Thieme: Imaging of the Temporal Bone

(1)

(2)

Imaging of the Temporal Bone

Fourth Edition

(3)

(4)

Imaging of the Temporal Bone

Fourth Edition

Joel D. Swartz, MD President

Germantown Imaging Associates Gladwyne, Pennsylvania

Laurie A. Loevner, MD

Professor of Radiology and Otorhinolaryngology—Head and Neck Surgery Department of Radiology

Neuroradiology Section

University of Pennsylvania School of Medicine and Health System Philadelphia, Pennsylvania

Thieme

New York

•

Stuttgart

(5)

Thieme Medical Publishers, Inc.

333 Seventh Ave.

New York, NY 10001

Executive Editor: Timothy Hiscock Editorial Assistant: David Price

Vice President, Production and Electronic Publishing: Anne T. Vinnicombe Production Editor: Heidi Pongratz, Maryland Composition

Vice President, International Marketing and Sales: Cornelia Schulze Chief Financial Officer: Peter van Woerden

President: Brian D. Scanlan Compositor: Thomson Digital

Printer: The Maple-Vail Book Manufacturing Group Library of Congress Cataloging-in-Publication Data

Imaging of the temporal bone / [edited by] Joel D. Swartz, Laurie A. Loevner.– 4th ed.

p. ; cm.

Rev. ed. of: Imaging of the temporal bone / Joel D. Swartz, H. Ric Harnsberger. 3rd ed. 1998.

Includes bibliographical references and index.

ISBN 978-1-58890-345-7

1. Temporal bone—Imaging. 2. Temporal bone—Diseases—Diagnosis. I. Swartz, Joel D. II. Loevner, Laurie A.

[DNLM: 1. Temporal Bone—radiography. 2. Magnetic Resonance Imaging. 3. Temporal Bone—pathology.

4. Tomography, X-Ray Computed. WE 705 I31 2008]

RF235.S93 2008 617'.514–dc22

2008026874

Copyright © 2009 by Thieme Medical Publishers, Inc. This book, including all parts thereof, is legally protected by copyright. Any use, exploitation, or commercialization outside the narrow limits set by copyright legislation without the publisher's consent is illegal and liable to prosecution. This applies in particular to photostat reproduction, copying, mimeographing or duplication of any kind, translating, preparation of microfilms, and electronic data processing and storage.

Important note:Medical knowledge is ever-changing. As new research and clinical experience broaden our knowledge, changes in treatment and drug therapy may be required. The authors and editors of the material herein have consulted sources believed to be reliable in their efforts to provide information that is complete and in accord with the standards accepted at the time of publication. However, in view of the possibility of human error by the authors, editors, or publisher of the work herein or changes in medical knowledge, neither the authors, editors, nor publisher, nor any other party who has been involved in the preparation of this work, warrants that the information contained herein is in every respect accurate or complete, and they are not responsible for any errors or omissions or for the results obtained from use of such information. Readers are encouraged to confirm the information contained herein with other sources. For example, readers are advised to check the product information sheet included in the package of each drug they plan to administer to be certain that the information contained in this publication is accurate and that changes have not been made in the recommended dose or in the contraindications for administration. This recommendation is of particular importance in connection with new or infrequently used drugs.

Some of the product names, patents, and registered designs referred to in this book are in fact registered trademarks or proprietary names even though specific reference to this fact is not always made in the text. Therefore, the appearance of a name without designation as proprietary is not to be construed as a representation by the publisher that it is in the public domain.

Printed in the United States 5 4 3 2 1

ISBN 978-1-58890-345-7

(6)

To Mrs. Charles Zale Swartz.

—Joel D. Swartz

To Joel Swartz—your passion, persistence, pride, and patience made this important project happen.

To my family, immediate and extended—thanks for your love and support.

—Laurie A. Loevner

(7)

(8)

Preface

Well, it wasn't easy! But then again, very few things that are worthwhile come easily. Losing a renaissance man such as Ric Harnsberger as an editor/contributor would certainly be expected to make any task more difficult, but 10 years between editions was more than we could have possibly anticipated!

