cavernosa [62]. Squamous cell carcinoma (SCC) access locations. usually presents as a hypoechoic lesion with heterogeneous • The development of newer radiotracers with different appearance. However, US cannot reliably identify the presmetabolic pathways may be promising in the evaluation ence of microscopic invasion. On MR imaging, penile carof NSGCT. cinoma is often a solitary, ill-defined, and infiltrating mass hypointense with respect to the adjacent corpora on both T1- and T2-weighted images [63]. The local extent, depth of tumor invasion, and extension to adjacent structures can Key Learning Points 18 be accurately depicted using MR, thus providing valuable • In patients with seminoma, [ F]FDG PET/CT has information for surgical planning [63]. high specificity, sensitivity, and NPV for evaluating However, both clinical examination and conventional post-chemotherapy residual masses, as PET imagimaging techniques, including US, CT, and MRI, are unreing is very effective in distinguishing residual tumor liable in accurately detecting lymph node metastases [62]. from fibrosis. Hence, prophylactic inguinal lymphadenectomy is often • Correct timing of the PET scan for assessing tumor recommended to improve survival, particularly in high-risk response to therapy is crucial, as it may substan18 patients. Nevertheless, the morbidity of groin dissection is tially reduce the false-positive results. [ F]FDG high, with complication rates between 30% and 50%. This PET/CT should be carried out not earlier than has stimulated interest in the search for other techniques able 6 weeks after the end (day 21) of the last chemoto noninvasively assess the lymph node status and possibly therapy cycle. 18 reduce the need for groin dissection [64]. • [ F]FDG PET/CT can be helpful in patients with elevated serum levels of tumor markers, whether or not a residual mass is seen on CT. • In non-seminomatous tumors, sensitivity of [18F] FDG PET is hampered by a high rate of false-negative results, particularly in mature teratomas.

35.3.2 Bone Scintigraphy Since bone metastases are rare in patients with penile cancer, bone scintigraphy is not commonly used.

35.3 Penile Cancer

35.3.3 Sentinel Lymph Node Biopsy

35.3.1 Clinical Background

35.3.3.1 Indications Sentinel lymph node biopsy for lymph node staging in penile cancer is indicated in patients with non-palpable lymph nodes. The sensitivity of SLNB in patients with penile cancer with non-palpable lymph nodes ranges between 70% and

Penile carcinomas, rare in most developed nations including Europe and the United States (with an incidence 1 cps) may indicate damage of the electronic circuit.

44.2.4 Sensitivity/Constancy 44.2.4.1 Purpose To check constancy and reproducibility of the count rate detected by the probe in standard conditions. Sensitivity is expressed in counts per second per unit activity (cps/MBq). 44.2.4.2 Procedure A long-lived reference source or a set of calibrated reference sources (such as 57Co, 133Ba, 68Ge, or 127Cs) should be available for daily checks of counting-rate constancy. Cobalt-57 is the preferred radionuclide, since it has a relatively long physical half-life (270 days) and a gamma energy of 122 keV (similar to that of 99mTc) and will often be available in the Nuclear Medicine Department in the form of a “pencil marker” or “disk marker” for routine use during patient positioning for nuclear medicine imaging. Perform measurements in a constant geometry with respect to the probe; ideally, each reference source should be incorporated into some sort of cap (source holder) that fits reproducibly over the probe, so as to avoid spurious differences in counting rates due to variations in the source–detector geometry. Record the count rates (in cps) obtained with all energy window settings and for each probe and collimator to be used clinically. A counting time sufficient to measure a minimum of 5000 counts should be set and the counts per second recorded. Note that some systems may only allow for a short-fixed counting time (e.g., 10 s); in this case it may be necessary to repeat this count interval several times until reaching a total of at least 5000 counts. Note that the expected value (the “reference mean value”) should be established when the quality assurance program is first set up. For initial assessment, measure sensitivity with exactly the same setup as for routine checking, recording at least 10,000 counts in a set time. Calculate the count rate (cps) and record this as the reference value. This reference value must be corrected as the standard source decays over time (e.g., 57Co has a half-life of 270 days).

1081

A change greater than ±5% (but smaller than ±10%) with respect to the baseline value suggests the need for further tests, as it may indicate an inappropriate energy window setting or some other technical problem. If the deviation from the reference value exceeds ±10%, stop to use the probe clinically.

Key Learning Points

• Constancy in sensitivity is a good test to evaluate proper operation of the probe (appropriate energy settings). • Check constancy before use. • Establish a reference value during reference tests.

44.2.5 Counting Precision 44.2.5.1 Purpose Counting precision is a measure of the stability of the whole system. The test is intended to check that the uncertainty in the measurement is primarily due to the random nature of radioactive decay; this test checks for stability of the overall electronics and cable connections. 44.2.5.2 Procedure A long-lived reference source or a set of reference sources should be used. Perform measurements in a constant geometry with respect to the probe, and record counts. A counting time sufficient to collect a minimum of 10,000 counts should be set. Repeat measurement ten times. Calculate mean value and standard deviation and apply the chi-squared test.

Key Learning Points

• Serial measurements of counting rate are useful to check for stability of the overall electronic circuit.

44.2.6 Energy Window and Energy Resolution 44.2.6.1 Purpose The test is intended to check that the energy windows set for the radionuclides are correct and to define the FWHM of the energy spectrum (energy resolution). It can only be performed with systems where the manufacturers have planned to include the possibility to assess the energy spectrum.

1082

Key Learning Points

• Correct energy window settings should be checked before use.

44.2.7 Sensitivity in Air 44.2.7.1 Purpose To test with periodic quality control the absolute value of sensitivity. 44.2.7.2 Procedure The procedure is the same as the sensitivity test, but a calibrated reference source should be used. Perform measurements in a constant geometry with respect to the probe (distance, scatter medium), and record cps or counts over a fixed time interval. Perform ten measurements, and calculate the mean value of cps/MBq. Note that the expected value (the “reference mean value”) should be established when the quality assurance program is first set up. Repeat measurements with all energy window settings and for each probe and collimator to be used clinically.

7×104 6×104 5×104 4×104 Cps

44.2.6.2 Procedure Select the default energy setting for the radionuclide tested. Observe the display to ensure that the corresponding photopeak is centered on the windows settings. Energy resolution may be evaluated by the percent FWHM of the photopeak energy. For semiconductor probes (that are characterized by an asymmetric energy spectrum versus the Gaussian-type spectrum of scintillator probes), the right side of the spectrum should be analyzed and the corresponding semi-width obtained. If the photopeak position is distant more than half of the energy resolution from the appropriate value, stop to use the probe.

A. Vaiano

3×104 2×104 1×104 0

0

2

4

6 x(cm)

8

10

12

Fig. 44.1 Example of sensitivity curve versus source-to-probe distance

44.2.8.2 Procedure Use a point-like source and PMMA slabs to simulate a scatter medium. Record cps moving the source in contact, then at growing distances from the probe, by positioning multiple PMMA slabs between the source and the probe. Place the point-like source in a source holder (PMMA slab with a hole for a 99mTc drop or a plastic holder for a solid source), in order to avoid variations in the source–detector geometry. Plot cps versus source-to-probe distance (see example in Fig. 44.1)

Key Learning Points

• Evaluate sensitivity for acceptance testing and when repairs can affect this parameter.

44.2.9 Spatial Resolution and Angular Resolution in a Scatter Medium

Key Learning Points

• Evaluate sensitivity in scatter medium to be in a situation close to clinical. • Use a well-defined source–detector geometry.

44.2.8 Sensitivity in a Scatter Medium 44.2.8.1 Purpose To test sensitivity in a physical setup that mimics the use of the system during clinical applications

44.2.9.1 Purpose To evaluate the spatial resolution and the angular resolution in a scatter medium 44.2.9.2 Procedure Use PMMA slabs to simulate a scatter medium and place a point-like source at a 30 mm distance from the probe surface (Z = 30 mm in Fig. 44.1). Use a point-like source placed in a source holder (PMMA slab with a hole for a 99mTc drop or a plastic holder for solid source). Select radionuclide and energy window.

44  Standard Operating Procedures for Quality Control of Instrumentation for Radioguided Surgery

1083

Fig. 44.2  Setup for spatial and angular resolution tests X

Source axis

PROBE

Probe axis

Source holder

Z

Move the source along the X direction (Fig. 44.2), and collect counts with prefixed counting time or preset cps at different X positions (2 mm steps). Plot cps or counts versus the X position. Spatial resolution is obtained from the FWHM (mm) of the curve (point spread function) (Fig. 44.3a). Calculate the semi-opening angle as

PMMA SLABS

a

Signal

θ1/ 2 = arc tg ( FWHM / 2 Z )

X

and angular resolution as 2θ1/2 (degree) (Fig. 44.3b).

Y

Target

Key Learning Points

• Place a point-like source at a 30 mm distance (far field) from the probe surface. • Measure spatial and angular resolution.

b

farfield

cm

source

30

44.2.10 C  ount Rate Capabilities in a Scatter Medium 44.2.10.1 Purpose To test the count rate performances of the probe in clinical situation; it is especially important when quantitative measurements are to be made. In particular, the maximum activity that the probe can correctly measure (linear response limit) is determined; this value depends on dead time of the electronics. Typical limits are 2500  cps for a scintillator probe and 10,000 cps for a semiconductor probe. It is not very important during clinical use because the activity in the field of view usually is much lower than the probe limit. This parameter is assessed during acceptance testing.

nearfield

angle of irradiation θ

probe axis

Fig 44.3  Spatial resolution test (a) and angular test (b)

44.2.10.2 Procedure Use a point-like 99mTc source with an activity that yields about 40  kcps. Acquire for a fixed geometry about 10,000 counts, and repeat the acquisition at different times during radioactive decay until the count rate is about 500 cps. Plot counts versus time. Fit low cps data with a linear curve. Determine cps and the value of activity for which the response of the probe is no more linear (i.e., measured cps

1084

A. Vaiano

Fig. 44.4  Setup for the lateral shielding test

C1

are 10% less than real cps): this is the activity limit to be adopted for a proper response of the probe.

Key Learning Points

• It is an important performance to identify/localize low activity sources embedded in a high background activity. • Check after physical damage of the probe.

Key Learning Points

• It is important for quantitative measurements. • It depends on electronic circuit. • Assessment to be performed during acceptance testing.

44.2.11 Lateral Shielding 44.2.11.1 Purpose To check proper lateral shielding of the probe. This parameter is very important since probes are often used intraoperatively to locate low activity sources near to high activity sources. Physical damage of the probe can compromise this parameter. 44.2.11.2 Procedure Use a point-like source. Place the source in contact with the surface of the probe, and acquire counts for a prefixed time or cps (C1), and then move the source in a lateral position as shown in Figure 44.4, and acquire counts for the same prefixed time or cps (C2). Calculate the transmission (T) from lateral shielding through the ratio:

C2

T=

C2

C1

An example of a practical spreadsheet to be used for recording of the data and calculation of the relevant performance parameters is accessible online by clicking on the link provided in the first page of this Chapter.

References 1. NEMA.  NEMA standards publication NU 3-2004. Performance measurements and quality control guidelines for non-imaging intraoperative gamma probes. Rosslyn, VA: National Electrical Manufacturers Association; 2004.. www.nema.org 2. EANM Physics Committee. Acceptance testing for nuclear medicine instrumentation. Eur J Nul Med Mol Imaging. 2010;37:672–81. 3. EANM Physics Committee. Routine quality control recommendations for nuclear medicine instrumentation. Eur J Nucl Med Mol Imaging. 2010;37:662–71.

Technical Aspects of Dual-Energy X-Ray Absorptiometry and Quantitative Computed Tomography for Assessment of Bone Mineral Density and Body Composition

45

Pat Zanzonico

Contents 45.1  Introduction 

 1086

45.2  Three-Compartment Model of Body Composition 

 1087

45.3  Alternative Techniques to DEXA and QCT  45.3.1  Conventional Radiography  45.3.2  Radiogrametry  45.3.3  Radiographic Absorptiometry  45.3.4  Single-Photon Absorptiometry  45.3.5  Ultrasonometry (QUS) 

 1087  1087  1087  1087  1087  1088

45.4  Production of Dual-Energy X-Ray Beams 

 1088

45.5  The DEXA Algorithm for Measurement of Bone Density and Body Composition 

 1090

45.6  Analysis of DEXA Studies 

 1092

45.7  Quantitative Computed Tomography (QCT) 

 1093

45.8  Quality Assurance and Calibration of DEXA and QCT Instrumentation 

 1095

45.9  Radiation Dosimetry of DEXA and QCT 

 1097

45.10  Concluding Remarks 

 1098

References 

 1098

Learning Objectives

• To understand the nature of osteoporosis and its scope as a public health issue. • To elucidate the basic role of dual-energy x-ray absorptiometry (DEXA) and quantitative computed tomography (QCT) in the characterization of osteoporosis. • To understand the characterization of body composition. • To enumerate the alternative approaches to DEXA and QCT for noninvasive measurement of bone mineral density (BMD). • To understand the limitations of alternative approaches to DEXA and QCT for noninvasive measurement of BMD.

P. Zanzonico (*) Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, NY, USA e-mail: [email protected]

• To describe the various approaches to producing dual-­ energy photon beams for DEXA. • To understand the basic physics of x-ray production. • To describe the x-ray beam geometries which have been used in DEXA systems. • To describe the basic mathematical algorithm for deriving BMD and body composition by DEXA. • To describe the calibration required for derivation of BMD and body composition by DEXA. • To identify the anatomic sites included in a typical DEXA study. • To describe the parameters commonly used to characterize the results of DEXA studies. • To describe the basic operating principle of modern CT scanners. • To describe the calibration phantoms used to derive body composition from CT studies.

© Springer Nature Switzerland AG 2019 D. Volterrani et al. (eds.), Nuclear Medicine Textbook, https://doi.org/10.1007/978-3-319-95564-3_45

1085

1086

P. Zanzonico

• To understand the basic quality assurance and calibration procedures for DEXA and QCT instrumentation. • To understand, qualitatively and quantitatively, the radiation doses to patients from DEXA and QCT procedures.

45.1 Introduction Osteoporosis is a systemic disorder of the skeleton which is characterized by low bone mass and micro-architectural deterioration of bone tissue (Fig.  45.1) with a resulting increase in bone fragility and susceptibility to fracture [2, 3]. Osteoporosis is a major national health problem which affects an estimated 25 million men and women. Osteoporotic fractures cause significant morbidity and mortality for the individual and incur a significant economic cost. Vertebral fractures cause loss of height, pain, and deformity. Hip fractures pose an even more significant risk of death and disability compared to vertebral fractures: of the 300,000 hip fractures which occur every year in the United States, more than 30,000 will die within the first year after the fracture. Most who suffer a hip fracture never fully recover, and many never walk again unassisted. Nearly half will require nursing home care after the fracture. The cost of hospitalization, rehabilitation, and long-term care of individuals who have suffered a hip fracture totals more than $10 billion annually. Since bone strength is proportional to bone mass, measurement of bone mass (i.e., hydroxyapatite mass), or what is commonly called “bone mineral density” (BMD), provides a

Normal Bone

Fig. 45.1  Scanning electron micrographs of bone biopsies from a normal individual (left panel) and an osteoporotic patient (right panel). The normal bone architecture is characterized by thick, structurally strong interconnections among plates of bone. Much of this bone min-

means of diagnosing osteoporosis and to estimate an individual’s future fracture risk. BMD is commonly expressed as the mass of mineralized tissue per unit area scanned (g/cm2) for two-dimensional modalities such as dual-energy x-ray absorptiometry (DEXA) or as the mass per unit volume of bone (g/cm3) for three-dimensional modalities such as quantitative computed tomography (QCT) [4]. Such quantitative measurements of BMD of the lumbar spine, hip, forearm, and heel have proven to be extremely accurate in estimating fracture risk [5]. DEXA and QCT are well-established x-ray imaging modalities for noninvasive measurement of BMD and body composition (i.e., relative amounts of fat, lean tissue, and bone). These techniques have been widely applied in the clinical management of osteoporosis and spontaneous and iatrogenic bone-wasting conditions and metabolic diseases. Since its introduction in the mid-1980s, DEXA has emerged as the modality of choice for measurement of BMD and body composition and rapidly superseded alternative modalities, including QCT, for performing such measurements in routine clinical practice. When evaluating BMD using DEXA to diagnose osteoporosis, there are several common measurement sites, including the lumbar spine, the proximal hip (i.e., femur), and the forearm (i.e., humerus). The standard protocol is to scan two of these sites, typically the spine and hip. If one of these sites is not available, then the forearm is used [6–9]. This chapter will review the basic physics and technology of DEXA and QCT.

Osteoporotic Bone

eral is lost in osteoporosis, with much thinner and, in some instances, discontinuous interconnections among the bone plates. This leads to an overall weakening in bone and a greater likelihood of fracture (from reference [1])

45  Technical Aspects of Dual-Energy X-Ray Absorptiometry and Quantitative Computed Tomography for Assessment of Bone…

Key Learning Points

• Osteoporosis is a systemic disorder of the skeleton characterized by low bone mass and deterioration of bone micro-architecture. • Bone mass (i.e., hydroxyapatite mass), or “bone mineral density” (BMD), provides a means of diagnosing osteoporosis and estimating an individual’s future fracture risk. • DEXA and QCT are well-established x-ray imaging modalities for noninvasive measurement of BMD and body composition. • Quantitative measurements of BMD have proven to be extremely accurate in estimating fracture risk.

45.2 T  hree-Compartment Model of Body Composition In the context of bone density and body composition measurements by DEXA, the body is assumed to be composed of three basic tissues based on their distinct x-ray attenuation properties: bone mineral, lipid (fat), and lipid-free soft tissue (or muscle) [4]. For each pixel in a DEXA image, the masses of these three tissue components are quantified. However, the distribution of the bone mineral, lipid, and non-lipid soft tissue within the volume projected onto the image pixel is not known. The model, however, “forces” all tissue types into these three components.

Key Learning Point

• In the context of DEXA, the body is assumed to be composed of three basic tissues, bone mineral, lipid (fat), and lipid-free soft tissue, based on their distinct x-ray attenuation properties.

45.3 Alternative Techniques to DEXA and QCT In addition to DEXA and QCT, a number of noninvasive techniques have been developed and applied for noninvasive measurement of BMD. None, however, has achieved the widespread clinical use of QCT and, in particular, DEXA. For completeness, however, these competing techniques are briefly reviewed [6, 10].

45.3.1 Conventional Radiography While plain radiographs are excellent for visualizing skeletal anatomy, the subjective evaluation of a radiograph is not

1087

quantitative and cannot reliably measure the amount of bone present at a skeletal location. More than 30% of the bone mass must be lost before the loss can be recognized radiographically.

45.3.2 Radiogrametry Radiogrametry, or radiographic morphometry, attempts to utilize radiographs to measure bone loss by quantifying the anatomic dimensions of specific bone sites, such as the thicknesses of the bone in the midshaft regions of the phalanx or metacarpal. Such cortical thickness data have been shown to be useful for comparative group studies of skeletal size of men and women of different races and of changes in skeletal size related to aging and various drugs. Since radiogrametry’s precision is limited (i.e., 3–5%) and the change in an individual’s cortical thickness is usually small, radiogrametry is unable to reliably detect early bone loss or to monitor bone change in an individual.

45.3.3 Radiographic Absorptiometry In radiographic absorptiometry (also known as photodensitometry), a plain radiograph of the hand and an aluminum step wedge is acquired using a conventional x-ray unit. The optical densities of the radiographic images of the bone and along the aluminum wedge are measured using a densitometer (optical density is the log10 of the ratio of light intensity incident on the radiographic film to the intensity of the light transmitted through the film). The bone mass is determined as equivalent to the mass of aluminum at which the optical densities corresponding to the bone and the aluminum wedge in the film are equal. Technical difficulties related to x-ray beam hardening and scatter limit this technique to measurement of the extremities. The precision of such measurements is 2–4%. While simple and inexpensive, the diagnostic reliability of radiographic absorptiometry has not been established.

45.3.4 Single-Photon Absorptiometry An important advance in the field of bone densitometry, and one that might be considered the predecessor of DEXA, was the development of single-photon absorptiometry (SPA) by Cameron and Sorenson in 1963 [6, 11]. This technique used a highly collimated radioactive source with activity of the order of 37–370 MBq (1–10 mCi) of either iodine-125 (125I) or americium-241 (241Am) to produce a monoenergetic gamma-ray beam with an energy of either 27 or 60  keV, respectively. The subject placed his or her arm (i.e., his or her humerus) in a water bath to provide a uniform path length

1088

through which the gamma rays would pass as the source and an opposed detector traveled in a raster pattern across and along the length of the humerus [11]. The data thus acquired allowed the calculation of the mass of bone tissue in the region scanned by means of subtraction of the photons attenuated by the soft tissue from those attenuated by bone and soft tissue. This technique proved to be useful for bone quantification but was limited to a peripheral site. In addition, it required the use of a radioactive source, introducing issues of relatively low photon flux and resulting pronounced statistical uncertainty (i.e., noise) in the measured data, periodic source replacement (at least in the case of 125I, which has a 60-day half-life), and source security.

45.3.5 Ultrasonometry (QUS) Clinical ultrasound units for quantitative assessment of skeletal status have been available since the 1990s, and there are currently a wide variety of such units approved for this clinical application. Because of its noninvasive nature, quantitative ultrasonometry (QUS) is attractive. The calcaneus is the most commonly used site for QUS because the heel is 90% trabecular bone (with a much higher metabolic turnover rate than cortical bone and thus exhibits metabolic bone alterations before cortical bone) and because it is a conveniently accessible anatomic site; the phalanges and tibia are other sites for QUS.  Different QUS systems measure different parameters, such as broadband ultrasound attenuation, the speed of sound in bone, and the modulus of elasticity. Although the precision of QUS-based measurements of BMD is generally poorer than that of QCT and DEXA, it has been shown to discriminate individuals with fracture from age-matched normal individuals with comparable statistical power to that of the x-ray absorptiometry techniques.

Key Learning Points

• In addition to DEXA and QCT, conventional radiography, radiogrametry, radiographic absorptiometry, single-photon absorptiometry, and ultrasonometry (QUS) have been used for noninvasive evaluation of BMD. • No alternative methods have achieved the widespread use and clinical utility of DEXA and QCT for evaluation of BMD.

45.4 P  roduction of Dual-Energy X-Ray Beams A key requirement for DEXA is a dual-energy x-ray beam. The term, “dual-energy x-rays,” refers to x-rays comprised

P. Zanzonico

of two energy spectra, as opposed to a single-energy spectrum characteristic of conventional x-ray systems. The predecessor of DEXA, dual-photon absorptiometry, utilized an ~37-GBq (~1-Ci) gadolinium-153 (153Gd) source whose energy spectrum includes photopeaks (nominally) at 44 and 100 keV (europium K x-rays, 42 and 48 keV; gamma rays, 97 and 103  keV) [12]. Replacing the 153Gd source with an x-ray tube not only yields a higher photon flux and therefore faster scans with less statistical uncertainty in the measured data but also eliminates the issues of security and periodic replacement of the radioactive source. However, adapting x-ray tubes to emit x-rays approximating the two well-­ defined photopeaks of 153Gd, 44 and 100 keV, is technically challenging. An x-ray tube basically consists of a filament (cathode) and a target (anode) within an evacuated glass housing (Fig.  45.2a) [13]. When the filament is energized by an electrical current (i.e., the tube current, expressed in units of milliamperes (mA)), it emits electrons. The emitted electrons are accelerated toward the target by a potential difference between the filament and target (i.e., the tube voltage, expressed in units of kilovolts (kV)). Since the tube voltage is provided by a rectified alternating current (AC) power supply, it is not perfectly constant but varies slightly in a cyclical fashion and is characterized by its peak (p) kV value, that is, its kVp. As the accelerated electrons strike the target and interact with the atoms therein, most of the deposited electron energy is ultimately dissipated as heat. However, a small fraction of the incident electrons interacts with the atomic nuclei in the target, producing bremsstrahlung (“brake radiation”) x-rays having a range, or spectrum, of energies (in kilo-electron volts, keV) from 0 to the kVp value. In addition, some of the incident electrons will eject k-shell orbital electrons from the atoms comprising the target, resulting in the production of characteristic k-shell x-rays; these appear as sharp peaks superimposed on the continuous bremsstrahlung energy spectrum (Fig.  45.2b). The beam exiting the x-ray tube is filtered, preferentially eliminating the lower-­energy x-rays; these less-penetrating x-rays are more likely be absorbed within the patient and thus increase the patient’s radiation dose without contributing to the imaging signal. Current DEXA systems use standard tungsten-anode x-ray tubes with focal spot sizes on the order of 0.5–1 mm2 in area. In current clinical practice, DEXA uses one of two approaches to generate a dual-energy spectrum from an x-ray tube [4, 6, 14]. One involves pulses (~10 ms in duration) of low- and high-kV voltages applied to the x-ray tube and corresponding filtration during alternate half-pulses of the power supply: 70  kVp and 4-mm aluminum filtration and 140 kVp with an additional “pre-hardening” 3-mm copper filtration. The resulting low- and high-energy spectra are then measured separately. The filters, voltage switching, and detector are electronically and mechanically synchronized

45  Technical Aspects of Dual-Energy X-Ray Absorptiometry and Quantitative Computed Tomography for Assessment of Bone… Voltage kilovolts, kV

a

b UNFILTERED α1 (IN VACUUM)

Target (Focal spot)

Electrons

X-rays Filter

Filament

Cathode (-)

Current milliamperes. mA

Bremsstrahlung + Characteristic (K) x-rays

K- CHARACTERISTIC RADIATION

α2

INTENSITY

Anode (+)

1089

β1

β2

BREMSSTRAHLUNG 0

50

100

Corresponds to a tube voltage of 150 kVp MAX PHOTON ENERGY 150

200

PHOTON ENERGY (keV)

Fig. 45.2 (a) General design and operating principle of an x-ray. See the text for details. (b) Energy spectrum of the photons emitted from an x-ray tube: intensity (expressed, e.g., as the number of x-rays/cm2/s) versus energy. The dashed curve represents the energy spectrum prior to any filtration of the x-rays emitted from the target and the solid curve the energy spectrum with filtration. Superimposed on the continuous

bremsstrahlung energy spectrum on the discrete characteristic k-shell x-rays emitted from the atoms comprising the target. By design, the filtration absorbs the lower-energy, less-penetrating x-rays emitted from the target, producing a high-energy, more-penetrating x-ray beam suitable for imaging or absorptiometry (adapted from [13])

to sequentially acquire the low- and high-energy information for each position of the x-ray gantry. The other approach applies a constant voltage to the x-ray source and uses a K-edge filter to separate the energy spectrum into two narrow energy bands. With a tube voltage of 100 kVp, one of several rare earth filters is used between the x-ray tube and the patient, depending on the manufacturer; these include cerium (atomic number, Z, 58; K absorption edge, 40.4 keV) and samarium (Z, 62; K absorption edge, 46.8 keV). In these systems, both high- and low-energy x-rays are intermixed, and an energy-discriminating detector with a dual-channel analyzer counts the resulting photons in the respective energy bands. So-called first-generation DEXA scanners use a highly collimated pencil x-ray beam coupled to a single detector, while second-generation scanners use a fan beam coupled to an opposed linear array of detectors (Fig. 45.3) [4, 6, 8, 14]. The pencil-beam scanner translates the x-ray tube and opposed detector in a two-dimensional raster pattern transversely across and then longitudinally along the patient. Pencil-beam systems do not magnify the body part being imaged and thus do not introduce any geometric distortion. Fan-beam scanners use a slit collimator to generate an x-ray beam that diverges in the transverse direction with an opposed linear array of solid-state detectors, so that transmission data can be acquired with a single longitudinal sweep along the patient. The fan-beam systems use higher-­energy photons and a higher x-ray flux, thus producing a better-resolution image faster than the older pencil-beam systems. For example, the lumbar

spine can be scanned in 30 s with a fan-beam scanner, as compared with the 3–10 min with a pencil-beam scanner. The improved image resolution with the fan-beam systems is achieved at the cost of a higher radiation dose to the patient. Additionally, the fan-beam geometry leads to magnification and thus some distortion of the image in the transverse direction. The degree of magnification will depend on the distance of the bone or tissue from the source: the closer the body part is to the source, the greater the magnification.