Production was complicated by a number of foresee- able and unforeseeable events and was not without high levels of drama and anxiety as well as an obligatory high- wire act. But after all is said and done, this story has a happy ending. We are very proud of this authoritative monograph.

Imaging of the Temporal Bonecontinues to evolve as a comprehensive reference book. The text has been rewritten and expanded throughout, the illustrations to a large extent have been replaced by more cutting edge high resolution CT and MR images, and the bibliography has been extensively updated. The index has been expanded as well and is now on par with other contem- porary reference books. Our main focus is centered on the imaging specialist, but we continue to hope that our clinical colleagues find our contribution of interest and importance as well. The chapter organization remains identical to previous editions. If it's not broken, why fix it?

This edition has substantially more contributors than the previous editions. This was necessitated by a number of factors, not the least of which are the exploding ad- vances in imaging technology, as well as the increasing subspecialization within neuro-otology which results in certain facilities seeing specific types of cases more than others.

Comments from dedicated readers were the driving force behind many of the changes in this fourth edition.

Foremost among these suggestions was the request for the introductory chapter to expand the “cookbook”

approach to evaluating and imaging the temporal bone.

Chapter 1 has accomplished that objective. Paul Caruso was the lead author and he and his colleagues Jennifer Smullen, Robert Liu, Mary Beth Cunane, and Hugh Curtin provided us with a highly detailed contribution useful to radiologists, otolaryngologists, and technologists alike.

Paul was also very helpful by providing us with many images utilized in this book, especially those pertaining to normal anatomy and congenital malformations.

Our good friend, Doug Phillips, spearheaded an outstanding contribution on the facial nerve for Chapter 7 with a very tight deadline and we are deeply indebted to him and his coauthors George Hashisaki and Francis Veillon. Doug was also very helpful to us in procuring a number of images used in this book. The editors also wish to thank Lucianna Ramos Taboada, Maher Abu Eid, and Sophie Riehm for their outstanding contributions.

Mauricio Castillo is a productive neuroradiologist, author, editor, administrator, and friend who took time from his increasingly busy schedule along with lead author Valerie Jewells to produce Chapter 2 on the external auditory canal.

Tim Larson provided considerable help with the postoperative middle ear and mastoid in Chapter 3. His experience and support allowed us to successfully update and expand this important section.

Gul Moonis, Ann Kim, and clinical colleague and friend Douglas Bigelow did a wonderful job with the subject of vascular anatomy and tinnitus in Chapter 4, and our friend Christine Glastonbury provided an outstanding contribution on imaging the cerebellopontine angle and internal auditory canal in Chapter 8. We are also indebted to Deborah Shatzkes and Edwin Wang for their contribution to temporal bone trauma, Chapter 6.

We would like to take this opportunity to thank our superb medical illustrator, Lori Goldstein Motis, for many of the beautiful drawings found throughout this book.

And an enormous thank you to the entire staff at Thieme

(11)

for their support, patience, and hard work in completing this project. And last, but not least, we especially want to thank our families, spouses Nina and Steve, and children Matthew and Laura, Daniel, Chuck, Benjamin, and Alexander. Where would we be without you?

To the readership, we especially thank you for your continued support. We hope that you find the information

and images that follow interesting and educational.

We are greatly interested in any of our readers' comments or suggestions. Please feel free to e-mail us at [email protected] or [email protected].

Joel D. Swartz Laurie A. Loevner

(12)

Douglas Bigelow, MD

Associate Professor of Otorhinolaryngology—Head and Neck Surgery

Department of Otorhinolaryngology—Head and Neck Surgery

University of Pennsylvania School of Medicine Philadelphia, Pennsylvania

Craig Buchman, MD

Associate Professor of Otolaryngology Chief of Otology

Department of Otolaryngology University of North Carolina Chapel Hill, North Carolina

Paul A. Caruso, MD Instructor of Radiology Department of Radiology Harvard Medical School

Massachusetts Eye and Ear Infirmary Boston, Massachusetts

Mauricio Castillo, MD

Professor of Neuroradiology, Section Chief Department of Radiology

University of North Carolina Chapel Hill, North Carolina

Mary Beth Cunnane, MD Instructor of Radiology Department of Radiology Harvard Medical School

Hugh D. Curtin, MD Professor of Radiology Department of Radiology Harvard Medical School

Christine M. Glastonbury, MBBS Associate Professor of Clinical Radiology Department of Radiology

University of California, San Francisco San Francisco, California

George Hashisaki, MD

Associate Professor Otolaryngology—Head and Neck Surgery

Department of Otolaryngology—Head and Neck Surgery University of Virginia Health Systems

Charlottesville, Virginia Valerie L. Jewells, DO

Assistant Professor of Neuroradiology Department of Radiology

University of North Carolina Chapel Hill, North Carolina Ann Kim, MD

Assistant Professor of Radiology Department of Radiology

University of Pennsylvania School of Medicine Philadelphia, Pennsylvania

Robert Liu, PhD

Instructor of Radiologic Physics Department of Radiologic Physics Harvard Medical School

Massachusetts General Hospital Boston, Massachusetts

Contributors

(13)

Laurie A. Loevner, MD

Professor of Radiology and Otorhinolaryngology–Head and Neck Surgery

Department of Radiology Neuroradiology Section

University of Pennsylvania School of Medicine and Health System

Philadelphia, Pennsylvania Gul Moonis, MD

Assistant Professor of Radiology Department of Radiology Harvard Medical School

Beth Israel Deaconess Medical Center Boston, Massachusetts

Suresh K. Mukherji, MD

Professor and Chief of Neuroradiology and Head and Neck Radiology

Professor of Radiology, Otolaryngology–Head Neck Surgery, and Radiation Oncology

University of Michigan Health System Ann Arbor, Michigan

C. Douglas Phillips, MD

Professor of Radiology, Neurosurgery, and Otolaryngology–Head and Neck Surgery Departments of Radiology, Neurosurgery, and

Otolaryngology–Head and Neck Surgery University of Virginia Health Systems Charlottesville, Virginia

Deborah Shatzkes, MD

Associate Professor of Radiology Department of Radiology

Columbia University College of Physicians and Surgeons Director of Head and Neck Imaging

St. Lukes–Roosevelt Hospital Center New York, New York

Jennifer L. Smullen, MD Instructor of Otology Department of Otology Harvard Medical School

Joel D. Swartz, MD President

Germantown Imaging Associates Gladwyne, Pennsylvania

Francis Veillon, MD

Professor of Medical Imaging

Head of Ear, Nose, Throat, Visceral Imaging Service de Radiologie I

Hôpital de Hautepierre Strasbourg Cedex, France Edwin Y. Wang, MD Diagnostic Imaging of Salem Salem, Oregon

(14)

This chapter on the technique for imaging the temporal bone is a practical “how to” written in two parts. In the first part, we explain how to image the temporal bone:

how to run the hardware and obtain the images. We present the imaging modalities and technical parameters routinely used for imaging of the temporal bone. In addition, we provide guidelines for the technologist or radiologist as to how to determine which parameters to enter into the computed tomography (CT) or magnetic resonance imaging (MRI) scanner. In the second part, we explain how to read and report the results of temporal bone imaging studies.

Here we provide a plan of action for the radiologist who, faced with a request for a temporal bone imaging study on a particular patient, must protocol the case, interpret the images, and report the findings in a way that answers the referring clinician’s questions. The major indications for imaging of the temporal bone are thus reviewed, and protocols, interpretive strategies, and template reports are provided for each of these indications.

Imaging Modalities and Technical Parameters

CT and MRI are primarily used for imaging of the temporal bone. We first present the standard technique and protocols most often used, then review the special considerations for both modalities. A brief overview of the roles of plain radiographs, ultrasound (US), positron emission tomography (PET), and PET/CT is given at the end of this section.

Computed Tomography

Routine Technique

The patient is placed supine in the gantry and positioned to place the lens of the eye as far as possible out of the path- way of the x-ray beam to minimize exposure to the lens.

Gantry tilt may need to be avoided to facilitate image reconstruction and reformats. A lateral topogram is then performed. The scan excursion is plotted from the arcuate eminence (the summit of the temporal bone) through the mastoid tip.