Key Learning Points

• The original DEXA systems utilized a radioactive gadolinium-153 (153Gd) source whose energy spectrum includes photopeaks at 44 and 100 keV. • Replacing the 153Gd source with an x-ray tube not only yields a higher photon flux and therefore faster scans with less statistical uncertainty but also eliminates the need for periodic replacement of the radioactive source. • X-ray tubes produce bremsstrahlung x-rays by high-energy electrons impinging on and being stopped by a high-atomic number (i.e., metal) target. • DEXA systems have utilized various x-ray beam geometries, including scanning pencil beams and fan beams.

1090

P. Zanzonico

Fig. 45.3  Schematic diagram of the x-ray beam and the x-ray source and detector motion for pencil-beam (left top panel) and for fan-beam (right top panel) DEXA systems. The bottom panel shows a typical DEXA system with the patient in position for a study

Transverse dimension of patient

Longitudinal dimension of patient

Linear array of X-ray detectors Single X-ray detector

Pencil Beam

Fan-Beam

X-ray detector (s)

X-ray source

45.5 T  he DEXA Algorithm for Measurement of Bone Density and Body Composition [4]

where μm = the mass attenuation coefficient (in cm2/g) of the absorber

The attenuation, or stopping, of x-rays and of electromagnetic radiations generally as they pass through a stopping medium of uniform composition is, of course, described by the following exponential formula:





I ( t ) = Ioe − µt

(45.1)

where Io  =  the intensity of x-rays (expressed, e.g., as the number of x-rays/cm2/s) incident on an absorber of linear attenuation coefficient μ (in cm) and thickness t (in cm) and I(t) = the intensity of x-rays transmitted through (i.e., exiting) the absorber. Equation (45.1) can be reformulated as follows:

I ( t ) = Ioe − µmσ

(45.2)

=

µ ρ

(45.3)

ρ =  the mass density (in g/cm3) of the absorber and σ  =  the areal density (in g/cm2), that is, the mass per unit cross-sectional area of the absorber. For a nonhomogeneous absorber composed of a mixture of soft tissue (s) and bone (b), Eq. (45.2) becomes:

I ( t ) = Io e 

− ( µm )s σ s

+e

− ( µ m )b σ b

 

(45.4)

where (μm)s, (μm)b  =  the mass attenuation coefficient (in cm2/g) of the soft tissue and bone, respectively, and σs, σb = the areal density (in g/cm2) of the soft tissue and bone, respectively. Equation (45.4) can be written for each of two monochromatic (i.e., monoenergetic) x-ray beams of low (L) and high

45  Technical Aspects of Dual-Energy X-Ray Absorptiometry and Quantitative Computed Tomography for Assessment of Bone…

(H) energies, approximated by the x-ray beams used in DEXA, as follows: I ( t ) = Io L e  L



L

− ( µm )s σ s

+e

L

− ( µ m )b σ b

 

1.5

100% lean 20% fat

(45.5) 1.4

40% fat

and (45.6)

100% fat

where the superscripts, “L” and “H,” indicate the corresponding quantities for the low- and high-energy x-ray beams, respectively. Simultaneously solving the two Equations (45.5) and (45.6), for the bone areal density σb, yields Eq. (45.7):

( µ m )s  I ( t ) H   I ( t ) L  ln  H  − ln  L  H ( µm )s  Io   Io  σb = L µ L H ( m )s µ − µ b b ( m) ( m) H ( µ m )s

80% fat

1.3

1.2 0

1



Defining the ratio, or “R value,” for soft tissue, Rs, as:

( µ m )s , H ( µ m )s L

Rs ≡

(45.8)

Equation (45.7) can be rewritten as follows:



 I ( t )H   I ( t )L   Rs ln − ln  L  Io H   Io    σb = H L ( µm )b − ( µm )b Rs

   



4

Fig. 45.4  Calibration curves for body composition measurements by DEXA. The solid and dashed horizontal lines yield the percent of fat and, by difference, the percent of muscle (lean tissue) in bone-free soft tissue projected onto a pixel of the DEXA image; these percentages depend on the soft-tissue R value, RS, and the transmission of the high-­

 I ( t )H energy x-rays, − ln   Io H 

  , as well as the areal density (in g/cm2) of  

the soft tissue, σS. As an example, if the measured values of RS,

 I ( t )H − ln  H  Io 

  , and σS in a DEXA pixel are 1.35, 1.0, and 5.0 g/cm2,  

respectively, the soft tissue projected onto that pixel is composed of 60% fat and 40% (=100–60%) lean tissue, as indicated by the “X” in the figure. Adapted from reference [4]

(45.9)

The R value for soft tissue, Rs, can be determined from the tissue surrounding bone that does not contain bone (i.e., where σb = 0) using the following equation:  I ( t )L   ln   Io L   Rs   I ( t )H   ln   Io H   

3

σs

I(t)H - In IoH

L

(45.7)

2

50 g/cm2

 

60% fat

15 g/cm2

+e

H

− ( µ m )b σ b

10 g/cm2

H

− ( µm )s σ s

5 g/cm2



H

Rs

I ( t ) = Io e  H

1091

(45.10)

All of the quantities in Eqs. (45.9) and (45.10) are either measurable by DEXA or for the mass attenuation coefficients in bone, (μm)bL and (μm)bH, known. The bone areal density, σb, is thus unambiguously determined. Equations (45.5) and (45.6) represent two equations in two unknowns. DEXA, therefore, can only determine the amounts of two tissue components (such as bone and soft

tissue) at a time. An equation analogous to Eq. (45.7) can be derived for soft tissue and the soft-tissue areal density, σs, above that of bone determined. The composition of the soft tissue in pixels corresponding to tissue containing no bone— in terms of relative amounts of fat and muscle—remains to be determined. For such tissue, the soft-tissue R value, RS,  I ( t )H  and the transmission of the high-energy x-rays, − ln  H   Io    , are related by calibration curves corresponding to the percent of fat and of muscle comprising the tissue, as shown in  I ( t )H  Fig. 45.4 [4]. Once RS and − ln  H  are measured, the  Io    percent fat and, by difference, the percent muscle in the soft tissue corresponding to each pixel can be extracted from the calibration curves.

1092 Fig. 45.5 (a) Patient positioning for a posteroanterior (PA) DEXA scan of the lumbar spine. (b) Generic report for a DEXA spine scan. The scan of the lumbar spine includes the L1–L4 vertebral bodies, from which a mean BMD of these four vertebrae is obtained. To assess the lumbar spine, at least two evaluable vertebrae are required; vertebrae with fracture or focal lesions are excluded from this analysis. The graphical display includes a marker indicating the bone “health” of the patient based on the T-score for his or her lumbar spine (from reference [6])

P. Zanzonico

a

BMD (g/m2) 1.42 Normal 1.30

b

1.18 0 1.06 -1 -2 0.94 Osteopenia -3 0.82 0.70 -4 Osteoporosis -5 0.58 20 30 40 50 60 70 80 90 100 Age (years)

Region L1 L2 L3 L4 L1-L4 L2-L4

Key Learning Points

• The stopping, or attenuation, of x-rays passing through matter is described by an exponential function. • Derivation of BMD and body composition is based on the differential attenuation of the two different x-ray energies produced in DEXA systems. • Manufacturer-provided calibration curves corresponding to the percent of fat and of muscle comprising tissue are required for body composition measurements.

45.6 Analysis of DEXA Studies For adults, the DEXA examination should include the lumbar spine (Fig. 45.5) and the proximal femur (Fig. 45.6); the forearm can also be examined when the hip or spine cannot be examined by DEXA.  In children, only measurement of

T-S core 2 1

BMD (g/m2) 0.987 0.930 1.053 0.961 0.983 0.982

Young-Adult Young-Adult (%) T-Score 87 -1.2 78 -2.2 88 -1.2 80 -2.0 83 -1.6 82 -1.8

the lumbar spine is performed. BMD is evaluated in regions of interest (ROIs) contouring the respective measurement sites, as illustrated in Figs.  45.5 and 45.6. These ROIs are automatically positioned by the software provided with the DEXA system but can be adjusted by the operator if required for proper positioning. The final BMD result is the lowest value between whichever two sites are examined. In addition to site-specific BMD values, the standard DEXA analysis yields the so-called T-score, the difference in number of standard deviations between the mean BMD value of the patient and that of a young adult reference population of the same gender. The criteria for osteopenia1 and osteoporosis are a T-score between −1.0 and −2.5 and a T-score less than −2.5, respectively. A “Z-score” may be included as well; the Z-score is the difference in number of standard deviations between the mean BMD value of the patient and the mean of a reference population of the same race, sex, and age. Osteopenia is a condition in which the BMD is lower than normal BMD but not low enough to be classified as osteoporosis. 1 

45  Technical Aspects of Dual-Energy X-Ray Absorptiometry and Quantitative Computed Tomography for Assessment of Bone… Fig. 45.6 (a) Patient positioning for DEXA scan of the left hip. (b) Generic report for a DEXA hip scan. The scan of the hip includes the femoral neck (“Neck”), Ward’s triangle (“Ward’s”; not labeled), trochanter (“Troc.”, not labeled), and diaphysis (not labeled) as well as the total hip. In this report, the Z-score and the T-score are provided. The curves in the graph are age-dependent plots of the mean normal BMD + and −1 standard deviation about the mean (from reference [6])

1093

a

b

Femoral Neck

BMD (g/m2) T-Score 2 1.22 Normal 1 1.10 0 0.98 -1 1.86 -2 0.74 Osteopenia -3 0.62 -4 0.50 Osteoporosis -5 0.38 20 30 40 50 60 70 80 90 100 Age (years)

Total Femur Region Neck Ward’s Troc. Diaphysis Total

Key Learning Points

• A typical DEXA study includes the lumbar spine and the proximal femur. • The standard DEXA analysis yields the T-score, the difference in number of standard deviations between the mean BMD value of the patient and that of a young adult reference population of the same gender. • The Z-score is the difference in number of standard deviations between the mean BMD value of the patient and the mean of a reference population of the same race, gender, and age.

BMD (g/m2) 0.992 0.811 0.873 1.331 1.095

Young-Adult (%) T-score 101 0.1 89 -0.8 111 0.8 110 0.8

Adj to age (%) Z-score 112 0.9 110 0.6 118 1.2 117 1.3

µ − µ water , where μ and μwater are the linear attenuation coefµ water

ficients of the tissue and of water, respectively. On the HU scale, water has a value of 0, air −1000, and (compact) bone nearly 1000. The HU values for most soft tissues are in a narrow range of −100 to 100; the differential attenuation among soft tissues is thus so limited that it is difficult to distinguish different soft tissues (at least without administration of radiographic contrast agents). In modern multi-slice CT (MSCT) scanners, the x-rays emanate as a diverging fan beam from the x-ray tube, pass through the subject, and strike an opposed bank of small-area detectors radiation detector; the transmission image thus acquired is called a “projection image.” The x-ray tube assembly rotates about the longitudinal axis of the subject, and another projection image is acquired. At the same time, the patient table is translated through the imaging gantry, 45.7 Quantitative Computed Tomography with the acquisition of projection image data in the longitu(QCT) dinal dimension of the patient; the length of travel of the Computed tomography (CT) is, of course, a well-established patient table may be sufficiently large to image the patient’s radiographic imaging modality which yields three-­ whole body. There are multiple (up to 256) contiguous banks dimensional “maps” of the relative attenuation of x-rays by of detectors in MSCT scanners, so that data required to different tissues [15]. CT images are commonly parameter- reconstruct images of several transverse sections (or slices) ized in terms of Hounsfield units (HUs), defined as 1000 are acquired simultaneously. The complete set of transmis-

1094 Fig. 45.7 (a) 2D (i.e., single-slice) and (b) 3D (i.e., volumetric) QCT images with the calibration phantom in place for converting HU values to BMD values (from reference [17])

P. Zanzonico

a

b

Calibration phantom

sion images acquired around the subject is mathematically reconstructed to yield a set of contiguous transverse images with excellent spatial resolution (of the order of 1 mm or better); these images can be re-sorted to yield coronal and sagittal images as well. The x-ray stopping power of a medium such as tissue is determined by its electron density (i.e., the number of electrons per cubic centimeter); this, in turn, is determined by the effective atomic number and the mass density of the medium. As a result, the bone (with a relatively high effective atomic number and electron density due to its calcium content), lung (with a low mass density and electron density due to its air content), and soft tissue can be readily distinguished from one another on CT scans. Different soft tissues, however, are difficult to resolve from one another, as noted. Intravenously administered iodinated contrast agents are therefore often used to temporarily increase the electron density of different tissues, with the resulting radiographic enhancement of different tissues related to their differential blood (and therefore contrast agent) content. QCT measures BMD (in g/cm3) using a standard CT system with a calibration phantom scanned with the patient in order to convert HUs of the CT image to BMD values [16] (Fig.  45.7). An advantage of QCT over DEXA is thus the ability to yield volumetric rather than areal BMD values (i.e., g/cm3 rather than g/cm2 values). QCT calibration phantoms contain various concentrations of material with similar x-ray attenuation characteristics to those of bone. From the regression equation relating HU to concentration of the calibration material in the phantom, the HU of bone can be converted to BMD.  Originally, these phantoms were liquid-filled, with compartments containing different concentrations of, for example, potassium phosphate. However, because of potential leakage and introduction of air bubbles into the liquid,

Calibration phantom

calibration phantoms composed of solid materials (such hydroxyapatite) were subsequently developed (Fig.  45.7) and are now used with specific analysis software packages (e.g., Mindways Software, Inc. Austin, TX; Image Analysis, Inc., Columbia, KY; Siemens, Erlangen, Germany). The results from different types of calibration phantoms are not interchangeable, however, unless a cross-calibration calculation can be performed. Phantoms, such as the European Spine Phantom [18], have been developed for such cross-­ calibration as well as for QCT quality assurance. In longitudinal studies, the same calibration phantom must be used. Originally, so-called two-dimensional (2D) QCT used individual, thick CT slice images through each of multiple vertebrae (Fig. 45.8a); this generally required tilting the CT scanner gantry to align each transverse slice with each vertebra. Today, three-dimensional (3D) QCT exploits the ability of spiral MSCT scanners to rapidly acquire multiple slices to construct 3D images (Fig.  45.8b), substantially reducing image acquisition time, improving reproducibility, and enabling QCT bone density analysis of the hip as well as vertebrae.

Key Learning Points

• Modern CT scanners are multi-slice, spiral scanning devices. • QCT calibration phantoms contain various concentrations of material with similar x-ray attenuation characteristics to those of bone. • From the regression equation relating HU to concentration of the calibration material in the phantom, the HU of bone can be converted to BMD.

45  Technical Aspects of Dual-Energy X-Ray Absorptiometry and Quantitative Computed Tomography for Assessment of Bone…

a

1095

b

Fig. 45.8 (a) For 2D (i.e., single-slice) QCT, three lumbar vertebrae are scanned (generally L1–L3). A lateral scout image is first obtained, and the slices to be imaged (8–10 mm in thickness) are identified as the midplane of each selected vertebra parallel to the vertebral end plates (top panel). On the transverse slice thus selected, either an oval (middle panel) or “PacMan” (bottom panel) ROI is selected to encompass as much of the trabecular bone and exclude as much of the cortical bone

as possible. (b) For 3D (i.e., volumetric) QCT, multi-slice sections through two or three lumbar vertebrae (L1, L2, and/or L3) are selected on a posteroanterior (PA) scout image (top panel). On the transverse section thus selected, either an elliptical cylinder or “PacMan” (osteo) trabecular bone volume of interest (VOI) or an integral VOI (trabecular plus cortical bone) is analyzed (adapted from [6])

45.8 Q  uality Assurance and Calibration of DEXA and QCT Instrumentation

mentation in particular. This is especially true when imaging instrumentation is used to generate numerical results (as in the case of DEXA and QCT), since the output of such instrumentation may be quantitatively unreliable even when the imaging data appear satisfactory. Regular scanning of a suit-

Regular performance of quality assurance (QA) procedures is required for instrumentation in general and imaging instru-

1096

P. Zanzonico

able phantom and analysis of the phantom image data are essential components of the quality assurance of DEXA and QCT instrumentation. In the case of but independent of QCT, the routine quality assurance procedures for CT scanners are, of course, required, and detailed protocols for evaluation of numerous CT performance parameters are well established. For example, daily testing of a CT scanner includes ­evaluation of tomographic uniformity, the accuracy of the CT number of water, and image “noise” (i.e., statistical uncertainty), based on scanning a water-filled cylinder phantom and using a clinically routine set of scan parameters [19]. For QCT as well as for DEXA systems, daily or weekly scanning of a

a

suitable phantom and analysis of the phantom image data are essential components of the quality assurance program for measurement of BMD. In general, such phantoms should be constructed of water-equivalent epoxy resin with inserts composed of hydroxyapatite so that the attenuation coefficients approximate as closely as possible those of soft tissue and bone. The inserts should span a range of hydroxyapatite concentrations so that the linearity of the BMD scale can be verified. The European Spine Phantom (Fig.  45.9), for example, has three simulated vertebrae corresponding to ­ BMD area densities of 0.5, 1.0, and 1.5 g/cm2 and volume densities of 0.05, 0.1, and 0.2 g/cm3, respectively [18].

b

Anterior/ Posterior view

c

Lateral view

d

Fig. 45.9  European Spine Phantom. (a) Photograph of the phantom. (b) Longitudinal-view diagrams of the spine inserts in the phantom. (c) Transverse-section diagrams and (d) CT images through the inserts for vertebrae L1 (top), L2 (middle), and L3 (bottom) (from reference [20])

45  Technical Aspects of Dual-Energy X-Ray Absorptiometry and Quantitative Computed Tomography for Assessment of Bone…

Key Learning Points

• Regular scanning of a suitable phantom and analysis of the phantom image data are essential components of the quality assurance of DEXA and QCT instrumentation. • Inserts in such a phantom should span a range of hydroxyapatite concentrations to verify the linearity of the BMD scale. • The widely used European Spine Phantom has three simulated vertebrae corresponding to BMD area densities of 0.5, 1.0, and 1.5 g/cm2 and volume densities of 0.05, 0.1, and 0.2 g/cm3, respectively.

1097

diagnostic CT scanning and therefore can be performed with low-dose techniques. Two-dimensional QCT of the lumbar spine performed using a tube voltage of 80  kVp and time current of 120 mAs delivers an effective dose (ED) of less than 0.2 mSv (less than 20 mrem); using a tube voltage of 120 kVp and time current of 200 mAs results in an ED of ~1 mSv (~100 mrem). Three-dimensional QCT of the spine (100 kVp, 100 mAs) and femur (120 kVp, 200 mAs) delivers EDs of ~1.5 mSv (~150 mrem) and ~3 mSv (~300 mrem), respectively. For the first-generation pencil-beam DEXA devices, the ED is very low—only ~0.001 mSv (~0.1 mrem) both for a spine and a femur study. However, the patient doses, while still very low, are considerably higher for the fan-beam devices, especially for children and adolescents (Fig. 45.11).

45.9 R  adiation Dosimetry of DEXA and QCT Key Learning Points

QCT and, in particular, DEXA are dosimetrically favorable procedures, delivering considerably lower radiation doses than most other radiological studies (Fig. 45.10) [4, 21–23]. QCT of the spine does not require the same image quality as 100000

Effective Dose (µSv)

a

10000 Effective Dose (µSv)

• QCT and, in particular, DEXA are dosimetrically favorable procedures, delivering considerably lower radiation doses than most other radiological studies.

1000

60 50

30 20 10 0

100

Array Fast Express

40

5

b

10 15 Age at time of DXA study (yr)

20

ll Xray Lum bar Spin e Xray 1-ye ar b ackg roun d Rou tine Bod y CT scan

Sku

ackg roun d ek b

X-ra y

1-we

Che st

DEX

AS can 1-da y ba ckgr ound

1

Common Radiation Exposures

Fig. 45.10  Comparative EDs (in μSv) for DEXA scan (i.e., a pencil-­ beam spine or femur scan) versus other radiation exposures, including those from common medical procedures as well as natural background radiation (From reference [4])

Effective Dose (µSv)

35 10

30 Array Fast Express

25 20 15 10 5 0

5

15 10 Age at time of DXA study (yr)

20

Fig. 45.11  EDs (in μSv) for a DEXA scan of the (a) spine and (b) hip as a function of patient age. Patient dose was estimated for a Hologic DEXA system using three proprietary imaging modes: array mode (60-s acquisition time), fast mode (30-s acquisition time), and express mode (10-s acquisition time) (from [21] as adapted from [22])

1098

• QCT studies deliver patient doses of well under ~1  mSv (~100 mrem) to ~3  mSv (~300  mrem), depending on the details of the study. • For pencil-beam DEXA devices, the ED is very low  – only ~0.001  mSv (~0.1  mrem) both for a spine and a femur study. However, the patient doses, while still very low, are considerably higher for the fan-beam devices.

45.10 Concluding Remarks DEXA and QCT are mature, low-dose x-ray imaging modalities for noninvasive measurement of BMD and body composition widely used in routine clinical practice. These techniques are accurate predictors of fracture risk and have been widely applied in the clinical management of osteoporosis and spontaneous and iatrogenic bone-wasting conditions and metabolic diseases.

References 1. Dempster DW, Shane E, Horbert W, et al. A simple method for correlative light and scanning electron microscopy of human iliac crest bone biopsies: qualitative observations in normal and osteoporotic subjects. J Bone Miner Res. 1986;1:15–21. 2. WHO. Prevention and management of osteoporosis. World Health Organ Tech Rep Ser. 2003;921:1–164. 3. US Department of Health and Human Services. Bone health and osteoporosis: a report of the surgeon general. Rockville, MD: US Department of Health and Human Services, Public Health Service, Office of the Surgeon General; 2004. p. 1–404. 4. IAEA. Dual energy x ray absorptiometry for bone mineral density and body composition assessment, in IAEA human health series. Vienna: International Atomic Energy Agency (IAEA); 2013. 5. Sabin MA, Blake GM, MacLaughlin-Black SM, et al. The accuracy of volumetric bone density measurements in dual x-ray absorptiometry. Calcif Tissue Int. 1995;56:210–4.

P. Zanzonico 6. Lorente Ramos RM, Azpeitia Arman J, Arevalo Galeano N, et al. Dual energy X-ray absorptimetry: fundamentals, methodology, and clinical applications. Radiologia. 2012;54:410–23. 7. Lorente-Ramos R, Azpeitia-Arman J, Munoz-Hernandez A, et al. Dual-energy x-ray absorptiometry in the diagnosis of osteoporosis: a practical guide. Am J Roentgenol. 2011;196:897–904. 8. Crabtree NJ, Leonard MB, Zemel BS. Dual-energy x-ray absorptiometry. In: Sawyer AJ, Bachrach LK, Fung EB, editors. Current clinical practice: bone densitometry in growing patients: guidelines for clinical practice. Totowa, NJ: Humana; 2007. 9. Chung KJ. Bone densitometry. Semin Nucl Med. 2011;41:220–8. 10. Wilson CR. Essentials of bone densitometry for the medical physicist. https://www.aapm.org/meetings/03AM/pdf/9873-13152.pdf. Accessed 20 Nov 2017. 11. Cameron JR, Sorenson J. Measurement of bone mineral in vivo: an improved method. Science. 1963;142:230–2. 12. Thorson LM, Wahner HW.  Single- and dual-photon absorptiometry techniques for bone mineral analysis. J Nucl Med Technol. 1986;14:163–71. 13. Christensen TE, Nunnally J. An introduction to the physics of diagnostic radiology. Philadelphia, PA: Lea & Febiger; 1973. 14. Coursey CA, Nelson RC, Boll DT, et al. Dual-energy multidetector CT: how does it work, what can it tell us, and when can we use it in abdominopelvic imaging? Radiographics. 2010;30:1037–55. 15. Bushberg JT, Seibert JA, Leidholdt EM Jr, et al. The essential physics of medical imaging. Philadelphia, PA: Lippincott, Williams & Wilkins; 2012. 16. Adams JE.  Quantitative computed tomography. Eur J Radiol. 2009;71:415–24. 17. Quantitative computed tomography. https://en.wikipedia.org/wiki/ Quantitative_computed_tomography. Accessed 21 Nov 2017. 18. Kalender WA, Felsenberg D, Genant HK, et  al. The European Spine Phantom – a tool for standardization and quality control in spinal bone mineral measurements by DXA and QCT. Eur J Radiol. 1995;20:83–92. 19. Zanzonico P. Routine quality control of clinical nuclear medicine instrumentation: a brief review. J Nucl Med. 2008;49:1114–31. 20. European spine phantom. http://www.qrm.de/content/products/ bonedensity/esp.htm. Accessed 21 Nov 2017. 21. Damilakis J, Adams JE, Guglielmi G, et al. Radiation exposure in X-ray-based imaging techniques used in osteoporosis. Eur Radiol. 2010;20:2707–14. 22. Blake GM, Naeem M, Boutros M. Comparison of effective dose to children and adults from dual X-ray absorptiometry examinations. Bone. 2006;38:935–42. 23. Mettler FA Jr, Huda W, Yoshizumi TT, et  al. Effective doses in radiology and diagnostic nuclear medicine: a catalog. Radiology. 2008;248:254–63.

Current Practical Guidelines for the Most Common Nuclear Medicine Procedures

46

Irene Marini, Onelio Geatti, and H. William Strauss

Contents 46.1      L  ymphoscintigraphy for Radio-­Guided Sentinel Lymph Node Biopsy in Patients with Breast Cancer 46.1.1   Indications 46.1.2   Procedure 46.1.3   Precautions 46.1.4   Radiopharmaceuticals 46.1.5   Volume and Activity 46.1.6   Modality of Injection 46.1.7   Image Acquisition 46.1.8   Radioguided Surgery 46.1.9   Interpretation Criteria 46.1.10  Reporting

 1102  1103  1103  1103  1103  1103  1103  1103  1103  1104  1104

46.2      L  ymphoscintigraphy for Radioguided Sentinel Lymph Node Biopsy in Patients with Cutaneous Melanoma 46.2.1   Potential Contraindications 46.2.2   Procedure 46.2.3   Radiopharmaceuticals 46.2.4   Administration of Radiopharmaceutical 46.2.5   Image Acquisition 46.2.6   Dynamic Imaging 46.2.7   Early Static Images 46.2.8   Delayed Static Images 46.2.9   SPECT/CT 46.2.10  Image Interpretation and Report 46.2.11  Pitfalls 46.2.12  Sources of False-Negative Interpretation of Images 46.2.13  Procedure in the Operating Room 46.2.14  Gamma Probe 46.2.15  Possible Causes of False-Negative Results

 1104  1104  1104  1104  1104  1105  1105  1105  1105  1105  1105  1105  1106  1106  1106  1106

46.3      [18F]FDG PET/CT in Oncology 46.3.1   Common Clinical Indications for [18F]FDG PET/CT in Malignant Diseases 46.3.2   Procedure/Specification of the [18F]FDG PET/CT Scan

 1106  1106  1107

I. Marini Nuclear Medicine Institute, Università Cattolica del Sacro Cuore and Nuclear Medicine Center, Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy O. Geatti (*) Nuclear Medicine Center, University Hospital of Udine, Udine, Italy H. W. Strauss Molecular Imaging and Therapy Service, Memorial Sloan Kettering Cancer Center, New York, NY, USA © Springer Nature Switzerland AG 2019 D. Volterrani et al. (eds.), Nuclear Medicine Textbook, https://doi.org/10.1007/978-3-319-95564-3_46

1099

1100

I. Marini et al. 46.3.3   Instructions to Patients 46.3.4   Image Acquisition 46.3.5   CT Protocols for [18F]FDG PET/CT Studies

 1107  1107  1107

46.4      Perfusion Brain SPECT 46.4.1   Common Clinical Indication for a Perfusion Brain SPECT Study 46.4.2   Instructions to Patients 46.4.3   Practical Instructions About Tracer Injection and Uptake Period 46.4.4   Image Acquisition 46.4.5   Reconstruction and Analysis

 1107  1108  1108  1108  1108  1108

46.5      Brain SPECT with 123I-Labelled Dopamine Transporter Ligands 46.5.1   Common Clinical Indication for a Dopamine Receptor Brain SPECT Study 46.5.2   Instructions to Patients 46.5.3   Practical Instructions About Tracer Injection and Uptake Period 46.5.4   Image Acquisition 46.5.5   Reconstruction and Analysis

 1108  1109  1109  1109  1109  1109

46.6      SPECT and PET with Dopamine D2 Receptor Ligands 46.6.1   Most Common Indication 46.6.2   Less Common Indications 46.6.3   Contraindications 46.6.4   Procedure 46.6.5   Timing of Injection 46.6.6   Administered Activity to Adults 46.6.7   Data Acquisition 46.6.8   Positioning 46.6.9   SPECT 46.6.10  Acquisition Parameters 46.6.11  PET 46.6.12  Reconstruction 46.6.13  Comparative Evaluation 46.6.14  Interpretation Criteria

 1109  1109  1109  1110  1110  1110  1110  1110  1110  1110  1110  1110  1110  1111  1111

46.7      Brain PET with [18F]FDG 46.7.1   Common Indications 46.7.2   Information to Be Recorded Before the Procedure 46.7.3   Procedure 46.7.4   Interpretation

 1111  1111  1111  1111  1111

46.8      Imaging of Brain Tumours with Amino Acid Tracers 46.8.1   Indications 46.8.2   Contraindications 46.8.3   Before the Procedure 46.8.4   Procedure 46.8.5   [11C]MET or 18F-FET PET 46.8.6   123I-IMT SPECT 46.8.7   SPECT Reconstruction 46.8.8   Interpretation

 1112  1112  1112  1112  1112  1112  1112  1113  1113

46.9      Myocardial Perfusion Scintigraphy 46.9.1   Food and Medication Before the Test 46.9.2   Radiopharmaceuticals 46.9.3   Activities to Be Injected 46.9.4   Stress Tests 46.9.5   Before a Stress Study 46.9.6   Exercise Stress Testing Procedure 46.9.7   Equipment and Protocols 46.9.8   Absolute Contraindications to Dynamic Exercise 46.9.9   Relative Contraindications to Dynamic Exercise 46.9.10  Absolute Indications for Early Termination of Exercise 46.9.11  Relative Indications for Early Termination of Exercise 46.9.12  Vasodilator Stress Testing with Adenosine, Regadenoson or Dipyridamole 46.9.13  Indications 46.9.14  Absolute Contraindications

 1113  1113  1113  1114  1114  1114  1114  1114  1114  1115  1115  1115  1115  1115  1115

46  Current Practical Guidelines for the Most Common Nuclear Medicine Procedures

1101

46.9.15   Relative Contraindications for Adenosine and Dipyridamole 46.9.16   Procedure 46.9.17   Combination with Low-Level Exercise 46.9.18   Administration of Vasodilators 46.9.19   Early Termination Vasodilator Infusion 46.9.20   Dobutamine Stress Test 46.9.21   Indications 46.9.22   Contraindications 46.9.23   Contraindications to Atropine Administration During the Dobutamine Stress Test 46.9.24   Administration of Dobutamine 46.9.25   Procedure 46.9.26   Indications for Early Termination of Dobutamine Infusion 46.9.27   Imaging Instrumentation 46.9.28   Imaging Protocols 46.9.29   SPECT/CT 46.9.30   Gated Myocardial Perfusion Imaging 46.9.31   SPECT

 1116  1116  1116  1116  1116  1116  1116  1116  1116  1116  1116  1117  1117  1117  1117  1117  1118

46.10      Ventilation/Perfusion (V/P) Scintigraphy 46.10.1   Pulmonary Embolus V/P Mismatch 46.10.2   V/P Match 46.10.3   Ventilation Scintigraphy 46.10.4   Perfusion Scintigraphy 46.10.5   Imaging Protocols 46.10.6   Interpretation of the V/P Scan for the Diagnosis of Acute PE 46.10.7   Chronic PE 46.10.8   Additional Diagnostic Outcomes 46.10.9   Combining Clinical Probability with Objective Testing for Acute PE 46.10.10  Clinical Algorithm for Investigating Patients with Suspected Acute PE 46.10.11  Follow-Up

 1118  1118  1118  1118  1118  1118  1119  1119  1119  1119  1119  1120

46.11      Bone Scintigraphy with 99mTc-Labelled Bone-Seeking Agents 46.11.1   Indications 46.11.2   Patient’s Preparation 46.11.3   Tracer Administration 46.11.4   Acquisition 46.11.5   Dosimetry

 1120  1120  1120  1120  1121  1121

46.12      Scintigraphy with 99mTc-HMPAO-­Labelled Autologous Leukocytes 46.12.1   Most Common Indications 46.12.2   Precautions 46.12.3   Radiolabelling Procedure

 1121  1121  1121  1122

46.13      Thyroid Scintigraphy 46.13.1   Clinical Indications 46.13.2   Patient Preparation 46.13.3   Pertinent Information 46.13.4   Radiopharmaceuticals 46.13.5   Image Acquisition 46.13.6   Interpretation Criteria 46.13.7   Reporting

 1122  1122  1123  1123  1123  1123  1124  1124

46.14      Radiometabolic Therapy with 131I-Iodide for Hyperthyroidism 46.14.1   Treatment Options for Hyperthyroidism and Nontoxic Goitre 46.14.2   Contraindications 46.14.3   Patient Preparation 46.14.4   Precautions 46.14.5   Information and Instruction to the Patients 46.14.6   Administration 46.14.7   Radioiodine Treatment in Children 46.14.8   Side Effects After 131I-Iodide Therapy 46.14.9   Follow-Up

 1124  1124  1124  1124  1125  1125  1125  1125  1125  1126

46.15      Radioiodine Therapy with 131I-Iodide for Differentiated Thyroid Cancer (DTC) 46.15.1   Indications

 1126  1126

1102

I. Marini et al. 46.15.2   Contraindications 46.15.3   Radioiodine Activities and Administration 46.15.4   Patient Preparation 46.15.5   Precautions 46.15.6   Potential Side Effects of Therapy with 131I-Iodide for DTC

 1126  1126  1127  1127  1128

46.16      Dynamic Renal Scintigraphy 46.16.1   Indications 46.16.2   Radiopharmaceuticals Currently Used for Dynamic Renal Scintigraphy 46.16.3   Patient Preparation 46.16.4   Positioning for the Scan 46.16.5   Activity to Be Administered 46.16.6   Image Acquisition 46.16.7   Diuretic Administration 46.16.8   Image Processing and Analysis 46.16.9   Acute Obstructive Uropathy 46.16.10  Children 46.16.11  Renovascular Hypertension

 1128  1128  1128  1128  1129  1129  1129  1129  1129  1130  1130  1131

46.17      PET/CT with 68Ga-PSMA-Ligands 46.17.1   Radiopharmaceuticals 46.17.2   Potential Clinical Indications 46.17.3   Emerging Clinical Applications 46.17.4   Patient Preparation 46.17.5   Recommendations for 68Ga-PSMA-Ligand Administration 46.17.6   PET/CT Acquisition Protocol 46.17.7   Image Reconstruction 46.17.8   Definitions of Volumes of Interest

 1132  1132  1132  1133  1133  1133  1133  1133  1133

46.18      Parathyroid Scintigraphy 46.18.1   Indications 46.18.2   Imaging Agents 46.18.3   Patient’s Preparation 46.18.4   Procedure of Single-Tracer, Dual-Phase Parathyroid Scintigraphy 46.18.5   Procedure of Double-Tracer Parathyroid Scintigraphy 46.18.6   SPECT/CT Imaging

 1133  1133  1133  1134  1134  1134  1134

46.19      PET/CT with 68Ga-DOTA-Somatostatin Analogues 46.19.1   Indications 46.19.2   Procedural Indications and Patient Preparation 46.19.3   Administration and Imaging

 1135  1135  1135  1135

46.20      Therapy with 131I-MIBG 46.20.1   Indications 46.20.2   Contraindications 46.20.3   Patient Preparation 46.20.4   Administration 46.20.5   Precautions and Follow-Up 46.20.6   Side Effects

 1136  1136  1136  1136  1136  1136  1136

Further Reading

 1137

Learning Objectives

• This chapter presents a focused overview of procedure guidelines published by the European Association of Nuclear Medicine (EANM) and the Society of Nuclear Medicine and Molecular Imaging (SNMMI). • This edited presentation of the guidelines describes the clinical indications, patient preparation, preferred radiopharmaceuticals, data processing and criteria for interpretation of common clinical procedures.

46.1 L  ymphoscintigraphy for Radio-­ Guided Sentinel Lymph Node Biopsy in Patients with Breast Cancer The sentinel lymph node (SLN) is the first lymph node that drains lymph and possibly metastatic tumour cells from the site of a tumour. Preoperative lymphoscintigraphy (LS) allows localisation of the sentinel lymph node(s) to be harvested during radioguided sentinel lymph node biopsy.

46  Current Practical Guidelines for the Most Common Nuclear Medicine Procedures

46.1.1  Indications The procedure should be performed in patients when surgical resection of the primary tumour and axillary lymph node clearance is planned for patients who have biopsy-proven breast carcinoma and no palpable axillary lymph nodes, nor other evidence of tumour metastasis in lymph nodes.

46.1.2  Procedure • No special preparation is needed. • Time of last menstruation, pregnancy and lactating status must be noted.

46.1.3  Precautions Pregnancy is not an absolute contraindication. The radiation dose to the foetus from this procedure is negligible. Breastfeeding should be temporarily discontinued for 24  h after the radiopharmaceutical administration.

1103

is generally sufficient, using activities in the upper range for surgical procedures delayed until the next day. The syringe should also contain a similar volume of air to ensure that the full amount of the agent is injected. A moderately larger volume (0.5 mL) can be used in case of deep injections.

46.1.6  Modality of Injection Except for the deepest tumours, either an intradermal injection over the lesion, a subdermal injection over the lesion or a subareolar injection site is suggested. The site of injection should be gently massaged (by the patient herself) after tracer administration, or at any time during the study. If the breast lesion is not palpable, ultrasound can be used to guide the injection. In case of upper quadrant tumours, a subareolar injection can be used.

46.1.7  Image Acquisition

The SLN is generally visualised within 2 h at most, and surgery should take place within 24 h after the injection (or up to 30 h when using 99mTc-tilmanocept).

With the patient lying supine under the gamma camera (the arm abducted in the same position as it will be during surgery), two views, e.g. anterior and 45° anterior oblique or anterior and lateral views, should be obtained. To facilitate anatomic correlation, a transmission scan with a cobalt-57 source can be recorded contemporaneously. The gamma camera is generally equipped with a low-­ energy, high-resolution collimator, and planar images are acquired (for 3–5 min, with minimum matrix a 128 × 128) starting within 15  min after the injection. If necessary (in case of slow lymphatic drainage), further imaging can be performed at 2–3 h (or even up to 16–18 h). In case of lack of visualization of drainage to lymph nodes, a further aliquot (boost) of the radiocolloid can be injected, and/or SPECT/CT images can be acquired, which improves SLN detection and localisation over planar imaging. As an optional step depending on local policies, the skin projection of any SLN visualised during lymphoscintigraphy can be marked with an indelible pen, to guide the surgeon to choose a surgical incision route for optimal cosmetic results. Nevertheless, the use of intraoperative hand-held gamma probe makes this step unnecessary in most cases.

46.1.5  Volume and Activity

46.1.8  Radioguided Surgery

Small volumes should be injected (approximately 0.1– 0.2 mL per aliquot in case of subdermal injection), as large volumes of the injectate may disrupt the pattern of local lymphatic drainage. A single aliquot of 5–20 MBq (depending on time elapsed between scintigraphy and surgery) in 0.2 mL

• A hand-held gamma detection probe provided with an audio signal fed by an instantaneous count rate will guide the surgeon to quickly detect location of the SLN(s), both prior to making the surgical field and, above all, in the exposed surgical field.

46.1.4  Radiopharmaceuticals A variety of lymphoscintigraphic agents can be used, including radiocolloid and non-radiocolloid tracers. Radiocolloids with particle size ranging between 100 and 200 nm are considered the best compromise between fast lymphatic drainage and optimal retention in the SLN(s). The radiopharmaceuticals most commonly used are: • 99mTc-sulphur colloid (particles’ size, 15–5000  nm if unfiltered, 1  mm and in selected cases with tumour thickness 1 mitosis/mm2, or regression with thickness ≥1  mm or regression of more than 50–75% of the whole pigmented lesion); some institutions extend the indication also to tumours ranging 0.75–1 mm in Breslow thickness. SLNB should be scheduled after histological confirmation of the diagnosis of melanoma following a diagnostic excision of the primary lesion with a narrow margin and should be combined with wide excision. SLNB is also indicated in patients with Merkel cell carcinoma.

46.2.1  Potential Contraindications • Poor general health status, severe concurrent disease, poor compliance and ascertained systemic metastatic status. • SLNB is a safe procedure even during pregnancy. • SLNB is also suitable and accurate in children and adolescents.

46.2.2  Procedure No special preparation is necessary. The following information must be available: clinical diagnosis, prior treatment(s), histopathology report of the excisional biopsy, history about prior surgery or trauma of the region of interest, comorbidities, pregnancy/nursing and administration of radiopharmaceuticals in the few days prior to lymphoscintigraphy. It is important to perform an accurate physical examination of the affected region. All clothes and jewellery in the affected region and along the lymphatic vessels should be removed, to avoid constriction and occlusion of lymphatic channels.

46.2.3  Radiopharmaceuticals See section above on SLNB for patients with breast cancer.

46.2.4  Administration of Radiopharmaceutical • Following administration of a topical anaesthetic when using 99mTc-sulphur colloid, the radiopharmaceutical is administered as four separate intradermal injections of 1.8 mBq (0.05 mCi) in 0.1 mL (for same day surgery, or ~3-fold more if surgery is the next day). With the single-­ day protocol, lymphoscintigraphy is performed a few

46  Current Practical Guidelines for the Most Common Nuclear Medicine Procedures







• •

hours prior to surgery, whereas with the 2-day protocol lymphoscintigraphy is performed the day before surgery. Intraoperative radiotracer injection and using the intraoperative probe without prior imaging should be avoided, since lymphatic drainage in melanoma may be aberrant, delayed, or occur to more than one lymph node basin— features that will all be missed without preoperative lymphoscintigraphy. The radiopharmaceutical should be injected intradermally in four or more aliquots on ~1  cm on each side of the centre of the surgical scar from prior excision biopsy (the most common occurrence). The distance between the scar and tracer injection may be increased if the planned site of injection is indurated. When the melanoma is still present (the more rare occurrence), the radiotracer will similarly be injected close to the primary tumour. The injected volume is around 0.1–0.2 mL per aliquot (if the volume is too large, lymphatics may collapse or the wheal on the skin surface may rupture). 25 or 27  G needles mounted on tuberculin syringes are generally used for injection. The amount of radioactivity injected is up to 37  MBq, depending on the protocol adopted (single-day or 2-day protocol).

46.2.5  Image Acquisition • All possible lymph node basins should be imaged. • Use of a dual-head gamma camera with large field of view detectors is preferred. • Use of low-energy, high-resolution collimator is highly recommended. • An energy window between 15% and 20% centred on the 140-keV photo peak is recommended. • Body contouring with a transmission scan using a flood source beneath the patient’s body (when using single-­ head gamma cameras) facilitates the anatomic localisation of hot spots.

46.2.6  Dynamic Imaging The tracer is injected when the patient is lying in a supine or prone position (depending on location of the primary tumour) on the bed of the gamma camera. Dynamic acquisition should be performed for 10–20 min (1 frame/min, 64 × 64 matrix) starting after tracer injection. This early imaging acquisition can visualise the overall pattern of lymphatic drainage, enable to detect “in-transit” lymph nodes and improve SLN localisation by subsequent static imaging. The procedure is strongly recommended especially in melanomas of the hand/forearm or foot/leg.

1105

46.2.7  Early Static Images At the end of dynamic acquisition, static planar 5-min views (anteroposterior and lateral) should be acquired with a large matrix (best with a 256 × 256 matrix) over the lymph node basin(s) in which the SLN is expected, based on the prior dynamic acquisition. Early images help to discriminate true SLNs from the second-echelon nodes. In melanomas of the trunk, usually bilateral static images of the axilla, trunk and groin are necessary, or a whole-body scan from the neck to the groin can be carried out.

46.2.8  Delayed Static Images After recording the early images, it is helpful to have the patient massage the area of the injection and move the body part (e.g. ambulate if the lesion is the lower extremity, or move the arms if the lesion is in the upper extremity, etc.) during the interval between early and delayed imaging. Late 3- to 5-min anteroposterior and lateral static images are acquired (1–3 h after tracer injection) to identify all relevant SLNs.

46.2.9  SPECT/CT SPECT/CT imaging greatly improves accuracy in SLN localisation and reduces the possibility of misinterpretation of images. SPECT/CT is extremely valuable for lesions in the head and neck. It is also useful for the groin or the axilla.

46.2.10  Image Interpretation and Report Early dynamic, early static and delayed static images identify SLNs in the majority of patients. The SLN is the lymph node first visualised along a pathway of lymphatic drainage; although it is not always the lymph node closest to the injection site, it is generally the hottest node along a certain pathway of lymphatic drainage. Lymph nodes that appear on later images may also be SLNs, unless they receive lymphatic drainage from a SLN detected earlier, e.g. during the dynamic image acquisition. The report should include radiopharmaceutical used, injection technique, injected activity and volume, time points of acquisitions and orientations of images. The visualised structures and their location should be described in detail, in particular, the number and location of SLNs.

46.2.11  Pitfalls 1. Skin contamination.

1106

2. Lymphangioma or lymphatic lakes may be misinterpreted as lymph nodes. 3. Second-echelon nodes may be misinterpreted as SLNs (if no early dynamic or static images are acquired). 4. Other tissues containing radioactivity may complicate image interpretation.

46.2.12  Sources of False-Negative Interpretation of Images 1. Two or more radioactive lymph nodes that are very close one to each other may be misinterpreted as one single SLN. 2. The SLN may be masked by the much greater radioactivity remaining at the injection site. 3. Only a small amount of the radiotracer drains from the injection site. 4. No tracer drainage from the injected site can be detected. In this case, gentle massage of the injection area can be helpful, as can also be warming the area with a hot water bag or, for melanoma of arms and legs, physical exercise. When no SLN is localised, the procedure should be repeated.

46.2.13  Procedure in the Operating Room The probe is used for guidance as exposure of the overlying tissues proceeds, in order to identify the location of the SLN(s) in the surgical bed (in vivo measurement). After SLN excision, the probe is used on the resected tissue, away from the patient, to confirm that it is the hot lymph node corresponding to the SLN.  Then the surgical bed is explored again with the hand-held gamma probe and checked for possible residual activity. Local policies dictate how many lymph nodes should be harvested for histologic analysis after the SLN (e.g. based on a 10% or 20% threshold of ex vivo count rate relative to the SLN). The surgeon must also check manually the region of interest to identify residual enlarged nodes without tracer uptake. In fact, a lymph node in which the lymphatic tissue has been completely replaced by metastatic cells will not be able to concentrate the radiopharmaceutical, but must be removed for histologic analysis and completeness of staging.

46.2.14  Gamma Probe The gamma probe used should be designed for intraoperative application and be routinely checked for quality assurance.

I. Marini et al.

The probe should be placed in a sterile sleeve and should provide instantaneous and cumulative counts. Conversion of count rate into an acoustic signal with a variable pitch facilitates SLN localisation.

46.2.15  P  ossible Causes of False-Negative Results 1. Imaging of the wrong nodal basin and failure to depict all potential drainage basins. 2. Failure to visualise the afferent lymph vessel and/or to detect an SLN in an unusual location. 3. Gross metastasis in the SLN may reduce tracer accumulation. If the time interval between lymphoscintigraphy and surgery is too long and the radioactive lymph nodes can no longer be detected, the patient can be reinjected before the surgical procedure.

46.3 [18F]FDG PET/CT in Oncology Many mammalian cells utilise glucose as their major source of energy. The radiolabelled glucose analogue [18F]FDG enters the cell via a family of transmembrane glucose transporters (GLUT), in similar fashion to exogenous glucose. Many cancer cells demonstrate increased glucose utilization. The mechanism of intracellular trapping of [18F]FDG is detailed in Chap. 3 (Positron-Emitting Radiopharmaceuticals). Although PET with [18F]FDG has important roles in the investigation of the central nervous system, in imaging infection/inflammation and in cardiology, the vast majority of the patients referred to a nuclear medicine centre for [18F]FDG PET/CT are patients with suspected or ascertained malignant disease.

46.3.1  C  ommon Clinical Indications for [18F] FDG PET/CT in Malignant Diseases • Differential diagnosis between benign and malignant tissues. • Staging of newly diagnosed cancers. • Localisation of the unknown primary tumour when metastatic disease has been diagnosed or in patients presenting with a paraneoplastic syndrome. • Monitoring the efficacy of antitumour therapy. • Detection of tumour recurrences post-treatment. • Differential diagnosis of tumour recurrence versus post-­ treatment fibrosis/necrosis.

46  Current Practical Guidelines for the Most Common Nuclear Medicine Procedures

• Molecular imaging guide to radiotherapy planning, for identifying metabolic tumour volume in patients scheduled for radiotherapy. • Identification of the optimal area for biopsy, in case of large tumour masses possibly containing necrotic tissues.

46.3.2  P  rocedure/Specification of the [18F] FDG PET/CT Scan Request from the referring physician/oncologist must include at least the (presumed) diagnosis and the clinical questions to be addressed; it should also include all known information that can be helpful for an accurate image interpretation, such as previous diagnostic imaging reports, laboratory tests and present clinical status. Personal information contained in the request for the scan should also include patient’s height, body weight, fasting blood sugar (which must be assayed before [18F]FDG injection), current medications (with special attention to metformin, glucocorticoids, phenothiazines, lithium, tricyclic antidepressants, phenytoin, thiazide diuretics, isoniazid, rifampin and ephedrine), allergies, claustrophobia (if present) and serum creatinine if an i.v. contrast medium will be used.

46.3.3  Instructions to Patients Patients should be fasting for a minimum of 6 h prior to [18F] FDG injection. Intense physical exercise should be avoided for 24 h before the [18F]FDG PET/CT study to reduce uptake of [18F]FDG in skeletal muscles. During the injection and the subsequent uptake time, the patient should remain seated (or recumbent) and silent, especially if the head and neck must be explored with special attention because of a cancer located in this region; in the latter case, patients should also be advised not to chew gum for 24 h before the study, in order to avoid metabolic activation of the muscles of mastication. To minimise [18F]FDG accumulation in activated brown fat, the patient should be kept warm, starting 30–60  min before the injection, throughout the uptake period and during the examination. The blood glucose level must be measured before administering [18F]FDG.  If blood glucose is 5 million detected events.

46.4.5  Reconstruction and Analysis Data must be filtered with a low-pass filter and corrected for attenuation, using iterative reconstruction. Traditionally, evaluation of perfusion brain SPECT has been based on visual analysis searching for areas with abnormal perfusion relative to other “normally” perfused areas of grey matter. An alternative semiquantitative approach for analysis was based on selection of regions of interest for visually evaluating blood flow abnormalities, by comparison with the corresponding structures in the contralateral hemisphere and with a normal control database. More recent approaches to the analysis of perfusion brain SPECT scans use of statistical parametric mapping (or similar methods) based on comparison with reference normal control databases; this type of analysis also allows for comparison of a group of patients with a certain disease with a groups of patients with another disease or—within the same patients’ group—comparison of post-therapy scans to baseline scans.

46.5 B  rain SPECT with 123I-Labelled Dopamine Transporter Ligands The two most commonly used radiopharmaceuticals for evaluating the presynaptic dopaminergic system in patients with movement disorders are [123I]2β-carboxymethoxy-3β(4-iodophenyl) tropane ([123I]β-CIT or [123I]Iometopane, commercially available as Dopascan™/SPECT) and [123I] N-ω-fluoropropyl-2β-carbomethoxy-3β-(4-iodophenyl) nortropane ([123I]FP-CIT or [123I]Altropane, commercially available as DaTscan®).

46  Current Practical Guidelines for the Most Common Nuclear Medicine Procedures

46.5.1  C  ommon Clinical Indication for a Dopamine Receptor Brain SPECT Study • Confirmation of deficiency of functional dopaminergic receptors in the striatum in patients with Parkinsonian syndromes, as an aid to distinguish essential tremor from Parkinson’s disease, progressive supranuclear palsy and multiple system atrophy. • [123I]FP-CIT is useful also to distinguish dementia with Lewy bodies from other dementias but cannot discriminate between Parkinson’s disease, multiple system atrophy and supranuclear palsy.

46.5.2  Instructions to Patients

1109

46.5.4  Image Acquisition • Dedicated SPECT gamma cameras or multiple detector systems should be used. • Use of fan-beam collimators is optimal, although use of low-energy high-resolution (LEHR) or low-energy ultra-­ high-­ resolution (LEUHR) parallel-hole collimators is acceptable as well. • Acquisition parameters include a 128  ×  128 matrix and adequate zoom to obtain a pixel size of one-third to one-­ half of the expected resolution. • Images are acquired with 3° angular sampling with a circular or elliptic orbit around the head, for a total of >3 million detected events for [123I]FP-CIT and >1 million for [123I]β-CIT.