The anatomy and pathology of the temporal bone involve small structures; resolution is thus highly important. Collimation is of optimal importance to achieve high resolution.

We routinely use a collimator of 0.6 mm and most commercially available units can be collimated to at least 1 mm. Collimation wider than 1 mm is not usually used, as the resolution is often insufficient.

For 40 to 64 detector scanners, the effective mAs (defined as the mA the gantry cycle time/helical pitch) is adjusted according to the age and head size. Usually, it is 150 effective mAs (CTDIvol[volume CT dose index] 34 mil- ligray [mGy]) for neonates, 200 effective mAs (CTDIvol

45 mGy) for children ages 1 to 10 years, 250 effective mAs (CTDIvol 57 mGy) for adolescents, and 320 (CTDIvol

72 mGy) for adults. The gantry cycle time is set at 1 cycle or gantry rotation/second. The kilovolt peak (kVp) is usually 120.

A helical mode is chosen. Although traditionally, some imagers maintain that nonhelical scans provide better resolution, it is our experience that the difference in resolution between nonhelical and helical acquisitions at thin collimation is not appreciable, and helical acquisitions allow for clearer coronal or oblique reformats and decrease susceptibility to motion artifact.

Intravenous (IV) contrast is usually of the low osmolar type; it is administered by power injector at standard doses of 1 mL/lb to a maximum of 80 to 100 mL for adults.

IV contrast is used for the evaluation of vascular pathology (e.g., dissection, tumors) and may be considered for some types of infections such as coalescent mastoiditis or for the evaluation of abscesses. However, it is not routinely used to evaluate for otomastoiditis or hearing loss.

The raw data from each ear are separated and reconstructed into 0.6 mm (slice thickness) axial images in bone algorithm at a dual field of view (DFOV) of 100 mm that effectively magnifies the images. Then the 0.6 mm images for each ear are brought up on the CT scanner console, where the raw data are displayed in three orthogonal planes. The technologist scrolls through the sagittal data to find an image where the anterior and posterior limbs of the lateral semicircular canal are displayed in cross section (Fig. 1.1). An axial dataset is then made in a plane parallel to the lateral semicircular canal (LSCC).

The technologist “connects the two dots” of the LSCC and

1 Temporal Bone Imaging Technique

Paul A. Caruso, Jennifer L. Smullen, Robert Liu, Mary Beth Cunnane, and Hugh D. Curtin

1

(15)

Fig. 1.1 Making the standard axial and coronal computed tomography (CT) dataset. (A)Once the source data are brought up in the three-dimensional (3D) viewer on the scanner console, the sagittal images are scrolled through until a sagittal image through the anterior and posterior limbs (short white arrows) of the lateral semicircular canal (LSCC) are found. (B)A set of axial images is then generated with

0.1 mm overlap parallel to this plane. (C)If the axial reconstructions have been done correctly, the axial dataset should produce an image where the entire lateral semicircular canal is displayed and, as in (D,E), where the cochlear fossette, modiolus, and stapes footplate are clearly delineated.

A B

C D

(16)

Chapter 1 Temporal Bone Imaging Technique

3

Fig. 1.1 (Continued) (F)The coronal reconstructions are made in a similar fashion. Starting with the sagittal image in (A), a plane perpendicular to the LSCC is established, and the coronal reformats are made along this

plane. (G)This technique should yield a set of coronal images where Prussak’s space (short white arrow) and (H)the facial nerve canal (short white arrow) are clearly delineated.

E F

G H

(17)

makes a 0.6 mm (image thickness) 0.5 (distance between images) axial dataset in this plane parallel to the LSCC;

0.6 0.5 mm coronal images are made in a plane perpendicular to the axial images. The raw data are also reconstructed into 2 mm axial images in soft tissue algorithm to include both ears and the brain at 180–210 mm DFOV.

This protocol generates seven sets of images, three for each ear—the source 0.6 mm images (in a variable axial plane), the 0.6 mm reformats in the axial plane parallel to the LSCC, the 0.6 mm reformats in the coronal plane, and a set of 2 mm axial images in soft tissue algorithm of the entire scan volume.

Multidetector Computed Tomography Reformats Multidetector CT provides shorter acquisition times, a decrease in tube current load, and improved spatial resolution.¹Short acquisition is useful in temporal bone imaging to reduce motion artifact, particularly in children who require sedation or are imaged postprandially without pharmacologic sedation. Although the radiation dose with multidetector scanners in high-quality mode remains an issue compared with single detector scanners, the improved spatial resolution allows for high-quality reformats that essentially obviate the need for rescanning the patient in a second coronal plane.¹

Reformats may, moreover, be obtained in sagittal or oblique planes to improve the detection of pathology in specific clinical settings such as superior semicircular canal dehiscence (SSCD) as discussed below in the section Vertigo and Dizziness.

Stenvers Reformat

Similar to the method explained above, for making the standard axial and coronal images, the 0.6 mm raw data are brought up on the console viewer in three orthogonal planes: axial, coronal, and sagittal. As above, the technologist scrolls through the sagittal plane until a view of the LSCC is obtained represented by the two

“dots” of the anterior and posterior limbs (Fig. 1.2). The axial plane is then established by connecting the two dots. The technologist scrolls through the axial dataset until an image of the summit of the SSC is viewed. The Stenvers reformats are then made by tracing a line perpendicular to the long axis of the summit of the SSC at 0.6 0.5 mm intervals. This plane is effectively perpendicular to the roof of the SSC and displays the roof of the SSC in cross section.

Pöschl Reformat

Similar to the method explained above, for making the standard axial and coronal images, the 0.6 mm raw data

are brought up on the console viewer in three orthogonal planes: axial, coronal, and sagittal. As above, the technologist scrolls through the sagittal plane until a view of the LSCC is obtained represented by the two “dots” of the anterior and posterior limbs (Fig. 1.3). The axial plane is then established by connecting the two dots. The technologist then scrolls through the axial dataset until an image of the summit of the SSC is viewed. The Pöschl reformats are then made by tracing a line parallel to the long axis of the summit of the SSC at 0.6 0.5 mm intervals. The line must be made as parallel as possible to the axis of the summit of the SSC. A slight obliquity may spuriously obscure a dehiscence by volume averaging with the temporal bone on either side of the summit of the SSC.

Computed Tomography Arteriography and Computed Tomography Venography

CT arteriography (CTA) or CT venography (CTV) of the temporal bone may be used to evaluate for tinnitus. At our institution, the standard CT protocol for temporal bone imaging is employed, but the injection rate is increased to 3 to 4 cc per second for CTA. A power injector is employed if a 22-gauge IV or larger is available.

Radiation Dose Reduction Techniques and Considerations for Pediatric Patients

Compared with most radiography procedures, CT exams deliver higher radiation dosages to patients. The quantity CTDIvolis used to describe the patient dose. CTDIvolrepre- sents the average dose in a given scan volume. When a scan is prescribed, the system displays the CTDIvolin mGy on the console. However, the dose displayed is not the true dose for the specific patient under examination. Instead, it is the dose value when the patient is replaced with an acrylic phantom while the same imaging parameters are used. The head phantom is a cylinder with a diameter of 16 cm and a height of 15 cm.

The effective dose E is used to assess the radiation detriment from partial-body as opposed to whole-body irradiation (e.g., irradiation of only the head or only the abdomen). The effective dose is a weighted sum of the doses to all exposed tissues. E (wtHt), where Htis the equivalent dose to a specific tissue and wt is the weight factor representing the relative radiosensitivity of that tissue. The unit of effective dose is sievert (Sv).

The effective dose for a typical CT exam of the temporal bone is 1 mSv (i.e., 1/1000 Sv). In comparison, the average effective dose from cosmic rays, radioisotopes in the soil, radon, and so on, is 3 mSv per year in the

(18)

5

C

A B

Fig. 1.2 How to make a Stenvers reformat. (A)Once the source data are brought up in the three-dimensional viewer on the scanner console, the axial images are scrolled through until an image through the summit of the superior semicircular canal (SSC; short white arrows) is found. (B)The Stenvers reformats are made in a plane perpendicular to the long axis of the SSC, and (C)should yield a cross-sectional view of the SSC (short white arrow).