Patients should avoid medications or habit-forming drugs that may influence the analysis (except if the aim of the study is to assess the effect of such medication). A withdrawal period of five times the drug’s biological half-life is suggested. It is not necessary to withdraw anti-Parkinsonian medications. Thyroidal uptake of radioiodide released during metabolic degradation of the 123I-labelled ligands should be prevented by adequate pretreatment with, e.g. saturated potassium iodide solution or Lugol’s solution. Pretest information and evaluation of the patient should include:

46.5.5  Reconstruction and Analysis

• Patient’s history with focus on neurological and psychiatric disorders. • Current neurological and psychiatric status. • Recent morphological studies of the central nervous system. • Current medication and when last administered. • Patient’s ability to lie and stand still for approximately 40–60 min. • If sedation is needed, it should be given just before the SPECT acquisition.

Multiple radiopharmaceuticals are available to evaluate distribution and density in the brain of the D2 receptors, which are mainly located in the postsynaptic dopaminergic system. They include the single-photon-emitting agents 123I-IBZM, 123 I-epidepride, [11C]raclopride and the positron-emitting 18 18 agents F-fallypride and F-desmethoxyfallypride 18 ( F-DMFP).

46.5.3  P  ractical Instructions About Tracer Injection and Uptake Period The two radiopharmaceuticals are delivered ready to use and should be injected slowly i.v. within the time frame recommended by the manufacturer. The standard activity to be injected is 185 MBq for adult patients. Acquisition should start 3–6  h after injection of [123I]FP-CIT and 18–24 h after injection of [123I]β-CIT.

Data must be filtered with a low-pass filter and corrected for attenuation, using iterative reconstruction. Regions of interest in the striatum and striatal subregions are defined for comparison with reference regions without specific dopamine transporter density (usually the cerebellum); the ratios so obtained are compared with those in normal controls.

46.6 S  PECT and PET with Dopamine D2 Receptor Ligands

46.6.1  Most Common Indication • Differential diagnosis of Parkinson’s disease from other neurodegenerative Parkinsonian syndromes characterized by loss of D2 receptors, such as multiple system atrophy and progressive supranuclear palsy.

46.6.2  Less Common Indications • Assessment of D2 receptor blockade obtained by treatment with dopamine D2 antagonists.

1110

I. Marini et al.

• Huntington’s disease. • Wilson’s disease. • Pituitary adenoma.

tion, for consistency across patients as well as in the same patient for easier follow-up.

46.6.8  Positioning

46.6.3  Contraindications

The patient should be instructed to avoid movements of the head during data acquisition. If necessary, sedation can be used.

• Pregnancy. • Breastfeeding (to be interrupted for 24 h). • Non-adequate compliance.

46.6.9  SPECT

46.6.4  Procedure Drugs known to affect D2 receptor binding should be withdrawn prior to the investigation; if clinically necessary, therapy with L-DOPA can be maintained. When using 123 I-ligands, thyroidal uptake of radioiodide released during metabolic degradation of the radiopharmaceuticals must be blocked with oral potassium iodide or Lugol’s solution. Ensure that the patient is able to cooperate; if sedation is necessary, it should be given at the earliest 1  h prior to imaging. The radiopharmaceutical must be slowly injected i.v. and flushed with saline.

46.6.5  Timing of Injection [11C]Raclopride, 18F-fallypride and 18F-DMFP should be injected immediately after production. For 123I-IBZM and 123 I-epidepride, instructions supplied by the manufacturer should be followed.

Multiple detector or dedicated SPECT gamma cameras for brain imaging should be used. Fan-beam collimators are to be preferred over parallel-­ hole LEHR or LEUHR collimators.

46.6.10  Acquisition Parameters • • • •

Rotational radius: the smallest. Matrix: 128 × 128 or higher. Angular sampling: ≤3° (360° rotation). Zoom: pixel size should be one-third to one-half of the expected resolution. • Acquisition mode: step-and-shoot or continuous mode. • Total detected events: >3 million. Total scan time: depending on the imaging device. A transmission scan or CT can be used for attenuation correction.

46.6.11  PET

46.6.6  Administered Activity to Adults

Acquisition parameters: • 150–250  MBq (typically 185  MBq) of 18 F-radiopharmaceutical. • 220–370 MBq of [11C]raclopride.

I- or

123

46.6.7  Data Acquisition Time from injection to start of imaging acquisition: • 123I-IBZM: 1.5–3 h (preferably 2 h). • 123I-Epidepride: 2–3 h. • [11C]Raclopride: 30–60 min. • 18F-DMFP: 1–1.5 h. • 18F-Fallypride: 2.5–3 h. Each nuclear medicine centre is advised to adopt a fixed time delay between injection and the start of data acquisi-

• Matrix: 128 × 128 or higher. • Zoom: pixel size should be one-third to one-half of the expected resolution. • Total scan time: 60 min for [11C]raclopride and 20–30 min for 18F-fallypride. • Transmission or CT scan can be used for attenuation correction.

46.6.12  Reconstruction • SPECT: iterative reconstruction. • PET: a final image resolution of 4–6  mm FWHM typically provides images of adequate resolution and noise. • Attenuation correction is mandatory for both SPECT and PET.

46  Current Practical Guidelines for the Most Common Nuclear Medicine Procedures

46.6.13  Comparative Evaluation Region-of-interest analysis is used for comparison and to assess nonspecific binding (e.g. frontal cortex, occipital cortex, cerebellum).

46.6.14  Interpretation Criteria • Visual interpretation: only to assist quantitative evaluation: • Quantification. • Relevant morphological information (CT, MRI) must be taken into account. • Pitfalls/sources of error: –– D2 receptor binding decline with age (about 6–8% per decade). –– Inappropriate level of contrast and background subtraction. –– Noncontinuous colour tables may overestimate findings. –– Technical artefacts: the images should be examined to detect artefacts. –– Differences in performance and accuracy of different scanners. –– Medications.

46.7 Brain PET with [18F]FDG [18F]FDG PET is the most accurate in vivo method for evaluating regional brain metabolism. Changes in neuronal activity induced by disease are mirrored by abnormalities in glucose metabolism, which can be accurately mapped with [18F]FDG-PET.

46.7.1  Common Indications • Dementing disorders (early diagnosis and differential diagnosis). • Neuro-oncology. • Epilepsy (preoperative evaluation to identify the functional epileptogenic focus). • Movement disorders (Parkinson’s disease/atypical Parkinsonian syndromes).

46.7.2  I nformation to Be Recorded Before the Procedure • History of diseases (neurological and psychiatric disorders, current neurological and psychiatric status), fasting

1111

state, prior surgery, trauma to the brain, radiation, diabetes and the use of corticosteroids. • Patient’s ability to lie still for 30 min to 1 h. If sedation is necessary, administer [18F]FDG prior to sedation (at least 20 min). • Recent morphological imaging studies and functional brain examinations. • Current medication and when last taken (particular focus on psychotropic drugs).

46.7.3  Procedure • Patients should fast for at least 4 h (some guidelines recommend 6 h). • Blood glucose levels should be checked prior to [18F]FDG administration. • Patients should void the bladder in order to keep as relaxed as possible. • The patient should lie comfortably in a quiet room (avoid to speak, read or be otherwise active) for 10 min before [18F]FDG injection (via an intravenous cannula previously placed) and for at least 20  min after administration. • For preoperative evaluation of epilepsy, continuous EEG recording is required. Monitoring should start before injection (to ensure that [18F]FDG is not administered in a postictal situation) and should be maintained at least until 20 min after injection. • Recommended activity: –– For adults: 300–600 MBq (typically 370 MBq) in 2-D mode; 125–250 MBq (typically 150 MBq) in 3-D mode. –– For children: minimum 26  MBq in 2-D mode; minimum 14 MBq in 3-D mode (3-D is recommended in children). • Careful positioning of the patient’s head is pivotal. The orbito-meatal line is often used to standardise positioning. • A CT scan can be used for attenuation correction; usually the tube voltage is set at 140 kV.

46.7.4  Interpretation • The images should be examined for the presence of movement or attenuation artefacts. It is preferable to use a normal database as reference. • Morphological changes (e.g. atrophy) should be considered in the interpretation. • It is helpful to fuse [18F]FDG images with the MRI or CT scan of the individual. • Fused PET/CT images can be easily visualised after image reconstruction.

1112

46.8 I maging of Brain Tumours with Amino Acid Tracers Brain tumours often overexpress amino acid transporters in a fashion closely correlated with altered tumour vasculature and tumour cell proliferation. These features allow for radiolabelled amino acids to reach better target-tobackground ratios than with [18F]FDG, thus improving their diagnostic performance in patients with brain tumours. Methyl-[11C]-l-methionine ([11C]MET) and several fluoro- and iodo-amino acids have been developed, including 3-123I-iodo-α-methyl-l-tyrosine (123I-IMT) for SPECT and O-(2-18F-fluoroethyl)-l-tyrosine (18F-FET) for PET.  The radiolabelled amino acid analogues are transported into cells through the same specific amino acid transport system as [11C]MET. However, once inside the cells, they are not incorporated into proteins; therefore, the labelling radionuclide does not undergo recirculation. Accumulation into brain tumours is independent from blood-brain barrier damage.

I. Marini et al.

• For 123I-IMT, block thyroidal uptake of radioiodide by an adequate regimen. • If sedation is necessary, it should be given at the earliest 1 h prior to the acquisition.

46.8.4  Procedure Activity in adults: • 123I-IMT: 100–400 MBq (typically 185 MBq). • [11C]MET: 200–250 MBq. • 18F-FET: 200–250 MBq. For children, a fraction of the adult activity should be administered (calculated from body weight). Time delay to begin of acquisition: • 123I-IMT: 15 min p.i. • A 40-min dynamic acquisition starts just after [11C]MET or 18F-FET injection. Positioning of the patient:

46.8.1  Indications • Detection of viable tumour tissue. • Confirmation of low-grade recurrences. • Tumour delineation (superior to CT and MRI in the estimation of tumour extension in low- and in high-grade gliomas, as they cannot differentiate between tumour oedema and tumour infiltration). • Non-invasive tumour grading. • [11C]MET and 123I-IMT uptake tends to correlate with the cell proliferation rate and may aid in distinguishing high-­ grade gliomas from histologically benign brain tumours or non-neoplastic lesions. • Uptake intensity is a reliable prognostic factor. • Guidance for stereotactic biopsy. • Evaluation of tumour response to treatment. • Radiotherapy planning.

46.8.2  Contraindications • Pregnancy. • Breastfeeding (interrupt breastfeeding for 24 h). • Lack of cooperation.

46.8.3  Before the Procedure • Patients should be fasting for at least 4  h and informed about the procedure.

• Patients should be informed about the total acquisition time and positioned for maximum comfort. • Patients’ comfort and perfect alignment of the head are the most important factors. • Inform the patient about the necessity to avoid movements of the head. • It is not recommended to rigidly fix the head in place.

46.8.5  [11C]MET or 18F-FET PET • Transmission scan. If attenuation correction is based on transmission images, better results are generally achieved when the images are acquired before [11C]MET of 18F-­ FET injection. • Emission scan. It is recommended to use a standardised acquisition protocol with fixed time for start of acquisition, to allow for more reliable comparison between the data of different patients or repeat scans in the same patient.

46.8.6  123  I-IMT SPECT • Multiple detectors (triple or dual head) or other dedicated SPECT gamma cameras for brain imaging should be used. • Fan-beam collimators are certainly preferred, although LEHR or LEUHR parallel-hole collimators are the most commonly available collimator sets for brain imaging.

46  Current Practical Guidelines for the Most Common Nuclear Medicine Procedures

• Acquisition parameters: –– Rotational radius: smallest possible. –– Matrix: 128 × 128. –– Angular sampling: 25,000  mmHg/min are considered sufficient, and values >30,000  mmHg/min are considered excellent for induced In patients with suspected coronary artery disease (CAD) hyperaemia. who are able to exercise to a workload of at least 85% of age-­ The exercise should be symptom-limited; if a test has to adjusted maximal predicted heart rate, physical exercise is the be stopped before reaching ≥85% of maximal heart rate, it is first test of choice, whereas for those who are unable to ade- recommended to switch or integrate the stress testing modalquately exercise, a pharmacological stress test is preferred. ity to a pharmacological stress test. Two groups of medication can be used for pharmacologiWhen the patient starts complaining for the symptoms for cal stress: which he/she was referred, despite a suboptimal increase in heart rate, administration of the radiopharmaceutical should 1. Vasodilators: adenosine, regadenoson and dipyridamole. be considered. 2. Sympathomimetic agents: dobutamine. Several treadmill exercise protocols are available; the most widely used are the Bruce and modified Bruce Appropriate facilities for cardiopulmonary resuscitation protocols. must be available, and the staff must have up-to-date knowledge of advanced life support techniques.

46.9.5  Before a Stress Study Detailed clinical history should be obtained (indication, symptoms, risk factors, current medication and prior proce-

46.9.8  Absolute Contraindications to Dynamic Exercise

• Acute coronary syndrome (the patient must be stable for at least 48 h). • Acute pulmonary embolism.

46  Current Practical Guidelines for the Most Common Nuclear Medicine Procedures

• • • • • •

Acute aortic dissection. Severe pulmonary hypertension. Uncontrolled cardiac arrhythmias. Severe aortic stenosis. Active endocarditis, acute myocarditis and pericarditis. Hypertrophic, obstructive cardiomyopathy.

46.9.9  R  elative Contraindications to Dynamic Exercise • Left bundle branch block. • Decompensated/non-adequately controlled congestive heart failure. • Active deep vein thrombophlebitis or thrombosis. • Ventricular paced rhythm. • Moderate or severe aortic stenosis. • Arterial hypertension (resting systolic or diastolic blood pressures >200/110 mmHg). • Recent stroke or transient ischemic attack. The ECG trace must be monitored continuously (during the exercise and for at least 5 min during recovery), recording or printing a 12-lead electrocardiogram at every stage of exercise, at peak and during recovery. Blood pressure should be measured every 2–3 min. The radiopharmaceutical should be injected close to the peak exercise, and the patients should be encouraged to continue the exercise for at least 2 min after the injection.

46.9.10  A  bsolute Indications for Early Termination of Exercise 1. Marked ST segment depression (≥3 mm). 2. Ischaemic ST segment elevation of >1  mm in leads without pathological Q waves. 3. Appearance of sustained ventricular tachyarrhythmia. 4. Occurrence of supraventricular tachycardia or atrial fibrillation with a high heart rate response. 5. A decrease in systolic blood pressure of >20  mmHg, despite increasing workload. 6. Marked increase of blood pressure (systolic blood pressure ≥250 mmHg; diastolic blood pressure ≥130 mmHg). 7. Angina (sufficient to cause distress). 8. Central nervous system symptoms. 9. Peripheral hypoperfusion (cyanosis/pallor). 10. Sustained ventricular tachycardia or fibrillation. 11. Inability of the patient to continue the test. 12. Technical difficulties in monitoring ECG or blood pressure.

1115

46.9.11  R  elative Indications for Early Termination of Exercise 1. ST segment depression >2  mm horizontal or downsloping. 2. Arrhythmias other than sustained ventricular tachycardia, especially if symptomatic. 3. Fatigue, dyspnoea, cramp or claudication. 4. Development of bundle branch block or intraventricular conduction defect that cannot be distinguished from ventricular tachycardia.

46.9.12  Vasodilator Stress Testing with Adenosine, Regadenoson or Dipyridamole Myocardial hyperaemia is induced by vasodilators via A2A adenosine receptors. Nevertheless, adenosine stimulates also A1, A2B and A3 adenosine receptors, which may provoke adverse effects. Instead, regadenoson selectively stimulates the A2A receptors. Dipyridamole prevents the intracellular reuptake and deamination of adenosine, thus increasing its tissue concentration. Vasodilators slightly increase heart rate, with a mild decrease in both systolic and diastolic blood pressures. Therapy with dipyridamole should be withdrawn for at least 24 h.

46.9.13  Indications Indications for the use of a vasodilator test for myocardial perfusion imaging are the same as for exercise myocardial perfusion imaging. In particular, vasodilators should be preferred to exercise testing in case of: • Left bundle branch block. • Severe chronic obstructive pulmonary disease (COPD). • Greater than first-degree heart block or sick sinus syndrome, without a pacemaker. • Symptomatic aortic stenosis and hypertrophic obstructive cardiomyopathy. • Systolic blood pressure 98% by trypan blue exclusion test). • Cell subset recovery test. • Chemotaxis or phagocytosis assays. • Efflux of 99mTc from the radiolabelled WBCs. 2. During routine clinical use: • Labelling efficiency (usually 40–80%). • Visual inspection of the radiolabelled WBC suspension. • Early in vivo lung uptake and liver-to-spleen activity ratio. 3. Periodical confirmation: • Sterility according to pharmacopoeia criteria (in case of positivity, the whole validation test must be repeated).

46.13 Thyroid Scintigraphy Thyroid scintigraphy obtained after intravenous injection of 99m TcO4− (99mTc-pertechnetate) or oral administration of 123Ior 131I-iodide allows mapping of the distribution of function in the thyroid; radioiodide for oral administration is available both as a capsule formulation and as a liquid formulation. In selected cases (e.g. vomiting, non-cooperating patients), also 123 I-iodide can be administered intravenously.

46.13.1  Clinical Indications • To evaluate how function is distributed in the thyroid gland and or in a thyroid nodule. • To calculate the activity of 131I-iodide to be administered for therapeutic purposes. • For the differential diagnosis of a neck or substernal mass.

46  Current Practical Guidelines for the Most Common Nuclear Medicine Procedures

• To confirm the diagnosis of subacute thyroiditis and to distinguish factitious hyperthyroidism from Graves’ disease and other forms of hyperthyroidism. • To localise ectopic thyroid tissue. • To evaluate patients with congenital hypothyroidism (thyroid agenesis, dyshormonogenesis, incomplete thyroid descent, etc.).

46.13.2  Patient Preparation No special preparation is required, except fasting for at least 2 h before oral administration of the radiopharmaceutical (no fasting is necessary when administering 99mTc-pertechnetate or 123I-iodide intravenously). Uptake of the radiopharmaceutical in the thyroid gland can be impaired by prior administration of exogenous thyroid hormones, antithyroid drugs or high iodine intake (e.g. iodinated contrast medium for radiological examinations, iodine-containing medications, iodide salt, etc.). In these cases, in order to avoid interferences that could lead to an incorrect diagnosis, thyroid scintigraphy should be delayed for as long as it is needed to eliminate any interference.

46.13.3  Pertinent Information Pregnant women are not eligible for thyroid scintigraphy; in case of doubt, a pregnancy test is mandatory. In case of breastfeeding mother, 99mTcO4− could be used for the scan— but only after comprehensive clinical evaluation and appropriate planning. For a correct evaluation, the patient should have a physical examination of the neck, and information should be available concerning: • Possible interfering medications and exposure to high iodine intake (see above). • Results of recent blood tests (particularly serum TSH, FT3, FT4) and of thyroid imaging procedures (ultrasound, prior scans, thyroid radioiodine uptake, etc.) • Recent diagnostic and/or therapeutic procedures with radionuclides.

46.13.4  Radiopharmaceuticals 1.

Tc-pertechnetate: it is the first-line radiopharmaceutical of choice because of its low radiation burden, ready availability in any nuclear medicine department and low cost. Activity to be administered to an adult individual ranges between 74 and 370 MBq (74 MBq in most instances). 99m

1123

I-iodide: although it yields images of better quality, it is used less frequently because it is quite expensive and generally not readily available. Nevertheless, it should be preferred when evaluating retrosternal masses, in children, and in case of low thyroid uptake. Activity to be administered to an adult individual ranges between 7.5 and 25 MBq (18.5 MBq in most instances). 3. 131I-iodide: owing to its unfavourable radiation dosimetry, its use is not recommended, unless for obtaining a curve of thyroid uptake for dosimetric purposes and when investigating patients submitted to thyroidectomy for differentiated thyroid cancer. Activity to be administered to an adult individual for a thyroid uptake curve ranges between 1.85 and 3.7 MBq (1.85 MBq in most instances). For a diagnostic whole-body scan in patients treated for differentiated thyroid cancer the activity to be administered is generally 18.5 MBq.

2.

123

46.13.5  Image Acquisition 1. Equipment: a gamma camera (preferably a small field of view), equipped with a pinhole collimator (5 mm aperture or less) for high-resolution imaging focused on the thyroid gland; a high-resolution, parallel-hole collimator may also be used for a preliminary image including in the field of view the lower part of the head and the upper part of the chest (with adequate zoom factor). 2. Patient positioning: an anterior view and both 45° anterior oblique views (with the same acquisition time) are obtained with the patient lying supine with the neck extended (the sitting position can be considered in patients unable to lie supine), at a distance allowing the thyroid image to correspond to about two-thirds of the field of view. 3. Timing: • 99mTc-pertechnetate: start imaging at about 20 min (not later than 30 min) after injection. • 123I-iodide: images can be obtained as early as 3–4  h after oral administration (although background is lower at 16–24 h, the count rate is significantly reduced). • 131I-iodide: images are generally obtained the day after administration. 4. Acquisition parameters: • 99mTc-pertechnetate: 100,000–200,000 counts or (5 min) • 123I-iodide: 50,000–100,000 counts (or 10 min) • 131I-iodide: up to 50,000 counts (or 10 min). 5. For correct image interpretation, radioactive or radi opaque markers can be used (e.g. to localise palpable nodules, the sternal notch, the thyroid cartilage, palpable masses in the neck). 6. Asking the patient to drink some water turns out to be useful to reduce accumulation of radioactivity in the mouth or oesophagus that is sometimes observed.

1124

46.13.6  Interpretation Criteria Complete pre-evaluation of the patient (see above) is mandatory for correct image interpretation in the clinical context. The thyroid should be described for anatomical location, size, symmetry, uniformity and intensity of tracer uptake as compared to surrounding background (e.g. the salivary glands); other focal areas of uptake outside the thyroid must be reported. When thyroid nodules are detected, intensity of uptake in the nodules compared to the extra-nodular thyroid tissue must be described: particular attention must be paid to possible partial suppression (which is easy to identify in case of total extra-nodular suppression). Knowledge of the TSH serum level facilitates the task.

46.13.7  Reporting After having described the images obtained, it is mandatory to report their meaning on the grounds of physical examination and patient’s history. The report must answer the questions raised by the referring clinician, in order for the examination to be helpful for patient management. In case of hyperthyroidism and diffusely increased radiopharmaceutical uptake, the report should indicate a diagnosis of Graves’ hyperthyroidism, while if uptake is limited to a single or multiple thyroid nodule/s suppressing the remnant tissue, a Plummer hyperthyroidism or a toxic multinodular goitre must be diagnosed. When observing a focal defect of uptake within the thyroid gland, special attention should be paid to assess correspondence of the “cold” area with a palpable nodule; although in the large majority of cases this is due to a benign disease, further diagnostic tests (e.g. fine needle aspiration cytology) should be suggested to exclude thyroid cancer. Possible Sources of Error Include: • Local contamination. • Oesophageal activity due to swallowing of saliva containing 99mTc-pertechnetate or radioiodide physiologically excreted through the salivary glands (drinking some water will clear the activity accumulated in the oesophagus). • Reduced thyroid uptake due to iodine containing drugs, iodinated contrast agents, iodine-containing foods or medications.

46.14 Radiometabolic Therapy with 131I-Iodide for Hyperthyroidism Hyperthyroidism from Graves’ disease, multinodular goitre and solitary hyperfunctioning thyroid nodule can be treated with 131I-iodide to achieve a “non-hyperthyroid status” (either euthyroid or hypothyroid).

I. Marini et al.

Euthyroid patients with large nodular goitre (NTG) causing mechanical compression symptoms who refuse surgery or for whom surgery is contraindicated can benefit from radiometabolic therapy with 131I-iodide, with the aim of reducing thyroid volume; an ad hoc specific evaluation is required to assess eligibility of patients for such therapy.

46.14.1  Treatment Options for Hyperthyroidism and Nontoxic Goitre Antithyroid drugs (ATDs) and radioiodine therapy are both used to treat hyperthyroidism. Radioiodine is the treatment of choice, as a first-line approach in patients with hyperthyroidism from solitary hyperfunctioning thyroid nodules and in patients with postoperative recurrent goitre. It is considered a second-line treatment after antithyroid drugs in case of resistance, adverse reactions or recurrence, especially in patients with medical contraindications to surgery and slight or moderate compressive symptoms, as well as in those patients who refuse surgery. Surgery, on the other hand, is preferred in selected patients with large nodular goitre with severe compression of nearby structures, Graves’ ophthalmopathy and need for immediate effectiveness or in the suspicion of coexisting malignancy.

46.14.2  Contraindications • Absolute: pregnancy and breastfeeding. • Relative: uncontrolled hyperthyroidism and active Graves’ ophthalmopathy.

46.14.3  Patient Preparation • Evaluation of patient history (special focus on previous and current treatments), laboratory test results (free T4, free T3, TSH, TPO and TSI-Ab) and thyroid target volume. A negative pregnancy test in female patients of childbearing age. • Thyroid 99mTc-pertechnetate scintigraphy and 24-h radioiodide uptake (if   1  ng/mL after primary treatments). In these patients, its diagnostic accuracy is >80%.

Case No. 48.40: PET/CT with [18F]FDG for Staging and Radiotherapy Planning in Penile Cancer

a Presentation [18F]FDG PET/CT was performed in a 36-year-old man submitted 1 year earlier to surgery because of penile carcinoma (initial stage pT2N2). The scan was performed as an aid to radiotherapy planning, after a CT scan had revealed a pelvic recurrence. Findings

b The PET/CT scan demonstrated a focally increased uptake of [18F]FDG uptake in the right deep inguinal (a) and iliac lymph nodes (b). Discussion [18F]FDG PET/CT is indicated in patients with recurrent penile cancer for the evaluation of local and distant metastases. The morphologic and metabolic information provided by the scan is particularly useful to plan adjuvant radiotherapy treatment.

48  Teaching Cases in Nuclear Medicine: Oncological Applications

1237

Case No. 48.41: PET/CT with [18F]FDG for Initial Staging in Testicular Cancer

Presentation [18F]FDG PET/CT was performed for initial staging in a 23-year-old man with newly diagnosed embryonal testicular carcinoma (cT3NxMx). Serum levels of beta-HCG, alphafetoprotein, and LDH were 1758 IU/L, 48.8  ng/mL, and 1151 IU/L, respectively. Findings The PET/CT scan reveals multiple foci with increased [18F] FDG uptake, both in the sovra- and in the sub-diaphragmatic

region (yellow arrows), as well as multiple areas with focally increased tracer uptake in the liver and in the spine (L3 body, red arrow). Discussion PET/CT with [18F]FDG allows a whole-body evaluation, thus providing complete staging of disease. In this patient, the presence of diffuse metastatic disease (stage IV) shifts the therapeutic approach from surgery to systemic chemotherapy.

1238

L. Evangelista et al.

Case No. 48.42: PET/CT with [18F]FDG for Suspected Recurrence in Testicular Cancer

Presentation [18F]FDG PET/CT was performed for suspected tumor recurrence in a 44-year-old man who had been submitted to surgery several years earlier because of testicular seminoma (initial stage I). Findings The PET/CT scan shows a focus of markedly increased [18F]FDG uptake (SUVmax 8.8) indicating lymph node recurrence in the right external iliac region. Discussion [18F]FDG PET/CT is recommended in patients with clinical/imaging suspicion for recurrent testicular cancer, especially if the histologic type is seminoma, and in patients with rising serum levels of tumor-associated markers (e.g., alpha-fetoprotein) after primary treatment.