United States. The effective dose can be estimated from the dose-length product (DLP CTDIvolscan length), which is also displayed on the CT scanner console. The effective dose for a head study in mSv is 0.0021 DLP (mGy cm).²

Radiation Risks The biological effect of radiation is either deterministic or stochastic. The deterministic effect will not occur unless a threshold dose is exceeded. However, the stochastic effects may occur at any dose level, and the probability of occurrence increases with dose linearly

according to the linear nonthreshold dose–response model.³

For CT of the temporal bone, the primary concern for deterministic effect is the dose to the lens. The minimum dose required to produce a progressive cataract is 2 Gy in a single exposure.⁴If the lens is in the direct x-ray beam, the dose to the lens from CT of the temporal bone is in the range of 0.03 to 0.06 Gy, but it could be as high as 0.13 Gy.

If the patient is positioned in such a way that the lens is outside the direct x-ray beam, the dose is in the order of 0.003 Gy.⁵Although the typical dose to the lens from a single

(19)

CT scan is much lower than the threshold value for a cataract, multiple nonoptimized scans in a short time with the lens in the x-ray beam can result in a lens dose close to the threshold. Every effort should be made to keep the lens outside a direct x-ray beam if it is possible.

Stochastic effects include carcinogenesis and the induc- tion of genetic mutations. Children are inherently more sensitive to radiation because they have more dividing cells, and radiation acts on dividing cells. Also, children have more time to express a cancer than do adults.⁶

Factors Influencing the Patient Dose The CT scanning protocols should be optimized such that the quality of images is sufficient for diagnosis and the patient dose is kept as low as reasonably achievable (ALARA). To get the best balance of the image quality and patient dose, it is important to understand the effects of imaging parameters on the dose and imaging quality.

Patient dose depends on three factors: equipment- related factors, patient-related factors, and application- related factors.

C

A B

Fig. 1.3 How to make a Poschl reformat. (A)Once the source data are brought up in the three-dimensional viewer on the scanner console, the axial images are scrolled through until an image through the summit of the superior semicircular canal (SSC) is found. (B)The Poschl reformats are made in a plane perpendicular to the long axis of the SSC, and (C)should yield a view of the entire excursion of the SSC (short white arrows) from anterior to posterior.

(20)

The factors in the first group include x-ray beam filtra- tion, x-ray beam collimation, system geometry, and detector efficiency. Although users do not have control of most of these factors, it is important to understand that the z-axis dose efficiency is reduced when the total x-ray beam width becomes very small for multidetector CT due to the need to keep the beam penumbra out of any detector row.

The dose is strongly dependent on patient size. If the same technique is used to image the heads of an average adult and a newborn, the dose to the newborn is signifi- cantly higher.

Imaging parameters such as kVp, mAs (the product of the tube current and the time in seconds per rotation), and pitch (the table travel per rotation divided by the total x-ray beam width) are selected by the operator.

If all other parameters are fixed, the patient dose is proportional to the effective mAs which is defined as the mAs (mA seconds per rotation) divided by the pitch.

The dependency of dose to kVp is more complicated. In general, the dose increases as a power function of kVp (DkVp^p) if all other parameters are fixed. The value of p is in the order of 2 to 3 depending on the type of scanner.

Image Quality Image quality is characterized by spatial resolution, contrast resolution, image noise, and other quantities. It is difficult to use a single variable to characterize completely the quality of an image. However, in practice, image noise has been widely used to judge the CT image quality because the detectability of low-contrast objects is strongly dependent on the contrast-to-noise ratio. The standard deviation of a region of interest (ROI) in the image is usually used to represent the noise. In CT, for a given reconstruction kernel, the noise is primarily due to the fluctuation of the x-ray photons reaching the detector.

The noise is approximately inversely proportional to the square root of the patient dose. To reduce the noise by a factor of 2, the dose must be increased by a factor of 4. In general, image quality is better when the patient dose is increased.