48  Teaching Cases in Nuclear Medicine: Oncological Applications

References 1. Som P, Atkins HL, Bandoypadhyay D, et al. A fluorinated glucose analog, 2-fluoro-2-deoxy-D-glucose (F-18): nontoxic tracer for rapid tumor detection. J Nucl Med. 1980;21:670–5. 2. Podo F.  Tumour phospholipid metabolism. NMR Biomed. 1999;12:413–39. 3. Kwee SA, Coel MN, Ly BH, et al. 18F-Choline PET/CT imaging of RECIST measurable lesions in hormone refractory prostate cancer. Ann Nucl Med. 2009;23:541–8. 4. Luboldt W, Küfer R, Blumstein N, et  al. Prostate carcinoma: diffusion-­weighted imaging as potential alternative to conventional MR and 11C-choline PET/CT for detection of bone metastases. Radiology. 2008;249:1017–25. 5. Huang Z, Zuo C, Guan Y, et al. Misdiagnoses of 11C-choline combined with 18F-FDG PET imaging in brain tumours. Nucl Med Commun. 2008;29:354–8. 6. Hara T, Kosaka N, Shinoura N, et al. PET imaging of brain tumor with [methyl-11C]choline. J Nucl Med. 1997;38:842–7. 7. Khan N, Oriuchi N, Ninomiya H, et  al. Positron emission tomographic imaging with 11C-choline in differential diagnosis of head and neck tumors: comparison with 18F-FDG PET. Ann Nucl Med. 2004;18:409–17. 8. Salem N, Kuang Y, Wang F, et al. PET imaging of hepatocellular carcinoma with 2-deoxy-2[18F]fluoro-D-glucose, 6-deoxy-6[18F] fluoro-D-glucose, [1-11C]-acetate and [N-methyl-11C]-choline. Q J Nucl Med Mol Imaging. 2009;53:144–56. 9. Beheshti M, Langsteger W. PET imaging of prostate cancer using radiolabeled choline. PET Clin. 2009;4:173–84. 10. Pascali G, D’Antonio L, Bovone P, et al. Optimization of automated large-scale production of [18F]fluoroethylcholine for PET prostate cancer imaging. Nucl Med Biol. 2009;36:569–74. 11. Youland RS, Kitange GJ, Peterson TE, et al. The role of LAT1 in 18 F-DOPA uptake in malignant gliomas. J Neuro-Oncol. 2013;111:11–8. 12. Gazdar AF, Helman LJ, Israel MA, et al. Expression of neuroendocrine cell markers L-dopa decarboxylase, chromogranin A, and dense core granules in human tumors of endocrine and nonendocrine origin. Cancer Res. 1988;48:4078–82. 13. F T, Ehmer J, Piroth MD, et al. The quantification of dynamic FET PET imaging and correlation with the clinical outcome in patients with glioblastoma. Phys Med Biol. 2009;54:5525–39. 14. Stadlbauer A, Prante O, Nimsky C, et  al. Metabolic imaging of cerebral gliomas: spatial correlation of changes in O-(2-18F-­ fluoroethyl)-L-tyrosine PET and proton magnetic resonance spectroscopic imaging. J Nucl Med. 2008;49:721–9. 15. Stadlbauer A, Pölking E, Prante O, et al. Detection of tumour invasion into the pyramidal tract in glioma patients with sensorimo-

1239 tor deficits by correlation of 18F-fluoroethyl-L-tyrosine PET and magnetic resonance diffusion tensor imaging. Acta Neurochir. 2009;151:1061–9. 16. Sharma MR, Maitland ML, Ratain MJ.  RECIST: no longer the sharpest tool in the oncology clinical trials toolbox. Cancer Res. 2012;72:5145–9. 17. Ruiz C, Lenkiewicz E, Evers L, et al. Advancing a clinically relevant perspective of the clonal nature of cancer. Proc Natl Acad Sci U S A. 2011;108:12054–9. 18. http://www.esmo.org/Guidelines 19. https://www.nccn.org/professionals/physician_gls/f_guidelines.asp 20. Boellaard R, Delgado-Bolton R, Oyen WJG, Giammarile F, Tatsch K, Eschner W, et  al. FDG PET/CT: EANM procedure guidelines for tumour imaging: version 2.0. Eur J Nucl Med Mol Imaging. 2015;42:328–54. 21. Delbeke D, Coleman RE, Guiberteau MJ, Brown ML, Royal HD, Siegel BA, et al. Procedure guideline for tumor imaging with 18F-­ FDG PET/CT 1.0. http://www.snm.org/guidelines

Further Reading Hara T, et al. PET imaging of prostate cancer using carbon-11-choline. J Nucl Med. 1998;39:990–6. Hogerle S, et al. Whole-body 18F-DOPA PET for detection of gastrointestinal carcinoid tumors. Radiology. 2001;220:373–80. Popperl G, et al. Analysis of 18F-FET PET for grading of recurrent gliomas: is evaluation of uptake kinetics superior to standard methods? J Nucl Med. 2006;47:393–403. Shreve PD, et al. Pitfalls in oncologic diagnosis with FDG PET imaging: physiologic and benign variants. Radiographics. 1999;19:61–7. Wahl RL, et al. From RECIST to PERCIST: evolving considerations for PET response criteria in solid tumors. J Nucl Med. 2009;50(suppl 1):122S–50S. Zhuang H, et  al. Dual time point 18F-FDG PET imaging for differentiating malignant form inflammatory processes. J Nucl Med. 2001;42:1412–7.

Links to Interesting Websites www.eanm.org. Accessed 16 Oct 2017. www.snmmi.org. Accessed 14 Oct 2017. www.nccn.org. Accessed 10 Oct 2017. www.esmo.org. Accessed 14 Oct 2017. www.ncbi.nlm.nih.gov/pubmed/. Accessed 26 Oct 2017.

Radiological Anatomy with CT: What the Nuclear Physician Should Know When Reading a PET/CT Scan

49

Davide Caramella, Matteo Revelli, and Alessandro Villa

Contents 49.1

Introduction

 1241

49.2  eck and Chest CT Anatomy N 49.2.1  Essential of Neck Anatomy for CT Interpretation 49.2.2  Cervical and Mediastinal Lymph Node Stations 

 1241   1242   1244

49.3  racheobronchial Tree and Pulmonary Segments T 49.3.1  Tracheobronchial Tree 49.3.2  Bronchopulmonary Segments

 1246   1246   1247

49.4

Mediastinum, Pleura, and Chest Wall

 1248

49.5  bdominal and Pelvic Anatomy and CT Technique A 49.5.1  Abdominal Anatomy 49.5.2  Pelvic Anatomy

 1248   1249   1255

Further Reading

 1256

Learning Objectives

• Describe the normal imaging features of anatomical structures in different body regions. • Localize image findings in specific anatomic compartments. • Get familiar with the spatial anatomy of cancer staging-­ related body regions.

D. Caramella (*) Diagnostic and Interventional Radiology, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy e-mail: [email protected] M. Revelli Department of Diagnostic Imaging and Laboratory Medicine, AUSL Reggio Emilia – IRCCS, Reggio Emilia, Italy A. Villa Unit of Radiology 2, ASL 5 “Spezzino”, Sarzana, La Spezia, Italy

49.1 Introduction In diagnostic imaging, knowledge of normal anatomy is the basic requirement for a correct interpretation of the morphological alterations that may involve the various organs of the body. An alteration in shape is most frequently associated with an alteration in function, which is why the image specialist will have to direct the clinician’s attention to certain pathologies, on the basis of the recognition of specific patterns and alterations. This chapter will provide basic elements of CT anatomy needed to recognize normal human anatomical structures, with some hints to the districts of interest in the staging of the most common tumors.

49.2 Neck and Chest CT Anatomy In this part of the chapter, the neck and chest anatomy will be discussed, considering their anatomical subdivisions and the main aspects that can help in the diagnosis or in the staging protocol. The main anatomical structures will be

© Springer Nature Switzerland AG 2019 D. Volterrani et al. (eds.), Nuclear Medicine Textbook, https://doi.org/10.1007/978-3-319-95564-3_49

1241

1242

D. Caramella et al.

a

b

c

d

e

f

Fig. 49.1  Main structures of the neck on the axial plane (caudocranial from a to f). T thyroid gland, CC cricoid cartilage, TVC true vocal cords, TC thyroid cartilage, LT lingual tonsil, EG epiglottis, OP oropharynx, NP nasopharynx, PT palatine tonsil, TT torus tubarius

discussed, and a complete analysis of cervical fascial anatomy and cervical and mediastinal nodal stations will be provided.

49.2.1 Essential of Neck Anatomy for CT Interpretation The main anatomical structures of the neck will be discussed in this paragraph: an image-oriented approach is provided in Fig.  49.1. The pharynx is divided into three portions, and, like many cervical structures, it is disposed symmetrically with respect to the median line: therefore, the search for asymmetries is often the first approach to the evaluation of several pathologies, especially when it comes to neoplastic disease. The rhinopharynx is the upper level, bordered by the sphenoidal sinus in the upper side, by the nasal cavities in the front, by the pharyngeal wall posteriorly, and it extends down to a level through the lower margin of the soft palate. The pharyngeal recess (the fossa of Rosenmüller) is located posterior and superior to the torus tubarius, the cartilage portion of the Eustachian tube. The

oropharynx extends from the soft palate to the vallecula, and is separated from the oral cavity by a ring formed by the palates and the tonsillar pillars. It contains the base of the tongue and the lingual and palatine tonsils. The hypopharynx extends down reaching the cricopharyngeal muscle; it contains three subregions, the pyriform sinus, the posterior wall, and the post-cricoid region (or pharyngoesophageal junction). The larynx is the continuation of the oropharynx, and it lies between the epiglottis and the inferior aspect of the cricoid cartilage and then continuing with the trachea. It consists of a cartilage skeleton divided in three regions: • Supraglottis, extending from the epiglottis to the laryngeal ventricles: it contains the epiglottis (supra- and infrahyoid), the pre-epiglottic fat tissue, the aryepiglottic folds, the false vocal cords, and the arytenoid cartilages. • Glottis, containing the true vocal cords and the anteroposterior commissure. • Subglottis, extending from the inferior aspect of the true vocal cords to the inferior surface of the cricoid cartilage and containing the cricoid ring mucosa.

49  Radiological Anatomy with CT: What the Nuclear Physician Should Know When Reading a PET/CT Scan

The lymphoid tissue is made up of the Waldeyer ring, whose main components are the palatine tonsils, pharyngeal tonsils, lingual tonsils, and adenoids. The latter are located in the vault of the rhinopharynx, and they can extend to the fossa of Rosenmüller; they are not usually recognizable in adults. The tonsils are symmetrical structures that may be affected by pathology, often unilaterally. MRI is the modality of choice for the study of lymphoid tissue in the neck. The salivary glands include  the parotid, consisting of glandular tissue and adipose tissue that give it a hypodense aspect on  CT.  From the front of the parotid originates the Stenone duct, which may be affected by lithiasis. The other salivary glands are the submandibular gland, located in the floor of the mouth, the sublingual gland, and a large number of minor salivary glands. The thyroid is a central and asymmetric gland located in front of the trachea and larynx. It has a hyperdense appearance in basal scans due to high iodine content. The parathyroid glands may not be visible on CT, unless they are abnormally enlarged because of disease.

49.2.1.1 F  ascial Anatomy: Basics of CT Approach to Neck Pathology The use of fascial anatomy is largely used in cross-sectional imaging for the evaluation of the cervical region, allowing the classification of suprahyoid and infrahyoid spaces (Fig. 49.2). A systematic approach to cervical masses, with the evaluation of the origin of the lesion, its extension, and infiltrative features, requires a precise knowledge of normal anatomy and a clear classification of cervical spaces. Three layers of the deep cervical fascia are the limits of the deep spaces in the neck. The deep fascia entirely surrounds the neck and splits to enclose the sternocleidomastoid and trape-

Fig. 49.2  Main fascial spaces of the infrahyoid neck on the axial plane. RS retropharyngeal space, VS visceral space, CS carotid space, PCS posterior cervical space, PVS perivertebral space

1243

zius muscles. The pretracheal fascia surrounds the content of the visceral space. The paravertebral fascia encloses the paraspinous and prevertebral muscles, forming the perivertebral space. The carotid sheath surrounds the vascular and nervous structures of the neck.

49.2.1.2 Suprahyoid Spaces Parapharyngeal space: it extends from the base of the skull to the hyoid bone, before the styloid process. It is made of adipose tissue and contains vascular and nervous structures; the secondary involvement from pathologies originating from adjacent spaces is frequent, and it can be displaced by space-occupying masses. Parotid space: it is located behind the masseter muscle and the mandible, lateral to the parapharyngeal space, surrounded by the projections of the deep cervical fascia; it contains the parotid, the facial nerve, and several vascular structures in addition to the parotid duct. Pharyngeal mucosal space: it contains a large variety of mucosal surfaces, which can result in squamous rhinopharyngeal, oropharyngeal, and hypopharyngeal carcinomas. It is surrounded by the middle leaf of the deep cervical fascia. Masticatory space: it is anterior to the parotid space, anterolateral to the parapharyngeal space, and behind the buccal space. Soft tissue sarcomas are frequently found in this space and may extend to the regional nerve structures to the cranial base. Buccal space: it is a horizontal space and is not delimited by the cervical fascia; it is almost completely occupied by adipose tissue; and the most common lesions are squamous carcinomas. 49.2.1.3 Infrahyoid Spaces Anterior cervical space: two symmetrical spaces in the anterolateral site of the infrahyoid neck. The major lesions that we encounter are neoplastic or infectious pathologies that extend from adjacent spaces. Posterior cervical space: as the anterior cervical space, it consists of two symmetrical spaces in the posterolateral region of the neck, with triangular morphology; the most common lesions are lymphomas and metastatic lymph node involvement from squamous cell carcinomas. Visceral space: it is the only space completely infrahyoid; it can be divided into a laryngeal area, a thyroid area, a parathyroid area, and an esophageal area. It has important clinical implications in the evaluation of thyroid, parathyroid, and squamous cell disorders of the hypopharynx and larynx. 49.2.1.4 C  ommon Suprahyoid and Infrahyoid Spaces Carotid spaces: they are symmetrical and extend from the jugular foramen to the aortic arch, involving both the supra-

1244

hyoid and the infrahyoid neck. They are located behind the parapharyngeal space and lateral to the retropharyngeal space. The lesions in the carotid space usually compress the adjacent vascular structures and are most frequently represented by lymph node metastases of the deep cervical chain and neurogenic tumors. Retropharyngeal space: it is a posterior median space extending to the upper mediastinum; its posterior portion is called the “danger space”, since it is particularly prone to tumor and flogistic dissemination of pathologies of the cervical region, which can extend to the cranial base and to the mediastinum. Prevertebral space: it extends posteriorly to the retropharyngeal space, and it can be further subdivided into two portions (the posterior part is also known as perivertebral space); the front compartment lesions displace the prevertebral muscle. This space contains muscular, vascular, and nerve structures, as well as the cervical vertebrae, and the bone lesions of this space have great clinical value.

49.2.2 Cervical and Mediastinal Lymph Node Stations The neck is the body area containing the largest number of lymph nodes. They appear on CT as oval- or round-shaped structures, usually isodense with respect to parenchymal organs. There are several classification systems used for the evaluation of cervical lymph nodes, but the most widely used by imaging specialists is the one by the American Academy of Otolaryngology, Head and Neck Surgery. This classification divides cervical lymph nodes into six levels (plus level 7 including superior mediastinal nodes).

49.2.2.1 Cervical Lymph Node Stations Level I: lymph nodes located below the mylohyoid muscle and above the lower margin of the hyoid bone, as well as nodes anterior to the posterior border of the submandibular glands. They can be classified as: • Level Ia: submental nodes. • Level Ib: submandibular nodes. Level II: superior jugular lymph nodes, including the superior internal jugular chain, nodes from the base of the skull to the inferior border of the hyoid bone, anterior to the posterior border of the sternocleidomastoid muscle, and

D. Caramella et al.

posterior to the posterior border of the submandibular glands. • Level IIa: anterior, lateral, or medial to the internal jugular vein. • Level IIb: posterior to the internal jugular vein. Level III: middle jugular lymph nodes, including the middle part of the internal jugular chain, nodes from the lower margin of the hyoid to omohyoid muscle, anterior to the posterior border of the sternocleidomastoid muscle, and from the lateral to the medial margin of the common carotid artery/ internal carotid artery. Level IV: lower jugular lymph nodes, including the lower internal jugular chain, nodes from the lower margin of cricoid cartilage to the level of the clavicle, anterior and medial to an oblique line drawn through the posterior edge of the sternocleidomastoid muscle and the posterolateral edge of the anterior scalene muscle and lateral to the medial margin of the common carotid artery. Level V: posterior triangle nodes, divided into: • Level Va: superior half, posterior to levels II and III (between the base of the skull and inferior border of the cricoid cartilage). • Level Vb: inferior half, posterior to level IV (between the inferior border of the cricoid cartilage and the level of the clavicles). Level VI: prelaryngeal/pretracheal lymph nodes, anterior to the visceral space, from the inferior margin of the hyoid bone to the manubrium, anterior to levels III and IV. Level VII: superior mediastinal lymph nodes, located between the common carotid arteries, below the superior aspect of the manubrium to the level of the brachiocephalic vein. Note that facial, occipital, retropharyngeal, and parotid nodes are not considered by this classification and should be referred to independently in the report.

49.2.2.2 Mediastinal Lymph Node Stations Regarding mediastinal lymph nodes, the most widely used classification is the one proposed by the International Association for the Study of Lung Cancer (IASLC). It divides mediastinal lymph nodes into the following stations (Fig. 49.3): Supraclavicular nodes (level 1), including low cervical, supraclavicular, and sternal notch nodes, located from the

49  Radiological Anatomy with CT: What the Nuclear Physician Should Know When Reading a PET/CT Scan

a

b

c

d

e

f

Fig. 49.3  Main mediastinal lymph node stations on the axial plane (craniocaudal from a to f). RLC, right low cervical/supraclavicular, LLC left low cervical/supraclavicular, RUP right upper paratracheal,

1245

LUP left upper paratracheal, PV prevascular, APW aortopulmonary window, RH right hilar and perihilar, SC subcarinal, LH left hilar and perihilar, PE paraesophageal, LPL left pulmonary ligament

1246

D. Caramella et al.

lower margin of the cricoid cartilage to the clavicles and the upper border of the manubrium. Superior mediastinal nodes (levels 2–4), including: • Upper paratracheal nodes (2R/2L), extending to the lateral border of the trachea, from the upper border of the manubrium to the intersection of the caudal margin of the left brachiocephalic vein with the trachea on the right or to the superior border of the aortic arch on the left. • Prevascular nodes (3A), located anterior to the vessels. • Prevertebral nodes (3P), located behind the esophagus. • Lower paratracheal nodes (4R/4L), extending from the intersection of the caudal margin of the left brachiocephalic vein with the trachea to the lower border of the azygos vein on the right or from the upper margin of the aortic arch to the upper rim of the left main pulmonary artery on the left. Aortic nodes (levels 5–6), including: • Subaortic nodes (5), located in the aortopulmonary window lateral to the ligamentum arteriosum. • Para-aortic nodes (6), including the ascending aorta or phrenic nodes lying anterior and lateral to the ascending aorta and the aortic arch. Inferior mediastinal nodes (levels 7–9), including: • Subcarinal nodes (7). • Paraesophageal nodes (8). • Pulmonary ligament nodes (9). Hilar nodes (level 10), including nodes adjacent to the main bronchus and hilar vessels. a

b

Fig. 49.4  Bronchial tree on the coronal planes (a, b) and sagittal plane (right lung in c). RMB right main bronchus, LMB left main bronchus, RULB right upper lobe bronchus, IB intermediate bronchus, LULB left

Key Learning Points

• CT is a widely used imaging technique for the evaluations of the neck pathologies. • Knowledge of the spatial and fascial neck anatomy is essential to reach the correct diagnosis. • CT plays a fundamental role in the evaluation of lymph node involvement in staging neck and pulmonary tumors.

49.3 Tracheobronchial Tree and Pulmonary Segments 49.3.1 Tracheobronchial Tree The tracheobronchial tree is the branching tree of airways beginning at the larynx and extending inferiorly and peripherally into the lungs. Moving toward the peripheral lung, there is a decrease of the luminal diameters of the branches. The tracheobronchial tree includes (Fig.  49.4) the trachea; the left and right main (primary) bronchi which bifurcate at the carina located in the transthoracic plane of Ludwig; the lobar (secondary) and segmental (tertiary) bronchi within the lungs; and smaller subdivisions, terminating in the alveoli in the periphery of the lungs. At the level of the tertiary segmental bronchi, sections of the lung called bronchopulmonary segments are defined: they represent the largest subdivisions of a lobe that can undergo surgical resection, and they are separated from adjacent segments by thin fibrous septa, forming the pulmonary interstitium. Segments have an independent supply of air (by a single segmental bronchus) and blood (by a tertiary branch of the pulmonary c

upper lobe bronchus, LILB left inferior lobe bronchus, MLB middle lobe bronchus, RILB right inferior lobe bronchus

49  Radiological Anatomy with CT: What the Nuclear Physician Should Know When Reading a PET/CT Scan

artery). The venous drainage follows intersegmental tributaries of the pulmonary veins that are located in the interstitium. The tracheobronchial tree is drained by lymphatic vessels which follow the course of the bronchi and pulmonary arteries toward the hilum. Ultimately, these lymph vessels drain to hilar lymph nodes and then ascend to the mediastinal lymph nodes. Fig. 49.5  Lung lobes on the right sagittal plane (a) and left sagittal plane (b). RUL right upper lobe, ML middle lobe, RIL right inferior lobe, LUL left upper lobe, LIL left inferior lobe

1247

49.3.2 Bronchopulmonary Segments The left lung is smaller and only contains two lobes (Figs. 49.5 and 49.6). In general, each lung has ten segments: the upper lobes contain three segments, the middle lobe/lingula two, and the lower lobes five. Bilaterally, the upper lobes have apical, posterior, and anterior segments and the lower lobes superior

a

b

a

b

c

d

Fig. 49.6  Lung lobes on the axial plane at different levels (craniocaudal from a to d). RUL right upper lobe, LUL left upper lobe, RIL right inferior lobe, LIL left inferior lobe, ML middle lobe, L lingula

1248

(apical) and four basal segments (anterior, medial, posterior, and lateral). With this basic symmetric anatomy shared between the lungs, there are a few differences: for example, the middle lobe on the right has a medial and a lateral segment, while the lingula on the left has a superior and an inferior segment; furthermore, there are two regions of the left lung in which two segments are joined as one as they have a common tertiary bronchus, in the left upper lobe apicoposterior segment and in the left lower lobe anteromedial segment. The right lung is divided into three lobes with ten segments (each segment can be referred to using a letter and a number, following Boyden classification):

49.3.2.1 Right Upper Lobe • Apical segment (B1). • Posterior segment (B2). • Anterior segment (B3). 49.3.2.2 Middle Lobe • Lateral segment (B4). • Medial segment (B5). 49.3.2.3 Right Lower Lobe • Superior segment (B6). • Medial segment (B7). • Anterior segment (B8). • Lateral segment (B9). • Posterior segment (B10). The left lung is divided into two lobes and eight segments:

49.3.2.4 Left Upper Lobe • Apicoposterior segment (B1/2). • Anterior segment (B3). • Superior lingular segment (B4). • Inferior lingular segment (B5). 49.3.2.5 Left Lower Lobe • Superior segment (B6). • Anteromedial segment (B8). • Lateral segment (B9). • Posterior segment (B10).

D. Caramella et al.

49.4 Mediastinum, Pleura, and Chest Wall The mediastinum is commonly divided into three compartments, the posterior mediastinum, the middle mediastinum, and the anterior mediastinum; however, the identification of real anatomical boundaries is difficult and very variable. The posterior mediastinum contains the descending aorta; the esophagus and some venous, such as the azygos system; and lymphatic structures, such as the thoracic duct. The middle mediastinum is centrally located and contains the heart and the major tracheobronchial branches. The anterior mediastinum is located before the pericardium, the ascending aorta, and the superior vena cava; it may physiologically contain descending thyroid portions and thyme residues in young adults. The absence of anatomical structures that separate the various mediastinal compartments makes it difficult to limit the expansion of any pathological process, which may therefore originate from a mediastinal portion and then extend to the others. Regarding the pleura and the thoracic wall, CT is not able to discriminate between the parietal pleura and visceral pleura if the two leaflets are not separated by the presence of fluid effusion, since the pleural space is normally only a virtual space. Therefore, the pleura appears indistinguishable from the endothoracic fascia and from the intercostal muscles. The intercostal vascular and nervous bundles run adjacent to the lower margin of each rib: this information is important for performing interventional procedures such as transthoracic biopsies of pulmonary nodules or pleural fluid drainages, in order to avoid to accidentally damage  these structures.

Key Learning Points

• CT is the technique of choice for the evaluation of mediastinal pathologies. • Knowledge of mediastinal compartments is essential to formulate a correct diagnosis.

Key Learning Points

49.5 A  bdominal and Pelvic Anatomy and CT Technique

• CT is the gold standard imaging technique for the evaluation of the pulmonary parenchyma. • Knowledge of the bronchopulmonary segmentation is essential for correct preoperative staging of lung cancer.

The boundaries of the  abdominal cavity are superiorly the diaphragm and inferiorly the upper boundary  of the pelvic cavity,  a plane  from the prominence of the sacrum to the pubic symphysis.

49  Radiological Anatomy with CT: What the Nuclear Physician Should Know When Reading a PET/CT Scan

49.5.1 Abdominal Anatomy 49.5.1.1 Spaces of the Abdomen Cavity The abdomen can be divided into the peritoneal and extraperitoneal (or retroperitoneal) spaces. Peritoneal space: the peritoneum is a serosal membrane that covers the abdominal wall (parietal peritoneum) and lines the peritoneal organs (visceral peritoneum). The parietal peritoneum forms the peritoneal ligaments, mesenteries, and omenta  by reflecting itself over parietal organs. The mesenteries are double layers  of peritoneum that attaches part of the small and great bowel to the posterior abdominal wall. The mesenteries contain  blood vessels, nerves, and lymphatic vessels. The peritoneum encloses the transverse colon forming the transverse mesocolon, the small bowel forming the small bowel mesentery, the appendix forming the mesoappendix, and the sigmoid colon forming the sigmoid mesocolon. Peritoneal omenta are  considered peritoneal ligaments (connecting different organs) and can be divided into the greater and the lesser peritoneal sac: the greater sac or greater omentum is a double-layer sheet derived from the inferior edge of the great curvature of the stomach that passes down the anterior surface of the small bowel. After a certain distance, it folds on itself reaching the anterior superior margin of the transverse colon. On the left the omentum becomes the gastrosplenic ligament. The greater omentum is not usually appreciable during a CT exam, but it becomes visable when peritoneal implants from gastrointestinal or ovarian malignancy cause irregular thickening of this surface or when it is distended by fluid. The lesser omentum is a peritoneal fold extending from the lesser curvature of the stomach to the porta hepatis of the liver. The peritoneal space can be divided into two compartments by the mesentery of the transverse colon: the supramesocolic compartment, which contains the liver, stomach, and spleen, and the inframesocolic compartment, which contains the small bowel and ascending and descending colon. These two compartments are connected laterally by the paracolic gutters. The pelvic space can be considered as a third compartment and will be discussed separately. Supramesocolic compartment: the fusion and the rotation processes of mesenteric structures due to embryological development lead to a complex anatomy of peritoneal space that can be classified into right and left peritoneal spaces. The left peritoneal space can be divided into anterior and posterior perihepatic spaces and anterior and posterior subphrenic spaces. The left anterior perihepatic space is located anterior to the left liver lobe, and it is confined at the right side by the falciform ligament. The left posterior perihepatic space is located at the medial side of the left hepatic lobe and reaches the venous ligament (Aranzio’s ligament).

1249

The anterior subphrenic space is located between the gastric fundus and the anterior part of the left hemidiaphragm. The posterior left subphrenic space is the perisplenic space at the left of the spleen. The right peritoneal space can be divided into the  perihepatic space (or right subphrenic space) and lesser omentum space. The right perihepatic space is located between the right h­ emidiaphragm and the right hepatic lobe. The caudal portion of this space is called Morrison’s pouch or right subhepatic space  which separates the anterior fascia of the right kidney from the posterior aspect of the right hepatic lobe. This space is a common site for fluid collection, especially in liver trauma due to gravity. The lesser sac space lies between the posterior wall of the stomach and the anterior surface of the pancreas. This virtual space joins the greater sac via Winslow’s foramen on the right. Medially it extends to the “porta hepatis.” Fluid collections such as in ascites can cause this space to be better visualized. Inframesocolic compartment: can be divided into two spaces—the right inframesocolic space, which is bounded by the transverse colon on the anterior superior side, located at the right of the small bowel mesentery, and the left inframesocolic space which presents a direct communication with the pelvic space. The right and left paracolic gutters are part of this space, and they allow a direct communication between the inframesocolic space and the supramesocolic space. In the pelvic space, the parietal peritoneum covers the bladder dome and the anterior surface of the rectum. In women the peritoneum also covers the anterior and posterior margins of the uterus. In males, when a fluid collection is present, it tends to accumulate in the rectovesical pouch, while in women  it accumulates  in the rectouterine pouch (pouch of Douglas) or the uterovesical pouch. Extraperitoneal (retroperitoneal) space: the retroperitoneal space can be defined as the space between the parietal peritoneum and the transversalis fascia. It can be divided into three spaces by the perirenal fascia: the anterior pararenal space, the perirenal space, and the posterior pararenal space. The great vessel space is the fourth retroperitoneal space not well defined by the fascial plane. Pararenal space: the space between the anterior perirenal fascia and the posterior parietal peritoneum is a complex area that contains a large part of the duodenum, the ascending and descending colon, and the pancreas; pathological processes that originate from one of these organs may spread in the anterior pararenal space and involve the other structures. These spaces are laterally demarcated by the lateroconal fascia, and anteriorly they communicate with the root of the small intestine mesentery, the root of the transverse mesocolon, and the lesser sac. Superiorly the anterior pararenal space is in continuity with the diaphragmatic dome pos-

1250

D. Caramella et al.

Great vessel space: the aorta and its main visceral branches and the inferior vena cava with its tributary veins pass in this anatomical space that is not well-defined and is  located in front of the vertebral body.  This space represents the caudal continuation of the posterior mediastinal space, but it is not delimited by true anatomical boundaries. Fluid collections that originate within this space can spread to other retroperitoneal spaces due to their communication. In case of rupture of an abdominal aorta aneurysm, the perirenal spaces may be involved, while the anterior pararenal space is less commonly involved.

Fig. 49.7  Pararenal and perirenal spaces. CECT in the portal phase for pancreatitis evaluation. The fluid collection in the abdominal cavity delineates the peritoneal fascia and the anatomical spaces. The white arrow depicts the Zuckerkandl fascia; the white arrowhead shows Gerota’s fascia. The dashed arrow shows  the lateroconal fascia and the black arrow localizes the perirenal space. The white star indicates the lesser sac between the pancreas and the posterior wall of the stomach. The black star is on the pancreatic parenchyma at the level of the body portion. The dashed arrowhead shows a fatty infiltration of the submucosa layer of the descending colon

terior to the esophagus, and inferiorly with the iliac fossa. To better understand the communication between these spaces, a patient with an acute pancreatitis can be considered: in this case there is a typical certain accumulation of abdominal fluid (Fig. 49.7). In this disease the free fluid usually extends posteriorly to the pancreatic tail (in the retromesenteric plane), going around the kidney and in the retrorenal space and running along the anterior lateral wall of the descending colon into the lateroconal plane. Fluid may extend in the iliac fossa and into the pelvis. Perirenal space: the perirenal space contains the kidneys, the adrenal glands, the renal vessels, the renal pelvis, the proximal part of the ureters, and a variable amount of fat. This space is enclosed by the renal fascia. The anterior part of the fascia is also called Gerota’s fascia, and the posterior part is also named Zuckerkandl fascia. Zuckerkandl fascia continues laterally with the lateroconal fascia that is in continuity with the parietal peritoneum. Medially the perirenal space is in communication with the great vessel space. In the perirenal space, there is a great amount of fat tissue containing fibrous septa that can loculate fluid collections. Posterior pararenal space (or retrorenal space): the posterior pararenal space is located laterally to the lateroconal fascia, within the transverse fascia. It does not contain organs, but it is in close communication with the ascending and descending colon’s posterior surfaces. In fact, inflammatory processes that originate from these colonic segments (e.g., diverticulitis, retrocecal appendicitis) have easy access to this space.

49.5.1.2 Abdominal Organs Liver anatomy: the liver occupies most of the abdominal upper right quadrant. It has variable shape and dimensions within the range of normality. The upper, lateral, and front sides are surrounded by the diaphragm. The diaphragm cruxes may, in their costal insertion, indent the liver surface, producing low attenuation or hypoattenuating superficial liver parenchyma defects that must not be confused with intrahepatic solid lesions. The liver comes into contact with the stomach, the duodenum, and the transverse colon medially and with the right colic flexure and the right kidney inferiorly. The upper portion of the right adrenal gland is close to the medial surface of the upper right posterior liver segment. Three liver fissures help to define the margins of liver lobes and major liver segments. The interlobar fissure is an incomplete structure on the lower hepatic margin, passing down the gallbladder fossa and over the middle hepatic vein: it is well visible in some patients, but it can be misidentified in others. The  interlobar fissure forms the lower margin between the left and right hepatic lobes. The left intersegmental fissure (round ligament fissure), which forms a clearly defined sagittal cleavage plane in the caudal surface of the left liver lobe, divides the left lobe into medial and lateral segments. The round ligament, which is generally surrounded by a small portion of fat, starts in the free margin of the falciform ligament. A third fissure, that of the venous ligament (Aranzio’s ligament), is oriented in a coronal or oblique plane between the back face of the lateral left segment of the liver and the front face of the caudate lobe. This fissure, which is in continuity with the intersegmental fissure, contains a portion of the gastro-hepatic ligament (lesser sac). The caudate lobe can be considered an autonomous part of the liver from a functional point of view, because it has its own vascularization compared to the rest of the liver. Liver segmentation is essential to localize and manage liver lesions appropriately. Couinaud nomenclature with Bismuth modification (segment IV is divided into IVa, superior, and IVb, inferior) is widely used to divide liver parenchyma. The most important landmarks are represented by the transverse fissure described by the portal veins that divide the cranial segments from the caudal segments, the hepatic veins (left, middle, and right),

49  Radiological Anatomy with CT: What the Nuclear Physician Should Know When Reading a PET/CT Scan

the round ligament fissure, the interlobar fissure, and the venous ligament fissure. Liver anatomy segmentation based on Couinaud nomenclature with Bismuth modification is shown below:

1251

Gallbladder and biliary tree anatomy: the gallbladder shows variable morphology and position. It is located along the medial profile of the right hepatic lobe caudally to the liver hilum. The density of gallbladder bile content is approximately close to water value but strictly dependent on bile saturation (high concentration or supersaturation leads to an increase of HU up to 25).

In the most common presentation of biliary tree anatomy, the anterior and posterior right intrahepatic segmental ducts converge to form the right hepatic duct. The right hepatic duct joins the left duct to form the common hepatic duct. When the cystic duct converges into the common hepatic duct, the resulting duct is called the common bile duct; it ends, together with the main pancreatic duct, in the papilla of Vater (Fig. 49.10). A large number of variants are described, but the most common type is the posterior segmental duct connecting to the left biliary duct or directly to the common hepatic duct. To obtain an optimal visualization of biliary tree anatomy and pathological alterations in CT multiphasic study, it is important to perform an arterial phase (cholangiocarcinoma shows an early contrast enhancement in arterial phase) and a venous phase. MPR reconstructions are useful to better visualize the long axis of the common bile duct. The physiological diameter of the biliary duct ranges from 1 mm (segmental ducts) to 3 mm (right and left hepatic ducts). The increase in diameter of the common bile duct (normally 6–7  mm) in patients with elevation of direct bilirubin or hepatic transaminases tests is suspicious for pathological findings that are most frequently due to the presence of a wedge calculus in the distal portion of the common bile duct. CT is not the modality of choice to visualize gallbladder or biliary tree stones (cholesterol type) due to the reduced contrast difference between the cholesterol component of the calculus and the bile itself, and the increase in size of the common bile duct diameter is usually the only way to suspect it. The dilatation of the common hepatic duct together with the right and left hepatic ducts is suspicious for pathological involvement of the biliary tree (Klatskin tumor). When the dilatation involves the common bile duct together with the peripheral biliary tree and the common pancreatic

Fig. 49.8  Hepatic segments at CECT in the portal phase. The dashed arrow shows the right hepatic vein, the white arrow depicts the middle hepatic vein, and the black arrow shows the left hepatic vein

Fig. 49.9  Hepatic segments at CECT in the portal phase. The white arrow depicts the middle hepatic vein, the white arrowhead shows the right portal vein, and the dashed arrow shows the right hepatic vein

• Segment I is the caudate lobe situated posteriorly to the venous ligament around the inferior vena cava. • Segments II and III are located in the left hepatic lobe, left of the left hepatic vein. Segment II is cranial and segment III is caudal to the portal plane. • Segment IV is located between the left and middle hepatic veins; segment IV is part of the left liver lobe, and it is divided into segments II and III by the intersegmental fissure. It is subdivided into IVa (superior) and IVb (inferior). • Segments V to VIII are part of the right liver lobe. Segment V is situated at or below the portal plane between the middle and right hepatic veins. The gallbladder makes up the medial edge of segment V. Segment VI is located below the portal plane, right of the right hepatic vein. On caudal CT scanning, segment VI is close to the kidney upper pole. Segment VII is situated at the level of the inferior vena cava, right of the right hepatic vein. Segment VIII is situated at the level of the inferior vena cava between the middle and the right hepatic veins (Figs. 49.8 and 49.9).

1252

duct (“double duct sign”), a cephalopancreatic lesion or a tumor of the papilla should be suspected. Particular attention is needed for patients who have undergone cholecystectomy because they exhibit an increase in diameter of the common bile duct (a cutoff value of 9  mm can be considered paraphysiological). The presence of air in the biliary tree can be observed after surgical treatment, in enteric biliary fistula (gallstone ileus) or in infections. Pancreas: the pancreas is a retroperitoneal gland located behind the stomach, and can be divided into four parts: Head portion with the uncinated process that is strictly connected with the duodenal C loop. The superior mesenteric vessels

Fig. 49.10  The gallbladder, common bile duct, and main pancreatic duct. CECT in the arterial phase. The white arrow depicts the gallbladder with water attenuation of the lumen content and  the  white arrowhead shows the common bile duct. The black arrowhead shows the main pancreatic duct and the black arrow shows a globular discontinuous pattern of hemangioma’s contrast enhancement during the arterial phase

a

Fig. 49.11  Dilation of the main pancreatic duct. CECT in the axial (a) and coronal (b) planes on the portal phase. The white arrowheads depict the dilation of the main pancreatic duct with water attenuation of the

D. Caramella et al.

lie on the left of the pancreatic head from which they are separated by a thin layer of adipose tissue. When a pancreatic lesion is observed in this portion of the pancreas, a fat plane between the tumor and the vessels should be searched for because its presence indicates that the lesion can be considered to be resectable; neck portion, localized anteriorly to the mesenteric vessels; body portion, covered anteriorly by the parietal peritoneum to form the posterior surface of the lesser sac. The proximal part of the splenic vein lies on a groove in the posterior surface of the body and tail portion, in continuity with the splenorenal ligament. The head of the pancreas has  a width of 2  cm, and the mean length from the body to tail ranges from 1 to 2 cm. The density of the parenchyma is variable due to the presence of fatty infiltration processes in relation to the age of the patient, functional alterations of the gland, or chronic diseases. The pancreas has two main ducts, the main pancreatic duct, which drains the dorsal portion and combines with the common bile duct to form the papilla major, and the accessory pancreatic duct (Santorini duct), which drains the ventral portion and forms  the minor duodenal papilla. The diameter of the head portion of the main pancreatic duct ranges from 1 to 3 mm. When a dilatation of the main pancreatic duct is observed, an endoluminal stone or a stenotic lesion should be considered (Fig. 49.11). The most frequent anatomic variant is the pancreas divisum in which the main pancreatic duct ends with the papilla minor and the posterior inferior head portion and uncinated process are drained by an accessory pancreatic duct that ends with the papilla major. Other anatomical variants include the annular pancreas and pancreas portion agenesia. The annular pancreas can be observed as a thickening of the vertical porb

lumen content and the white arrow shows the inferior vena cava. The dashed white arrow shows the portal vein and the white star shows the gallbladder’s lumen

49  Radiological Anatomy with CT: What the Nuclear Physician Should Know When Reading a PET/CT Scan

tion of the duodenum with a density comparable to the pancreas parenchyma. The pancreas portion agenesia (dorsal) is a very rare entity. The contrast enhancement CT study protocol for pancreatic lesions should include an arterial parenchymal phase after 35 s and a venous phase after 60–65 s. Spleen: the spleen is a secondary lymphoid organ localized in the left hypochondrium. The normal  measurements are 12–15 cm in length, 4–8 cm in width, and 3–4 cm in thickness. An increase in diameter or spleen index (the product of the three major measurements is considered normal between 120 and 480 cm) could be suggestive of splenomegaly. On NECT the parenchymal density of the spleen is between 55 and 65 HU, similar to hepatic density (an increase in spleen density can be observed in hemochromatosis due to the presence of iron deposits in reticuloendothelial system). A heterogeneous contrast enhancement of the spleen during an arterial phase is observed, and it can be useful to demonstrate the presence of other accessory spleens or splenosis (Fig. 49.12). On portal venous phase, the spleen parenchyma is homogenous, similar to the liver. Normal and congenital variants include the splenic lobulations (prominent splenic tissue), migrant spleen (most frequent in women due to a laxity of splenic ligament that leads the spleen away from the hypochondrium), accessory spleens (due to the lack in fusion of numerous splenic tissue lobules during embryogenesis), polysplenia (congenital anomaly with multiple little splenic nodules in association with other organ malformations), asplenia, and splenorenal fusion. Kidney: the kidneys are localized in the retroperitoneal space at the level of D10–L2; they are made up of an external cortical portion, an internal medullary portion, and a renal sinus. On NECT the density of the medullary and the cortical portion is quite similar. The renal sinus is a virtual space

Fig. 49.12  The spleen. CECT on the arterial phase. White arrow depicts the heterogeneous contrast enhancement of the spleen’s parenchyma during CECT acquisition on the arterial phase

1253

within the kidney which is occupied by the renal pelvis, renal calyces, blood vessels, and fat. The kidneys have their own renal capsule, and the anterior perirenal fascia envelops the adrenal glands as well. The posterior part of the perirenal fascia is called Zuckerkandl fascia: it continues a­ nterolaterally as the lateroconal fascia and fuses with the parietal peritoneum. The space between the renal fascia and the perirenal fascia is the perirenal space. The perirenal space is made up of  fat tissue and fibrous septa (Fig.  49.13). This fascia becomes more visible in  the case of fluid collection and inflammatory processes involving the peritoneum, as in the case of pancreatitis. The most frequent congenital renal abnormalities are ectopia and fusion defect. The most common ectopia is pelvic ectopia, often associated with rotation defect. The most common renal fusion defect is the horseshoe kidney, in which the lower poles are fused in front of the abdominal aorta. The kidneys are usually vascularized by a renal artery (but accessory renal arteries are not infrequent) and drained by one or more renal veins. To obtain an optimal visualization of the renal artery, especially in the cases of preoperative planning, it is essential to perform an arterial phase on CECT study. A pure arterial phase from 20 to 30  s after contrast injection is mandatory to provide the surgeon with the best visualization of the renal arteries and their possible anatomic variants. Using the common protocol of image acquisition on CECT exam, an arterial parenchymal phase after 35  s (from 25 to 40 s) is useful to visualize the cortical portion that has higher density with respect to the medullary portion. The nephrographic phase starts after 100  s and continues until 120  s: during this phase renal parenchyma contrast

Fig. 49.13  The kidneys on CECT in portal phase. The white arrows depict the renal arteries; the  black arrow shows a renal sinus cystic lesion; the dashed white arrow is on the anterior perirenal fascia; asterisk is at the level of perirenal space; white arrowhead is on the renal pelvis

1254

enhancement is homogenous with a demarcation of the fat of the renal sinus. The excretory phase (from 7 to 10  min) makes it possible  to visualize the renal collecting system, ureters, and bladder. The MIP reconstruction is useful to obtain a 3D map of the collecting system. The NECT is used for evaluation of renal stones. Adrenal glands: the adrenal glands are retroperitoneal organs enveloped by the perirenal fascia. The right adrenal gland lies directly behind the inferior vena cava and between the right liver lobe and the right diaphragmatic dome. The left adrenal gland lies laterally to the abdominal aorta and superior to the ipsilateral renal vein. They are located at the upper pole of the kidneys and are made up of  a medial and lateral arm with a “Y” or “L” shape (Fig. 49.14). During NECT the adrenal glands have a density similar to the liver. The presence of fat tissue inside a nodule on NECT is diagnostic for lipid-rich adenoma. Some other conditions  can mimic adrenal nodules such as:  splenosis in a  patient with previous splenectomy (the typical arterial heterogeneity of splenic parenchyma can help with correct interpretation); splenic artery aneurysm on vein dilatation (a correct artery and venous phase can help to distinguish these two entities); and exophytic gastrointestinal stromal tumor (GIST). Small bowel: the duodenum extends from the pylorus to the duodenum-jejunal junction at the level of the ligament of Treitz (peritoneal fold). It is divided into four parts: the bulb portion (first portion), descending duodenum (second portion), horizontal duodenum (third portion), and ascending duodenum (fourth portion). At the descending portion on the medial side, we can identify the minor papilla cranially (where the accessory pancreatic duct ends) and the major papilla inferiorly (where the common bile duct and the main pancreatic duct combine). The duodenal portions from the first to the third form the duodenal “C” closely linked to the pancreas head. Pancreatic head adenocarcinoma may infiltrate the duode-

Fig. 49.14  The adrenal glands during CECT in the arterial phase. White arrows show the Y shape of the adrenal glands

D. Caramella et al.

nal “C.” The virtual space between the head of the pancreas and the descending part of the duodenum is called the pancreatic duodenal groove. The small bowel (jejunum and ileus) starts from the ligament of Treitz and ends at the ileocecal valve. The folds of the small intestine, more numerous in the jejunum, are called conniventes valves. Normal ileal wall thickness is less than 4  mm. Flogistic processes, secondary, for example, to inflammatory bowel disease, cause significant wall thickening and increase the density of the surrounding adipose tissue. The jejunum and ileum are intraperitoneal organs suspended through small bowel mesenteries, connecting them with the posterior abdominal wall. Large bowel: the colon extends from the ileocecal valve to the anus and consists of the cecum, appendix, ascending colon, transverse colon, descending colon, sigma, and rectum. The cecum lies below the ileocecal valve, caudally has a blind end and cranially continues with the ascending colon. The ileocecal valve occasionally may appear thickened showing a fatty attenuation of the lips; this paraphysiological condition is also called lipomatosis of the ileal valve and should not be confused with other pathological conditions. The cecum is partially covered with the peritoneum with an exception of a small posterior area. The appendix arises from the cecum in variable locations. It has its own specialized mesentery called the  mesoappendix. During a CT examination, it is possible to observe small air bubbles inside the appendix lumen or small calcified deposits called appendicoliths. The ascending colon and descending colon are retroperitoneal. The transverse colon is intraperitoneal and suspended through the transverse mesocolon. The sigma is also intraperitoneal, suspended through its own sigmoid mesentery (also called mesosigmoid). The rectum is located within the pelvis and extends from the rectosigmoid junction, where the sigmoid colon loses his own mesentery, to the anorectal sphincter. The peritoneal pouch, between the upper part of the anterior wall of the rectum and the posterior wall of the bladder in males and posterior wall of the uterus in females, is the lower and most dependent area of the peritoneum. For this reason fluid collections of peritoneal space collect here. The anatomic space between the anterior part of the lower rectum and the posterior part of the bladder and the prostate is not well appreciable on CT exams due to the reduced difference in contrast between these structures and organs. For this reason, the CECT is not accurate in detecting or local staging of prostate or rectal cancer, but it is useful in  advanced disease to localize enlarged lymph nodes (also retroperitoneal) or metastases. The presence of multiple diverticula is a frequent finding in the sigmoid and descending colon. Fat stranding around inflamed diverticula is a typical condition that can be commonly encountered during a CECT exam performed for a suspected diverticulitis.

49  Radiological Anatomy with CT: What the Nuclear Physician Should Know When Reading a PET/CT Scan

The colonic wall has a normal thickness of less than 4 mm and has four layers (mucous, submucosal, muscular, and serous layers). Although these layers are not always recognizable due to the reduced contrast resolution of CT, an acute intestinal inflammatory condition may result in a contrast enhancement of the mucosal layer, or a chronic inflammatory state may result in infiltration of adipose tissue in the submucosa layer. The longitudinal outer muscular layer of the muscolaris tonaca is incomplete and forms the taenia coli. The folds of the colonic wall are called haustra coli.

49.5.2 Pelvic Anatomy Considering the reduced sensitivity of CT exam in the evaluation of pelvic organs and anatomical spaces due to the poor contrast between these structures, only the main organs will be discussed below.

49.5.2.1 Bladder The bladder is a pelvic organ that is part of the excretory system. It is made up of a dome, a body, an apex, and a neck. Axial scans feature an anterior wall that is part of the apex, a posterior wall, and two side walls. The bladder dome (the upper part) is covered by the peritoneum, and the bladder neck (the inferior part) is in close anatomical relationship with the prostate gland. The bladder walls are lined by a trabeculated epithelium (transitional cells) with the exception of the trigone. The trigone is a triangular area in the posterior bladder wall with the apex pointing down towards the bladder neck and from which the urethra originates and a base facing up where the vertices end the ureters. In a CT study, it often appears containing urine with a density similar to water (Fig.  49.15). During the CECT study, a homogenous contrast enhancement of the bladder wall is detected both in the arterial phase and in the portal

Fig. 49.15  The bladder on NECT. White arrow depicts the rectal wall; black star shows the prostate; dashed white arrow is on the posterior perivesical space; white star is at the level of the mesorectal space; black arrow shows the anterior wall of the bladder

1255

phase. Any filling defect observed in the excretory phase (between 5 and 10 min after contrast injection) of the bladder should be considered as a suspicious lesion. The bladder is surrounded by the fat tissue of the perivesical space. The posterior part of the perivesical space encloses the cervix in female and the prostate with seminal vesicles in male.

49.5.2.2 Prostate The prostate is a pelvic organ with pyramidal shape with the apex pointing downwards. It is made up of a base (that is in close contact with the inferior surface of the bladder) and an apex ending at the urogenital diaphragm level. The proximal urethra is surrounded by the prostatic gland. The three portions of the gland (peripheral zone, transitional zone, and central zone) cannot be clearly identified on a CECT examination. The peripheral zone (hypodense) can be differentiated from the transitional zone (hyperdense) by properly adjusting the windows and levels, but the difference in contrast remains too low for the assessment of pathologic prostatic lesions. The CECT exam remains useful for metastatic spread of prostatic cancer. 49.5.2.3 Uterus and Ovaries The uterus is a pelvic organ and it can be divided into a body and a cervical portion. Two other parts can be described: the isthmus that is located at the middle portion of the uterus body (where a slight narrowing can be appreciable) and the fundus that is the upper part of the uterus body. On a coronal plane, the lumen of the uterus can be imagined as a triangular shape with the corners of the base in communication with the fallopian  tubes and the apex in continuity with the cervix. The uterus wall is formed by an inner layer of endometrium, a middle layer of myometrium, and an outer layer of serosa. During a CECT exam, the uterus walls show a heterogeneous contrast enhancement depending on the age and menstrual cycle of the patient. Many anatomical variants of the uterus can be considered, including the arcuate uterus, bicornuate uterus, septate uterus, etc.; CT exam is not the modality of choice to correctly evaluate these malformations, due to low contrast difference between the layers of the uterus walls and the virtual lumen. The ovaries are symmetrical pelvic organs localized laterally to the uterus. The ovaries are connected to the body of the uterus by the ovarian ligaments. The suspensory ovarian ligaments are part of the broad ligament that is a parietal peritoneum fold in continuity with the lateral pelvic wall. During a CECT study, normal ovary appearance mainly depends on the age of the patients. Normal ovaries during a CT study may sometimes be difficult to observe due to their proximity to other anatomical structures that exhibit similar density and therefore difficult to dissociate. To increase the confidence in locating the ovaries during a CT study, some techniques may be used such as: to follow back ovarian veins

1256

Fig. 49.16  The right ovary with mature cystic teratoma in CECT during the portal phase. White arrow shows a fat-containing lesion with a Rokitansky nodule (dashed arrow)

(that are tributary to the inferior vena cava and left kidney vein) to the area where the ovary should lie; to  follow the course of the tubes if observable, ending in close contact with the ovaries; to track the ureter from the renal pelvis to where it becomes the pelvic ureter (the ovaries usually are anterior lateral to this ureter tract); or to look for cystic structures close to the uterus body that represent follicules in women with menstrual cycles. When an expansive ovarian lesion is present, the localization of the ovary is much simpler, but in this case if there are dimensional criteria of alarm related to the patient’s age and the menstrual cycle, it is necessary to undergo a diagnostic workup with MRI that for its multiplanarity and multiparametric characteristics is the gold standard exam for the characterization of such lesions. Mature cystic teratoma is an ovarian lesion that can be diagnosed with CT when there is fatty tissue, calcification, or Rokitansky nodule (Fig. 49.16).

Key Learning Points

• CT is the gold standard imaging technique for the evaluations of the abdominal pathologies. • Knowledge of the peritoneal and extraperitoneal (or retroperitoneal) spaces is important in the peritoneal diseases, ascites, intraperitoneal collections, or peritoneal metastases. • Knowledge of liver anatomy segmentation based on Couinaud nomenclature with Bismuth modification is fundamental to correctly localize liver lesions for presurgical planning. • A large number of variants are described for the gallbladder morphology and biliary tree anatomy. MPR reconstructions are useful to better visualize the long axis of the common bile duct.

D. Caramella et al.

• The pancreas is a retroperitoneal gland. When a dilatation of the main pancreatic duct is observed, an endoluminal stone or a stenotic lesion should be considered. • The spleen is a secondary lymphoid organ localized in the left hypochondrium. The normal  size is 12–15 cm. • The kidneys are localized in the retroperitoneal space. NECT is used for the evaluation of renal stones. To obtain an optimal visualization of the renal artery, especially in the cases of preoperative planning, it is essential to acquire an arterial phase on CECT study. • The adrenal glands are retroperitoneal organs. They are located at the upper pole of the kidneys and are made up of a medial and lateral arm with a “Y” or “L” shape. • The duodenum extends from the pylorus to the duodenum-jejunal junction at the level of the ligament of Treitz. The small bowel (jejunum and ileus) starts from the ligament of Treitz and ends at the ileocecal valve. The folds of the small intestine, more numerous in the jejunum, are called conniventes valves. Normal ileal wall thickness is less than 4 mm. • The colon extends from the ileocecal valve to the anus. The anatomic space between the anterior part of the lower rectum and the posterior part of the bladder and the prostate is not well appreciable on CT exams due to the poor  contrast between these structures. • The bladder is a pelvic organ that is part of the excretory system. It is made up of a dome, a body, an apex, and a neck. Any filling defect in the bladder observed in the excretory phase (between 5 and 10  min after contrast injection) should be considered as a suspicious lesion. • The prostate is a pelvic organ. The CECT exam remains useful for prostate tumor staging for metastatic spread. • The uterus and the ovaries are pelvic organs. To increase the confidence in locating the ovaries during a CT study, the course of the tubes can be observed.

Further Reading Kitapci MT.  Atlas of sectional radiological anatomy for PET/ CT. New York, NY: Springer-Verlag; 2012. Lee JKT, et al. Computed body tomography with MRI correlation (2 volume set). 4th ed. Philadelphia, PA: LWW; 2003. Olivetti L.  Atlas of imaging anatomy. New  York, NY: Springer International Publishing; 2015.

Radiological Anatomy with MR: What the Nuclear Physician Should Know When Reading a PET/MR Scan

50

Davide Caramella and Fabio Chiesa

Contents 50.1

Introduction ������������������������������������������������������������������������������������������������������������������������������  1257

50.2 50.2.1  50.2.2  50.2.3 

MR Imaging of Brain Structures �������������������������������������������������������������������������������������������  1258  R Imaging Appearance of Gray and White Matter �����������������������������������������������������������������  1258 M The Subarachnoid Spaces and the Ventricular System ��������������������������������������������������������������  1259 Signal Characteristics of Normal Blood Vessels ������������������������������������������������������������������������  1261

50.3

MR Imaging of the Neck Compartments�������������������������������������������������������������������������������  1262

50.4  R Imaging of the Chest���������������������������������������������������������������������������������������������������������  1264 M 50.4.1  Lungs and Pleura ������������������������������������������������������������������������������������������������������������������������  1264 50.4.2  Heart and Mediastinum ��������������������������������������������������������������������������������������������������������������  1264 50.5 50.5.1  50.5.2  50.5.3  50.5.4  50.5.5 

MR Imaging of the Abdomen��������������������������������������������������������������������������������������������������  1267  iver and Biliary System������������������������������������������������������������������������������������������������������������  1267 L Pancreas��������������������������������������������������������������������������������������������������������������������������������������  1270 Spleen������������������������������������������������������������������������������������������������������������������������������������������  1270 Kidneys���������������������������������������������������������������������������������������������������������������������������������������  1271 Adrenal Glands���������������������������������������������������������������������������������������������������������������������������  1271

50.6 50.6.1  50.6.2  50.6.3  50.6.4  50.6.5 

MR Imaging of the Pelvis���������������������������������������������������������������������������������������������������������  1272 Uterus������������������������������������������������������������������������������������������������������������������������������������������  1272 Ovaries����������������������������������������������������������������������������������������������������������������������������������������  1273 Urinary Bladder��������������������������������������������������������������������������������������������������������������������������  1274 Prostate����������������������������������������������������������������������������������������������������������������������������������������  1274 Rectum����������������������������������������������������������������������������������������������������������������������������������������  1275

50.7

MR Imaging of the Spine���������������������������������������������������������������������������������������������������������  1276

50.8

MR Imaging of Soft Tissues�����������������������������������������������������������������������������������������������������  1277

Further Reading

Learning Objectives

• Illustrate the most relevant imaging features of various anatomical structures commonly evaluated in MR scans. • Describe the most useful pulse sequences to depict anatomical details of various organs and tissues in different anatomical regions. D. Caramella (*) Diagnostic and Interventional Radiology, Department of Translational Research and Advanced Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy e-mail: [email protected] F. Chiesa Unit of Radiology, ASL 5 “Spezzino”, Sarzana, La Spezia, Italy

 1278

50.1 Introduction Cross-sectional imaging techniques like CT and MRI depict anatomical regions of interest in selected imaging planes. For this reason correct interpretation of cross-sectional images requires deep knowledge of systematic anatomy and of the most relevant topographic relationships between adjacent structures. The image specialist must have a precise three-dimensional mental map of normal anatomy of the body region under consideration, in order to identify anatomical structures and compartments and to correctly define the anatomical location of lesions and pathological processes.

© Springer Nature Switzerland AG 2019 D. Volterrani et al. (eds.), Nuclear Medicine Textbook, https://doi.org/10.1007/978-3-319-95564-3_50

1257

1258

This chapter intends to provide the  basic concepts of radiological anatomy needed when reading MR images in PET/MR hybrid scans, including signal characteristics of normal organs and tissues adequately supported by teaching pictorial material organized by anatomic regions.

50.2 MR Imaging of Brain Structures MR imaging depicts with great detail the anatomy of the brain owing to the superior soft tissue contrast. Differences in signal intensity between brain structures correlate with specific tissue composition, which in turn influences the T1and T2-related relaxation mechanisms. In this section, aspects of brain anatomy that can be helpful in the interpretation of MR images will be discussed.

50.2.1 MR Imaging Appearance of Gray and White Matter The central nervous system is mainly composed of two types of tissues: the gray matter and the white matter. The gray a

Fig. 50.1  Axial view of the brain at the level of the basal ganglia. T1-weighted image (a) and T2-weighted image (b) are shown. Head of the caudate nucleus (c), lentiform nucleus composed of putamen (p)

D. Caramella and F. Chiesa

matter mainly contains the neuronal cell bodies, the dendritic processes, and the axon terminals forming the synapses. The white matter contains the myelin-sheathed axons with variable spatial arrangement in different regions of the neuraxis connecting the different gray matter structures. In the completely myelinated brain, gray matter shows higher signal intensity compared to white matter on T2-weighted images, whereas the opposite is seen on T1-weighted images. The higher T1 signal intensity of white matter mainly correlates with the amount of myelin deposited around axonal processes which is associated with a lower content of free water compared to gray matter. In the cerebral hemispheres gray matter is distributed along the surface of brain gyri constituting the cortical gray matter and in deep nuclei, symmetrical paired structures located in the basal forebrain between lateral ventricles and insular regions, including the basal ganglia and thalami. The basal ganglia are represented by the lentiform nucleus, which is formed by the putamen in its lateral two-­ thirds and the globus pallidus at the medial apex, and the caudate nucleus. The deep gray matter nuclei are separated from each other by the anterior and posterior limbs of the internal capsule (Fig. 50.1). The caudate nucleus, putab

and globus pallidus (gp), and thalamus (t) are separated by internal capsule (dashed line). Left Sylvian fissure (white arrow) and trigone of right lateral ventricle (white arrowhead) are indicated

50  Radiological Anatomy with MR: What the Nuclear Physician Should Know When Reading a PET/MR Scan

men, and thalamus appear isointense to cortical gray matter on all imaging sequences. The globus pallidus typically appears mildly hypointense compared to the putamen on T2-weighted images, an imaging feature that becomes more prominent with aging, likely due to the gradual accumulation of iron-containing compounds. A rather frequent normal finding that can be observed in the region of the basal ganglia is the presence of dilated perivascular spaces appearing as round structures isointense to the cerebrospinal fluid, often located on the inferior aspect of lentiform nucleus. In cerebellar hemispheres, the gray and white matter are arranged to form thin folds called folia oriented parallel to the calvaria. The cerebellar folia are grouped into lobules that are composed by a core of white matter covered by a layer of gray matter with similar signal characteristics as those described for cerebral hemispheres. In the deep portion of cerebellar hemispheres, the dentate nuclei can be recognized on T2-weighted images as mildly hypointense areas located symmetrically near the fourth ventricle, owing to iron deposition (Fig. 50.2). The brainstem is the central axis of the brain and can be subdivided into midbrain, pons, and medulla oblongata; it is

a

Fig. 50.2  Axial view of the brain at the level of the posterior cranial fossa. T1-weighted image (a) and T2-weighted image (b) are shown. Dentate nuclei (dn), the largest of deep cerebellar nuclei, can often be recognized as hypointense paired structures located in the paravermial

1259

located in the posterior cranial fossa between the diencephalon and the spinal cord and is connected to the cerebellum via the cerebellar peduncles. The brainstem has a complex anatomy owing to its intricate architecture composed of gray matter nuclei intermixed with white matter fiber tracts that cannot be completely resolved with MR imaging. Red nucleus and the pars reticulata of the substantia nigra located in the midbrain are frequently recognizable as symmetrical paired areas of low signal intensity on T2-weighted images due to their higher iron content (Fig. 50.3).

50.2.2 The Subarachnoid Spaces and the Ventricular System The subarachnoid space is an anatomical space containing cerebrospinal fluid (CSF) that lies between the meningeal layers pia mater, firmly adhering to the surface of the brain, and arachnoid. Over the cerebral convexity the subarachnoid space fills the sulci between cortical gyri, whereas in the basal region wide separations between the pia mater and the arachnoid form the CSF-filled cisterns including:

b

zones of cerebellar hemispheres (c) on T2-weighted images. Fourth ventricle (white arrowhead) and the basilar artery demonstrating flow-­ related signal void (white arrow) are indicated

1260

a

D. Caramella and F. Chiesa

b

Fig. 50.3  Axial view at the level of the midbrain. T1-weighted image  (a) and T2-weighted image  (b) are shown. Red nucleus (rn) and substantia nigra pars compacta (sn) can often be recognized as hypointense paired structures located in the midbrain on

T2-weighted images. The horizontal segment of the right middle cerebral artery (white arrowhead) normally exhibits low signal intensity due to flow void

• The interpeduncular, prepontine, and premedullary cisterns located along the anterior surface of the brainstem • The suprasellar cistern located above the sella turcica surrounding the hypothalamic infundibulum • The quadrigeminal cistern located posterior to the mesencephalic colliculi between the cerebellar vermis and the splenium of corpus callosum • The cisterna magna, the largest of subarachnoid cisterns situated in the inferior portion of posterior fossa • The cerebellopontine angle cisterns, symmetrical paired subarachnoid spaces located between the cerebellar hemisphere, the pons, and the tentorium • The Sylvian cistern situated over the insular cortex

tricular foramina of Monro (Fig. 50.4). Each lateral ventricle is composed of a body and three horns: the anterior or frontal horn, the posterior or occipital horn, and the inferior or temporal horn. The frontal horn is bounded laterally by the head of nucleus caudatus, medially by the septum pellucidum, and anteriorly by the genus of the corpus callosum. The trigone or atrium of the lateral ventricles is a triangular-shaped area formed by the confluence of the ventricular body, the occipital horn, and the temporal horn (Fig. 50.1). The temporal horn has a close relationship with the hippocampus inferiorly, the tail of the caudate nucleus superiorly, and the choroidal fissure medially. The cerebral aqueduct (of Sylvius) is a thin canal connecting the third ventricle with the fourth ventricle, the latter located in the posterior fossa between the cerebellum and brainstem. The fourth ventricle opens via the median foramen of Magendie in the cisterna magna. The paired foramina of Luschka are lateral apertures connecting the fourth ventricle to the cerebellopontine cisterns. The CSF-filled ventricles and subarachnoid space show the signal characteristics of clear fluid, that shows marked

The brain ventricles are ependymal-lined cavities filled with CSF communicating with each other and with the subarachnoid space through foramina. The ventricular system continues inferiorly with the central spinal canal. The lateral ventricles are paired structures located in the cerebral hemispheres and are  connected to the third ventricle, a median cavity situated in the diencephalon, by the interven-

50  Radiological Anatomy with MR: What the Nuclear Physician Should Know When Reading a PET/MR Scan

a

1261

b

Fig. 50.4  Midsagittal (a) and coronal (b) T2-weighted images of the brain are shown. The brainstem is composed of midbrain (m), pons (p), and medulla oblongata (mo) in a rostrocaudal direction. Cerebellar vermis (v), fourth ventricle (4), and pituitary gland or hypophysis (h) are indicated in a. Lateral ventricles (L) are bounded superiorly by corpus

callosum (c). The foramen of Monro (white arrowhead in b) connects the third ventricle (3) to each lateral ventricle (L). Sylvian aqueduct (black arrowhead) connects the third ventricle (3) with the fourth (4). The right middle cerebral artery in the Sylvian fissure (white arrow in b) exhibits flow-related low signal intensity

hyperintensity on T2-weighted images, low signal intensity on T1-weighted images, and complete signal suppression on FLAIR. In some circumstances an increased signal intensity on FLAIR images may be seen in the subarachnoid space owing to pulsatile CSF flow. This imaging artifact more frequently occurs in the basal cisterns, often in the prepontine cistern and near ventricular openings, where flow phenomena are more prominent.

images. The utilization of angiographic techniques such as TOF (i.e. time of flight) and contrast-enhanced MR imaging allows a more accurate evaluation of vessel anatomy as well as patency.

50.2.3 Signal Characteristics of Normal Blood Vessels Both arterial and venous vessels are well depicted on MR images and demonstrate signal characteristics dependent on the presence of flowing blood. The arterial vessels supplying brain parenchyma can be divided in anterior and posterior circulation, the former originating from internal carotid arteries and the latter from the vertebrobasilar system, together forming the polygon-shaped anastomotic circle of Willis that lies at the base of the brain. The cerebral venous system consists of the dural venous sinuses, the superficial cortical veins, and the deep veins, most notably the vein of Galen and the internal cerebral veins. Normally patent arteries and veins demonstrate a flow-related loss of signal called “flow void,” more consistently seen on T2-weighted images (Figs. 50.1, 50.2, 50.3, and 50.4). Occasionally the presence of turbulent or slow flowing blood in non-occluded arteries or veins may be associated with absence of the expected signal void, which is more frequently seen on T1-weighted

Key Learning Points

• In the completely myelinated brain, gray matter located in the cerebral hemispheres, in both cortical gyri and deep nuclei, cerebellum, and brainstem, shows higher signal intensity compared to white matter on T2-weighted images, whereas the opposite is seen on T1-weighted images. • Globus pallidus, dentate nucleus, red nucleus, and pars reticulata of the substantia nigra often appear hypointense  compared to gray matter on T2-weighted images due to iron deposition. • The ventricular system and subarachnoid space contain cerebrospinal fluid (CSF) that shows marked hyperintensity on T2-weighted images, low signal intensity on T1-weighted images, and complete signal suppression on FLAIR. • Normally patent arteries and veins demonstrate a flow-related signal void more consistently seen on T2-weighted images. Turbulent or slow flow in non-occluded arteries or veins may be associated with absence of the expected signal void more frequently seen on T1-weighted images.

1262

50.3

D. Caramella and F. Chiesa

 R Imaging of the Neck M Compartments •

The neck represents a very complex anatomical region containing intimately associated visceral, vascular, nervous, and muscular structures. From a radiological perspective, it is convenient to describe the neck as composed of different anatomical spaces bounded by the layers of the deep cervical fascia. It is also practical to separate the neck into two contiguous regions relative to the position of the hyoid bone, namely, the suprahyoid and the infrahyoid neck. This approach may prove useful not only to better understand the relationships between different anatomical structures but also to evaluate disease processes based on their location and pattern of spread in neck compartments. Some neck spaces are located exclusively in the suprahyoid or infrahyoid neck, whereas some compartments span across both regions. The suprahyoid neck is made up of (Fig. 50.5): • The sublingual space located deep to the mylohyoid muscle. Its contents include the lingual glands, the ducts of a

Fig. 50.5  Axial T1-weighted image (a) and T2-weighted image (b) of the suprahyoid neck are shown. Neck compartments are delimited by colored dashed lines (yellow, pharyngeal mucosal space; green, parapharyngeal space, light blue, prevertebral portion of the paravertebral space; white, paraspinal portion of the paravertebral space; red, carotid space; purple, masticator space; pink, parotid space). Retropharyngeal

• • •





the submandibular glands, the neurovascular structures, and the extrinsic tongue muscles. The submental and submandibular spaces located superficial to the mylohyoid muscle. The paired submandibular spaces are located lateral to the submental space and contain the submandibular glands. Both neck spaces contain lymph nodes. The buccal and masticator spaces located on each side of the facial region contain the masticator muscles. The pharyngeal mucosal space that is the innermost neck compartment containing the pharyngeal wall. The parapharyngeal spaces, paired compartments located lateral to the pharyngeal mucosal space. They mainly contain fat tissue and neurovascular structures. The superior portion of the retropharyngeal space located posterior to the pharyngeal mucosal space contains fat tissue and lymph nodes. The retropharyngeal space is separated posteriorly from the so-called danger space by a thin fascial layer not appreciable on MR images in normal conditions. The parotid spaces, paired pyramidal-shaped compartments with their apex facing medially toward the

b

space is a virtual compartment located between the pharyngeal mucosal and the prevertebral spaces that may become visible when distended in disease processes (not shown). Pharyngeal lumen (p), left internal carotid artery (ic), left jugular vein (j), left parotid gland (g), and spinal cord (s) are indicated

50  Radiological Anatomy with MR: What the Nuclear Physician Should Know When Reading a PET/MR Scan

­ arapharyngeal spaces. The parotid space contains the p parotid gland and neurovascular structures including the external carotid artery and intraparotid lymph nodes. • The prevertebral and paraspinal portions of the paravertebral spaces that continue in the infrahyoid neck contain neurovascular structures, including the vertebral vessels and part of the brachial plexus, prevertebral, and paraspinal muscles. • The superior portion of the carotid spaces located deep to the sternocleidomastoid muscles, bounded anteriorly by the parapharyngeal spaces and laterally by the parotid spaces. The suprahyoid carotid space contains the internal carotid artery, the jugular vein, and the cranial nerves including the glossopharyngeal nerve, the vagus nerve, the accessory nerve and the hypoglossal nerve, the sympathetic plexus, and lymph nodes. • The superior portion of the posterior cervical spaces, a thin extension of the larger infrahyoid portion located deep to the sternocleidomastoid and trapezius muscles between the carotid spaces anteriorly and the paravertebral spaces medially. The posterior cervical space contains mainly fat tissue, nervous structures, and lymph nodes. The infrahyoid neck is composed of (Fig. 50.6): • The visceral space that is located in the anterior midline region of the neck and contains the larynx, hypopharynx, trachea, esophagus, thyroid and parathyroid glands,

a

Fig. 50.6  Axial T1-weighted image (a) and T2-weighted image (b) of infrahyoid neck are shown. Neck compartments are delimited by colored dashed lines (yellow, visceral space; green, posterior cervical space; light blue, prevertebral portion of the paravertebral space; red,

1263

recurrent laryngeal nerves, and perivisceral lymph nodes. • The anterior cervical spaces located between the visceral space and the carotid spaces, mostly composed of fat tissue. • The inferior and more extensive portion of the posterior cervical spaces that mainly contain abundant fat tissue and lymph nodes. • The inferior portion of the retropharyngeal space that consists of a thin layer of fat tissue behind the esophagus. On T1- and T2-weighed images, fat displays high signal intensity that is useful to differentiate the anatomical structures contained in the various neck spaces, attribute a pathological process to a specific compartment, and evaluate the trans-spatial spread of disease. Muscles show intermediate signal intensity on both T1- and T2-weighted images and are usually separated by variable amounts of fat. Air inside hollow organs appears  as absence of signal in all imaging sequences (Figs. 50.5 and 50.6). Parenchymal organs show variable signal features based on specific tissue compositions. The thyroid gland demonstrates homogeneous signal intensity, slightly higher than muscle on both T1- and T2-weighted images, and moderate homogeneous contrast enhancement after  contrast administration. Salivary glands are composed of a variable admixture of glandular and adipose tissue. Submandibular and sublingual glands usually contain small amounts of intraglandular fat and appear

b

carotid space; white, paraspinal portion of the  paravertebral space). Laryngeal lumen (l), left common carotid artery (c), left jugular vein (j), left vertebral artery in transverse foramen (v), vertebral body (b), and spinal cord (s) are indicated

1264

mildly hyperintense compared to muscles on both T1- and T2-weighted images. Parotid glands are made up of nearly equal proportions of glandular tissue and fat that results in higher signal hyperintensity relative to the submandibular and sublingual glands, occasionally approaching the signal intensity of fat. The glandular tissue of salivary glands shows moderate and homogeneous enhancement after contrast administration. Arterial blood vessels typically demonstrate a signal void in their lumen on both T1- and T2-weighted images resulting from flow-related signal loss, especially when the imaging plane is acquired perpendicular to the vessel direction (Figs. 50.5 and 50.6). Turbulent flow may cause high signal intensity areas more frequently seen on T1-weighted images. Veins demonstrate variable signal depending on blood flow velocity and orientation relative to the imaging plane. On T2-weighted images in the axial plane, the jugular veins frequently appear as signal voids, although high signal intensity secondary to slow flow or sometimes reversed flow may occasionally be found especially on the left side. Veins of smaller caliber may show prevalent high signal intensity on T2-weighted images due to slow flow or demonstrate a central area of signal void surrounded by high signal intensity attributable to laminar flow with reduced flow velocities near the vessel wall and higher flow at the center of the vessel lumen. Contrast-enhanced images are commonly used to depict anatomy and detect pathological changes of neck vascular structures.

Key Learning Points

• From a radiological perspective, it is useful to describe the neck as composed of different anatomical spaces bounded by the layers of the deep cervical fascia and subdivided in suprahyoid and infrahyoid regions. • Fat displays high signal intensity on both T1- and T2-weighed images which is useful to differentiate the anatomical structures contained in neck spaces. • Air inside the lumen of hollow organs appears as signal void. • Thyroid and salivary glands usually appear hyperintense compared to muscles on both T1- and T2-weighted images. • Parotid glands may contain high amounts of fat tissue that results in higher signal intensity relative to submandibular and sublingual glands. • Arterial blood vessels usually exhibit a flow-related signal void on both T1- and T2-weighted images. • Veins may show high signal intensity due to slow or reversed flow.

D. Caramella and F. Chiesa

50.4 MR Imaging of the Chest 50.4.1 Lungs and Pleura The lungs are very difficult to assess with MR imaging owing to their predominant air content which correlates with a low density of signal-generating molecules and generates susceptibility artifacts that tend to cancel air-tissue interfaces. For these reasons anatomic details of the lungs are not well depicted on commonly utilized imaging pulse sequences. Nevertheless, MR imaging may prove useful in evaluating specific lung pathologies mainly in the pediatric population, such as malformations and cystic fibrosis, and is gaining wider applications in assessment of pulmonary masses and airway and vascular diseases by employing adequately tailored imaging protocols. Lung parenchyma normally appears as a signal void on all imaging sequences due to the predominant air content. Pulmonary vascular structures and large-­ caliber central airways may be better displayed on breath-hold three-dimensional T1-weighted gradient echo sequences that may allow to detect even small lung nodules, airway wall thickening, and parenchymal consolidations thanks to the high contrast difference between solid tissue and air spaces. Fast T2-weighted images acquired during breath-hold are complementary to T1-weighted images for assessment of lung pathology, since most disease processes are associated with an increase in water content that appears as high signal intensity; nevertheless the technically achievable spatial resolution and the low proton density of normal lung parenchyma do not allow for accurate depiction  of anatomic details. Evaluation of pulmonary arterial vessels usually requires the utilization of contrast-enhanced images or steady-state free precession sequences (Fig. 50.7), the latter depicting both static fluids and flowing blood with high signal intensity without administration of contrast agents. Normal pleura is made up of a thin membrane not demonstrable in MR images in normal conditions. The extrapleuric fat is well displayed on both T1-weighted and T2-weighted images as a high signal intensity layer that separates pleura surface from muscular chest wall. Even small pleural effusions can be readily detected on T2-weighted images owing to their high signal intensity (Fig. 50.7). Pleural layers may show contrast enhancement in pathological conditions.

50.4.2 Heart and Mediastinum MR imaging is extremely useful to evaluate anatomic structures located in mediastinal compartments owing to its superior soft tissue contrast resolution, although respiratory motion, cardiac pulsations, and blood flowing in large-­ caliber vessels require application of particular imaging

50  Radiological Anatomy with MR: What the Nuclear Physician Should Know When Reading a PET/MR Scan

a

b

c

d

1265

Fig. 50.7  Axial view of the chest at the level of pulmonary arteries. T1-weighted image (a), T2-weighted image (b), steady-state free precession  image (c), and contrast-enhanced image  (d) are shown. Ascending aorta (aa), descending aorta (da), main pulmonary artery (mp), right (rp) and left (lp) pulmonary arteries, and right (rb) and left

(lb) main bronchi are indicated. Flowing blood appears as a signal void in T2-weighted images (b) and shows bright signal on steady-state free precession images (c). In (b) a small pleural effusion along the right major fissure (white arrow) associated with a low signal intensity pleural plaque (white arrowhead) is demonstrated

techniques to obtain artifact-free images. A combination of breath-hold image acquisition, respiratory motion compensation, cardiac gating, and blood signal suppression allows detailed images of the heart, intrathoracic vessels, large airways, and other mediastinal structures to be acquired. The conventional division of mediastinum in prevascular (anterior), visceral (middle), and paravertebral (posterior) compartments utilized in cross-sectional imaging is helpful to characterize mediastinal pathology based on anatomical location. Mediastinal compartments contain  variable amounts of areolar fat-rich connective  tissue showing high signal intensity on T1- and T2-weighted images, that allows to delineate the borders of anatomical structures. The prevas-

cular compartment contains the left brachiocephalic vein, lymph nodes, and variable amounts of thymic tissue depending on age. The visceral compartment contains central large airways, the heart, large blood vessels (including thoracic aorta, intrapericardial pulmonary arteries and superior vena cava), the  esophagus, and lymph nodes. The paravertebral compartment contains the thoracic spine and paravertebral soft tissues (Fig. 50.7). Large airways demonstrate thin walls with intermediate signal in all imaging sequences and an air-containing lumen appearing as signal void. The esophagus is a tubular structure with muscular walls of intermediate signal intensity on all imaging sequences. The blood vessels demonstrate vari-

1266

D. Caramella and F. Chiesa

able signal, depending on the pulse sequences employed, which can be qualitatively divided into two broad categories: dark blood and bright blood imaging. In dark blood imaging, based on turbo spin echo pulse sequences, flowing blood appears a signal void (Fig.  50.7), whereas in bright blood imaging, usually obtained with gradient echo techniques such as steady-state free precession sequences, blood and static fluids both exhibit high signal intensity (Fig. 50.7). Bright blood images can also be acquired with cardiac gating to create dynamic series of images that allow assess-

ment of cardiac wall motion and blood flow. The complex three-dimensional anatomy of the heart can be better evaluated employing adequately oriented planes according to the cardiac axes. The standard cardiac planes include the vertical long axis corresponding to a two-chamber view that displays the left atrium and ventricle, the horizontal long axis that corresponds to a four-chamber view with both atria and ventricles visualized, and the short axis that is acquired perpendicular to the long axis and is commonly used to assess both ventricles (Fig.  50.8). Other commonly acquired imaging

a

b

c

d

Fig. 50.8 Bright blood images depicting cardiac anatomy. Two-­ posterior papillary muscle (pm), interventricular septum (white arrowchamber view (a), four-chamber view (b), short axis view (c), and left head), aortic root (ar), and anterior leaflet of mitral valve (black arrowventricular outflow tract or three-chamber view (d) are shown. Left head) are indicated atrium (la), left ventricle (lv), right atrium (ra), right ventricle (rv), left

50  Radiological Anatomy with MR: What the Nuclear Physician Should Know When Reading a PET/MR Scan

planes are the left ventricular outflow tract or three-­chamber view (Fig. 50.8), the right ventricular outflow tract view, the aortic valve plane, and the oblique sagittal plane for evaluation of  the thoracic aorta. Systemic and pulmonary blood vessels can also be visualized with the utilization of contrastenhanced angiographic techniques, which require the synchronization of image acquisition with the passage of contrast agent inside the vascular lumen.

Key Learning Points

• The lungs are very difficult to assess with MR imaging owing to their predominant air content that results in low signal. • Pulmonary vascular structures, large central airways, and small nodules may be better displayed on three-dimensional T1-weighted gradient echo sequences. • T2-weighted images are useful for assessment of lung pathology, since most disease processes show high signal intensity due to an increase in water content, and for detecting pleural effusions. • Accurate evaluation of pulmonary arteries usually requires the utilization of contrast-enhanced images or steady-state free precession sequences. • Mediastinal compartments contain variable amounts of areolar fat-rich connective tissue showing high signal intensity on both T1- and T2-weighted images that allows to delineate anatomical structures. • Dark blood, bright blood, and contrast-enhanced angiographic images can be used to depict blood vessels. • Bright blood cardiac-gated images acquired in adequately oriented planes allow assessment of cardiac wall motion and blood flow.

50.5 MR Imaging of the Abdomen 50.5.1 Liver and Biliary System The liver is a parenchymal organ located in the upper right and mid-abdomen occasionally extending to the left upper abdomen (Figs. 50.9 and 50.10); it has a convex surface in close relation with the diaphragm and the anterior abdominal wall and a visceral concave surface where the hilar fissure is found. The liver is anatomically separated into the right and left lobes by the falciform ligament; each lobe is further ­subdivided into segments using the portal venous system and hepatic veins as anatomical landmarks. Considering a trans-

1267

verse plane passing through the main portal vein branches, the liver can be conventionally divided into superior and inferior halves. From left to right, the superior part is constituted by segments II, IVa, VIII, and VII, whereas the inferior part is constituted by segments III, IVb, V, and VI. Segments II and III are separated by the left hepatic vein. Segments IVa and IVb are delimited by the falciform ligament on the left and by the middle hepatic vein and gallbladder fossa on the right. The right hepatic vein separates the posterior segments VI and VII from the anterior segments V and VIII. The caudate “lobe” represents segment I and is delimited by the inferior vena cava posteriorly and by the venous ligament anteriorly (Fig. 50.11). The most relevant topographical relationships of the visceral surface of the right lobe are with the right kidney and the hepatic flexure of the colon. The visceral surface of the left lobe lies on top of the small curvature of the stomach. The hepatic veins drain into the tract of the inferior vena cava that passes in a deep groove on the posterior aspect on the liver, partially surrounded by the caudate “lobe” that in fact represents a segment of the right lobe. The biliary system is formed by the confluence of intrahepatic biliary ducts that follow the segmental anatomy of liver parenchyma and merge into the left and right hepatic ducts. The hepatic ducts converge at the hilum to form the common hepatic duct that in turn unites with the cystic duct to form the common bile duct, also known as ductus choledocus; its length is variable depending on the level of insertion of the cystic duct. The common bile duct in most cases joins the main pancreatic duct within the duodenal wall terminating at the major papilla of Vater (Fig. 50.12). The cystic duct drains the gallbladder that can be anatomically divided into the fundus, body, infundibulum, and neck. The normal liver parenchyma appears hyperintense compared to muscle tissue on T1-weighted images. On fat-­ suppressed T1-weighted images, the liver typically shows high signal intensity second only to the pancreas. On dual-­ phase chemical shift imaging, the liver parenchyma may demonstrate variable signal loss in cases of diffuse fatty infiltration. On T2-weighted images, the liver parenchyma exhibits intermediate signal intensity, slightly higher than muscle tissue but significantly lower than spleen (Fig. 50.9). Hepatic vessels usually demonstrate low signal intensity on T1-weighted images and stand out against hyperintense parenchyma. On gradient echo sequences, a phenomenon known as flow-related enhancement may determine a high signal intensity of slow flowing blood in venous vessels, more typically seen in hepatic veins and in the inferior vena cava; this signal may rarely simulate pathology and can be eliminated by using appropriate saturation techniques commonly employed in liver imaging protocols. On T2-weighted images, flowing blood can show variable signal intensity depending on velocity and direction of blood flow relative to the plane of image acquisition, type of pulse

1268

D. Caramella and F. Chiesa

a

b

c

d

Fig. 50.9  Axial view of upper abdomen. T2-weighted image (a), fat-­ saturated T2-weighted image (b), and T1-weighted in-phase image (c) and out-of-phase image (d) are shown. Right lobe (r) and left lobe (l) of

Fig. 50.10  Coronal T2-weighted image of upper abdomen. Liver (l), kidneys (k), spleen (s), and gastric lumen (g) are indicated

the liver, right adrenal gland (a), left kidney (k), pancreas (p), spleen (s), inferior vena cava (c), and gastric lumen (g) are indicated

sequences used, and technical parameters. Veins running parallel to  the imaging plane, such as hepatic and splenic veins, may demonstrate high signal intensity although the inferior vena cava and main portal vein usually appear as flow-related signal voids (Fig. 50.9). Visceral arteries typically demonstrate the flow void phenomenon owing to the high velocity of flowing blood. On diffusion-weighted images, hepatic vessels lose signal due to the bulk movement of blood during the application of diffusion gradients, generating the so-called “black blood” effect (Fig. 50.13). Low b-value images (b-value 3 in the left nasal pledget. The right nasal pledget has a ratio of 1.5 (ratio of 2 is suspicious; >3 suggests a leak), suggesting the presence of a CSF leak. CONCLUSIONS: Probable CSF leak in the left nasal cavity.

51.2.7 123I-FP-CIT Brain SPECT

Ventricles/sulci/ cisterns: Cortex: White matter: Basal ganglia: Cerebellum: Skull/soft tissue/skin:

No abnormal uptake There is a focus of decreased [18F]FDG uptake in the left anterior temporal lobe No abnormal uptake No abnormal uptake No abnormal uptake No abnormal uptake

51  Interpreting and Reporting the Results of Radionuclide Tests

51.2.11 Gated Blood Pool Scan at Rest CLINICAL STATEMENT: A 62-year-old woman with breast cancer treated with doxorubicin. COMPARISON: Gated scan performed on [DATE]: LVEF 60%. CORRELATION: None. RADIOPHARMACEUTICAL: 740  MBq 99mTc-red blood cells (in vitro technique). PROCEDURE: Following reinjection of autologous radiolabeled red blood cells, planar and SPECT EKG-gated images of the cardiac blood pool were recorded. FINDINGS: The heart is of normal size and configuration, with normal left and right ventricular regional wall motion. The left ventricular ejection fraction is 62%. IMPRESSION: Normal resting gated blood pool scan. No significant change from prior examination (>5% change considered significant).

51.2.12 R  est/Redistribution 201Tl-Chloride Scan CLINICAL STATEMENT: A 55-year-old diabetic hypertensive man had transient ST changes and a slightly elevated troponin serum level. COMPARISON: None. CORRELATION: Contrast CT coronary angiography performed on [DATE] revealed 50–70% stenoses in the mid LAD and proximal circumflex. The RCA had no significant narrowings. RADIOPHARMACEUTICAL: 111 MBq 201Tl-chloride. PROCEDURE: Following i.v. administration of the radiopharmaceutical, EKG-gated SPECT images of the heart were recorded about 20 min after injection. A second acquisition was performed 6 h after injection. The patient was fasting (except for water) during the 6 h interval. FINDINGS: The initial images demonstrate a small area of decreased perfusion involving the inferior and lateral walls, with decreased regional wall motion. The delayed images demonstrate improvement in the relative concentration of 201Tl, with persistence of the regional wall motion abnormality. IMPRESSION: Rest ischemia of the inferolateral wall.

51.2.13 Stress/Reinjection 201Tl-Chloride Scan CLINICAL STATEMENT: A 65-year-old diabetic hypertensive man with a strong family history of CAD. COMPARISON: None. CORRELATION: Coronary calcium score of 1500. RADIOPHARMACEUTICAL: 111  MBq  +  37  MBq of 201 Tl-chloride. PROCEDURE: Following intravenous administration of 111 MBq 201Tl-chloride at peak stress, EKG-gated SPECT

1285

images of the heart were recorded. Acquisition commenced about 5 min after injection. When the patient returned about 4  h later, an additional 37  MBq of 201Tl-chloride was injected, and gated SPECT images were recorded about 20 min later. FINDINGS: STRESS: The patient exercised for 8 min on the Bruce protocol, achieving a workload of nine METS.  The resting heart rate and blood pressure were 72 BPM and 105/70  mmHg, respectively. At peak exercise, the heart rate was 145 BPM (92% of predicted maximum), and blood pressure was 200/110 mmHg (rate pressure product 29,000). Exercise was terminated due to fatigue. The patient had no complaints of chest pain during the procedure. EKG will be reported separately. FINDINGS: The post exercise thallium scan SPECT/CT images demonstrated a small focal region of severely decreased thallium uptake in the inferolateral wall, best seen on the vertical long axis images. The lesion was markedly improved on the redistribution images. IMPRESSION: Exercise-induced ischemia of the inferolateral wall.

51.2.14 S  tress/Rest Myocardial Perfusion Scan (1-Day Protocol) CLINICAL STATEMENT: A 65-year-old diabetic hypertensive woman with a strong family history of CAD. COMPARISON: None. CORRELATION: Coronary calcium score 750. RADIOPHARMACEUTICAL: 400 MBq REST + 1200 MBq 99mTc-tetrofosmin [or 99mTc-sestamibi] [STRESS]. PROCEDURE: MODE OF STRESS: [Treadmill/pharmacologic]. STRESS IMAGING: [INTERVAL XX] minutes following injection of 1200  MBq of 99mTc-tetrofosmin, gated SPECT attenuation corrected images of the heart were recorded. REST: Following i.v. administration of 400 MBq 99mTc-­ tetrofosmin at rest, gated SPECT images with attenuation correction of the heart were recorded. FINDINGS: STRESS TEST: The patient exercised for [XX] minutes on the Bruce protocol, achieving a workload of [XX] METS. The resting heart rate and blood pressure were [XX] and [XX], respectively. At peak exercise, the heart rate was [XX] and blood pressure [XX]. Exercise was terminated due to [fatigue/dyspnea/arrhythmia/chest pain]. The patient had [no] complaints during the procedure. SCAN FINDINGS: The post-stress images demonstrate a small area of decreased perfusion involving the anterior wall, with hypokinesis in the region of decreased perfusion. Normal perfusion is seen in the right ventricular myocardium. The rest-injected images demonstrate improvement in

1286

perfusion of the anterior wall, with improvement of the regional wall motion abnormality. The left ventricular chamber is [normal/dilated at stress]. LVEF at stress is [XX]%. IMPRESSION: Exercise-induced ischemia of the anterior wall.

51.2.15 S  tress/Rest Dual Tracer Myocardial Perfusion Scan CLINICAL STATEMENT: A 75-year-old diabetic hypertensive man with a strong family history of CAD. COMPARISON: None. CORRELATION: Coronary calcium score 800. RADIOPHARMACEUTICAL: 201Tl-Chloride 111  MBq i.v. at rest and 99mTc-tetrofosmin [or 99mTc-sestamibi] 1110 MBq i.v. at stress. PROCEDURE: REST: Following intravenous administration of 111 MBq 201 of Tl-chloride at rest, attenuation corrected gated SPECT images of the heart were recorded. MODE OF STRESS: [Treadmill/pharmacologic]. STRESS IMAGING: [XX] minutes following injection of 1110 MBq of 99mTc-tetrofosmin at peak stress, gated SPECT attenuation corrected images of the heart were recorded. FINDINGS: STRESS TEST: The patient exercised for [XX] minutes on the modified Bruce protocol, achieving a workload of [XX] METS. The resting heart rate and blood pressure were [XX] and [XX], respectively. At peak exercise, the heart rate was [XX] and blood pressure [XX]. Exercise was terminated due to [fatigue/dyspnea/chest pain, arrhythmia]. The patient had no complaints during the procedure. SCAN FINDINGS: The post-stress images demonstrate no [small/medium/large] area of [mild/moderate/severe] decreased perfusion [of XX [anterior, inferior, septal, lateral] walls, with [decreased/normal] regional wall motion. Normal perfusion is seen in the right ventricular myocardium. The rest-injected images demonstrate improvement in perfusion of the XX wall, with [normal/hypokinetic/akinetic] regional wall motion. The left ventricular chamber is [normal/dilated]. LVEF at stress is [XX]%. IMPRESSION: [Normal myocardial perfusion scan].

H. W. Strauss and F. Orsini

RADIOPHARMACEUTICAL: 1110 MBq 82Rb-chloride at rest and 1110 MBq 82Rb-chloride at stress (*). PROCEDURE: A prospectively EKG-gated CT was recorded for coronary calcium scoring followed by a low-­ dose CT for attenuation correction. Gated list mode data was recorded for 7  min beginning with infusion of 1110  MBq 82Rb-chloride to determine myocardial perfusion at rest. Pharmacologic stress was performed with dipyridamole, 0.14 μg/kg/min i.v. for 4 min [or regadenoson 0.4 mg i.v. over 30 s]. About 1 min following conclusion of vasodilator administration, a second gated list mode acquisition was recorded for 7 min beginning with infusion of 1110 MBq 82Rb-chloride to determine myocardial perfusion at stress. The stress procedure was supervised by Dr. [XX]. The images and EKG were reviewed with Dr. [YY]. EKG will be reported separately. FINDINGS: STRESS RESPONSE: Baseline HR [XX] and BP [XX]; peaks stress HR [XX] and BP [XX]. The patient was asymptomatic during stress. LEFT VENTRICULAR SIZE: Normal. PERFUSION: Homogeneous/if lesion present give location—anterior inferior septal lateral/describe size (small medium large) and severity (mild, moderate, severe)]. REGIONAL WALL MOTION: Normal. LVEF: [XX]% at rest and [XX]% at stress. CORONARY CALCIUM SCORE: [XX]. Increase in relative risk with increasing coronary artery calcium scores in asymptomatic persons: Calcium score Calcium score Calcium score Calcium score

1–100 100–400 400–1000 >1000

relative risk 1.9 relative risk 4.3 relative risk 7.2 relative risk 10.8

Low risk equals less than 1% per year. Intermediate risk equals 1–2% per year. High risk equals greater than 2% per year. Oudkerk et al. Eur Radiol 2008;18:2785–2807. IMPRESSION: No evidence of myocardial ischemia or scar. (*) Before using the 82Sr/82Rb generator, the daily QC and 82 Sr breakthrough procedures must be performed and documented.

51.2.16 82Rb-Chloride PET/CT Myocardial Perfusion Scan

51.2.17 [18F]FDG PET/CT

CLINICAL STATEMENT: A 58-year-old man with cardiac risk factors of hypertension, hyperlipidemia, diabetes, and known coronary artery disease. COMPARISON: None. CORRELATION: None.

CLINICAL STATEMENT: A 39-year-old woman with fever of unknown origin. COMPARISON: None. CORRELATION: Contemporaneous CT scan. RADIOPHARMACEUTICAL: 240 MBq [18F]FDG.

51  Interpreting and Reporting the Results of Radionuclide Tests

1287

PROCEDURE: Following i.v. administration of [18F] FDG and a waiting time of 70 min, images were recorded from the mid skull to the mid thigh on the PET/CT scanner [Manufacturer: XXX; Model: YYY]. Tissue attenuation correction was obtained through low-dose CT. No intravenous contrast was administered. Plasma glucose at the time of injection was 86 mg/dL. The standardized uptake values are normalized to the patient’s body weight and indicate the highest activity concentration (SUVmax) at a given site. FINDINGS: HEAD and NECK: Physiologic uptake in the brain and large salivary glands. CHEST: Intense tracer uptake at wall of ascending aorta, aortic arch, and thoracic aorta (SUVmax 9.2). Physiologic [18F] FDG uptake is seen in the myocardium and residual blood pool. LUNGS: No abnormal uptake in the lungs. ABDOMEN and PELVIS: Physiologic uptake is present in the bowel, liver, and spleen, and excreted activity is seen in the urinary bladder. BONES and SOFT TISSUES: No abnormal uptake. IMPRESSION: [18F]FDG uptake of the wall of ascending aorta, aortic arch, and thoracic aorta likely due to large-­ vessel vasculitis.

and right-side pleuritic chest pain, for evaluation of possible pulmonary embolism. COMPARISON: None. CORRELATION: Portable chest X-ray performed [DATE]. PROCEDURE: Following inhalation of aerosol from a reservoir containing 1670 MBq of 99mTc-DTPA, ventilation images of the chest were recorded in the anterior, posterior, as well as left and right anterior and posterior oblique views. Following i.v. injection of [XX] MBq 99mTc-MAA, images of the chest were recorded in the anterior, posterior, as well as left and right anterior and posterior oblique views. FINDINGS: PERFUSION: There are multiple segmental regions of decreased perfusion involving the right middle lobe and right lower lobe seen on the planar images. VENTILATION: Ventilation is normal. CHEST X-RAY: The chest film demonstrates well-­ aerated lung in the areas of decreased perfusion. IMPRESSION: High probability of multiple pulmonary emboli.

51.2.18 Perfusion SPECT/CT Lung Scan

CLINICAL STATEMENT: A 70-year-old woman with right lower lobe lung mass, for possible resection. COMPARISON: None. CORRELATION: None. PROCEDURE: Following inhalation of an aerosol from a reservoir containing 1300 MBq of 99mTc-DTPA, images of the chest were recorded in the anterior, posterior, as well as left and right lateral views. Following intravenous i.v. of 160  MBq 99mTc-MAA, images of the chest were recorded in the anterior, posterior, as well as left and right lateral views. FINDINGS: Quantitative analysis results:

CLINICAL STATEMENT: A 62-year-old man with metastatic prostate cancer and sudden onset of left pleuritic chest pain, for evaluation of possible pulmonary embolism. COMPARISON: None. CORRELATION: CT scan of the chest performed [DATE]. RADIOPHARMACEUTICAL: 150 MBq 99mTc-MAA. PROCEDURE: Following non-contrast low-dose CT, 99m Tc-MAA was injected intravenously. SPECT images were obtained. FINDINGS: SPECT/CT: There are multiple segmental regions of decreased perfusion involving the right middle lobe and left lower lobe seen on both the planar and SPECT images. The CT scan demonstrates well-aerated lung in these regions. IMPRESSION: High probability of pulmonary embolism.

51.2.19 Ventilation/Perfusion Lung Scan CLINICAL STATEMENT: A 50-year-old woman with a history of metastatic breast cancer had sudden onset of dyspnea

51.2.20 Quantitative Ventilation and Perfusion Lung Scan

PERFUSION: RIGHT UPPER LUNG [XX%] RIGHT MID LUNG [XX%] RIGHT LOWER LUNG [XX%] RIGHT LUNG TOTAL [XX%] VENTILATION: RIGHT UPPER LUNG [XX%] RIGHT MID LUNG [XX%] RIGHT LOWER LUNG [XX%] RIGHT LUNG TOTAL [XX%]

LEFT UPPER LUNG [XX%] LEFT MID LUNG [XX%] LEFT LOWER LUNG [XX%] LEFT LUNG TOTAL [XX%] LEFT UPPER LUNG [XX%] LEFT MID LUNG [XX%] LEFT LOWER LUNG [XX%] LEFT LUNG TOTAL [XX%]

IMPRESSION: Quantitative lung scan as described above.

1288

H. W. Strauss and F. Orsini

51.2.21 Bone Scan

51.2.23 Labeled Autologous Leukocyte Scan

CLINICAL STATEMENT: A 60-year-old woman with breast cancer and recent onset of back pain. COMPARISON: No prior scan available. CORRELATION: CT of the chest, abdomen, and pelvis performed [DATE]. RADIOPHARMACEUTICAL: 740 MBq 99mTc-MDP. PROCEDURE: Approximately 4 h after i.v. injection of the radiopharmaceutical, total body images, supplemented by spot views of the head and chest (SPECT/CT), were recorded. FINDINGS: SKULL: Symmetric uptake in the skull with no focal lesions. Activity in the paranasal region, probably due to inflammation. Spine, ribs, sternum, and clavicles: Focal lesions involving T5, T7, and T9 vertebrae and the seventh, eighth, and ninth right lateral ribs. These lesions correlate with lytic findings on the CT of [DATE]. Pelvis: No abnormal uptake. Extremities: Increased uptake in the shoulders, knees, ankles, and feet, probably degenerative. Soft tissue: No abnormal uptake. Kidneys: Both kidneys are seen. There is excreted activity in the urinary bladder. IMPRESSION: Focal lesions in thoracic vertebrae and ribs suggest osseous metastases.

CLINICAL STATEMENT: A 62-year-old woman with suspected prosthetic joint infection of the left hip (implanted 4 years ago). COMPARISON: No prior scan available. CORRELATION: CT of the hip and three-phase bone scan performed [DATE]. RADIOPHARMACEUTICAL: 200 MBq 99mTc-HMPAO-­ labeled autologous leukocytes. PROCEDURE: Approximately 30 min and 6 h after intravenous reinjection of labeled autologous leukocytes, total body images, supplemented by spot views of the pelvis and hip (SPECT/CT), were recorded. FINDINGS: Markedly increased uptake of the superior region of the acetabular component and moderate focal uptake of the trochanteric region. The intensity and extent of WBC uptake has increased since the previous examination, suggesting progression of the infection. IMPRESSION: Progression of acute infection of left hip prosthesis.

51.2.22 Three-Phase Bone Scan CLINICAL STATEMENT: A 50-year-old man with a history of compound fracture of the left humerus with persistent drainage, low-grade fever, and swelling of the left arm. COMPARISON: None. CORRELATION: Plain film of the left humerus. RADIOPHARMACEUTICAL: 740 99mTc-MDP. PROCEDURE: Serial images of the left forearm were recorded immediately after i.v. injection of the radiopharmaceutical. Approximately 4  h later, additional spot views of the left forearm and total body images were recorded. FINDINGS: Markedly increased perfusion, increased accumulation on the immediate static images, as well as focal uptake on the 4 h images of the left mid humerus. The remainder of the tracer distribution is physiologic. IMPRESSION: Probable osteomyelitis of left mid humerus.

51.2.24 18F-Fluoride PET/CT Bone Scan CLINICAL STATEMENT: A 42-year-old man with lung cancer. COMPARISON: None. CORRELATION: Contemporaneous CT scan. RADIOPHARMACEUTICAL: 222 MBq 18F-fluoride. PROCEDURE: Following i.v. administration of the radiopharmaceutical and a waiting time of 35  min, whole-body images were recorded on the PET/CT scanner [Manufacturer: XXX; Model: YYY]. Tissue attenuation correction was obtained through low-dose CT. No intravenous contrast was administered. FINDINGS: Presence of multiple focal areas of intense uptake in the axial skeleton, involving the spine (C5–C7, D4, D8, D11–L3), several ribs bilaterally, the left clavicle, and the sternum; additional sclerotic lesions are also evident in the appendicular skeleton (left humeral head and both femora). Moderately increased uptake is also observed in the shoulders, knees, ankles, and feet, probably degenerative. IMPRESSION: Multiple bone metastases of both axial and appendicular skeleton.

51  Interpreting and Reporting the Results of Radionuclide Tests

51.2.25 153Sm-EDTMP Therapy for Intractable Bone Pain CLINICAL STATEMENT: A 66-year-old man with prostate cancer and severe persistent bone pain is referred for radionuclide therapy. After review of bone scan performed [DATE], CBC [DATE], and serum creatinine [DATE]. Radiopharmaceutical: 2590  MBq 153Sm-EDTMP administered intravenously [37 MBq/kg body weight is the standard dose]. PROCEDURE: The radiopharmaceutical was administered intravenously over about 2  min. The radiation safety officer (RSO) was present at the time of administration. Prior to 153Sm-EDTMP administration, the RSO had described the procedure, appropriate radiation safety precautions, and the requirement for good hydration to both the patient and immediate family members residing in the home with the patient. FINDINGS: The patient tolerated the dose without immediate side effects. The 153Sm-EDTMP bone scan is scheduled for 1 week from the date of therapy [DATE].

51.2.26 [18F]FDG PET/CT CLINICAL STATEMENT: A 55-year-old man with lymphoma, for restaging. COMPARISON: Scan performed [DATE]. CORRELATION: Contemporaneous CT scan. RADIOPHARMACEUTICAL: 210 MBq [18F]FDG. PROCEDURE: Following i.v. administration of [18F]FDG and a waiting time of 60 min, images were recorded from the mid skull to the mid thigh on the PET/CT scanner [Manufacturer: XXX; Model: YYY]. Oral contrast was administered during the uptake interval. Plasma glucose at the time of injection was 103 mg/dL. The standardized uptake values are normalized to the patient’s body weight and indicate the highest activity concentration (SUVmax) at a given site. FINDINGS: HEAD/FACE: Physiologic uptake in the brain and large salivary glands. NECK: There is no abnormal uptake in the thyroid. CHEST: Intense focal uptake is present in both axillae, SUVmax 15.2 (previously 12.5). The uptake sites correspond to enlarged (3  ×  4  cm) axillary lymph nodes seen on

1289

CT. Physiologic [18F]FDG uptake is seen in the myocardium and residual blood pool. LUNGS: No abnormal uptake. PLEURA/PERICARDIUM: No abnormal uptake. THORACIC NODES: No abnormal uptake. ABDOMEN and PELVIS: The kidneys are visualized, and excreted activity is present in the urinary bladder. Physiologic uptake is present in the bowel, liver (average SUV 2.5), and spleen. BONES and SOFT TISSUES: No abnormal uptake. IMPRESSION: Focal uptake in axillary nodes, unchanged, is likely due to lymphoma.

51.2.27 Thyroid Uptake and Scan [123I-Iodide] CLINICAL STATEMENT: A 50-year-old woman with acute onset of weight loss, tachycardia, and extreme nervousness, TSH