Strategy for Dose Reduction To optimize the CT technique, the image quality required for the specific indication is assessed based on the radiologist’s experience. The imaging parameters are then selected based on the patient size and organ type under exam such that the required image quality is achieved, while the patient dose is kept as low as possible. Weight- or age-based pediatric protocols should be established and special attention should be paid to children under age 2 because their heads are small and under rapid development.

In general, the technologist and radiologist should keep in mind that

Dose mA, time in seconds per rotation of the gantry, kVp^2.5, and 1/pitch

and adjust these parameters to reduce dose, while main- taining image quality.

Often these adjustments are done empirically, and the technique described above (see Routine Technique section) represents our experience with such adjustments.

Magnetic Resonance Imaging

Routine Technique

The standard MRI protocol for evaluation of the temporal bone in adults is detailed below for a 1.5T (Tesla) magnet.

The patient is placed in the supine position in the head coil.

Sagittal T1-weighted, axial T2-weighted, axial fluid attenuated inversion recovery (FLAIR), and axial diffusion weighted images (DWIs) are obtained through the whole brain.

Axial T1-weighted images are obtained through the temporal bone from the arcuate eminence through the mastoid tip using the following parameters: TR (time to repetition) 300 milliseconds; TE (echo time) 12 milliseconds; flip angle 90 degrees; slice thickness 3 mm;

distance factor 0.10; matrix 192 256 (phase to frequency encoding steps); FOV 180 mm; two acquisitions, one saturation, time 3 minutes, 15 seconds.

Axial CISS (constructive interference in steady state;

Siemens AG, Berlin/Munich, Germany) or 3D (three- dimensional) FIESTA (fast imaging employing steady-state acquisition; General Electric Healthcare, Waukesha, WI) images are obtained through the internal auditory canals and pons using the following parameters: TR 12.25;

TE5.9; flip angle 70 degrees; one slab, slab thickness 32 mm; effective thickness0.7 mm; number of partitions 46; matrix 230 512; FOV 200; swap left (L) to right (R), no saturation, time 4 minutes, 20 seconds.

Gadolinium is then administered.

Axial T1-weighted images are obtained through the whole brain.

Thin section axial T1-weighted images of the temporal bone are performed in two interleaved sets, using the following parameters: TR 450 milliseconds; TE 15 milliseconds; flip angle 90 degrees; slice thickness 2 mm;

distance factor 0.10; matrix 192 256 (phase to frequency encoding steps); FOV 170 mm; two acquisitions, swap L to R, one saturation, time 4 minutes, 20 seconds for each set; total time 8 minutes, 40 seconds.

Coronal T1-weighted images are obtained through the internal auditory canals using the following parameters:

TR450 milliseconds, TE 15 milliseconds, flip angle 90 degrees, slice thickness 3 mm, no gap, matrix 192 256 (phase to frequency encoding steps), FOV 170 mm, three acquisitions, swap L to R, one saturation, time 4 minutes 22 seconds.

7

(21)

Additional Considerations

Coronal high-resolution T1-weighted images may be useful for more detailed imaging of the temporal bone; the indications for the use of this sequence are reviewed in the Indications section. The technical parameters are as follows: TR 528 milliseconds; TE 12 milliseconds; flip angle 90 degrees; slice thickness 3 mm; distance factor 0.20; matrix 338 512 (phase to frequency encoding steps); FOV 200 mm; two acquisitions, swap L to R, no saturation, time 5 minutes, 7 seconds.

Magnetic Resonance Angiography

The imaging parameters for MRA (provided here for a 3T magnet) are TR 25 milliseconds; TE 3.5 milliseconds;

flip angle 20 degrees; slice volume 140; R to L fold-over direction, superior venous saturation band, distance factor 0.10; matrix 496 284; FOV 200 mm;

one acquisition, time 6 minutes.

The most common indication for MRA is in the evaluation for tinnitus. For this indication, MRA is used to evaluate for dural arteriovenous fistulas, aneurysms, vasculopathies such as fibromuscular dysplasia, or arteriovenous malformations (AVMs).

There are two important considerations for MRA in the evaluation of tinnitus: data acquisition and postprocessing.

Suppression of normal venous flow-related signal is key to increase the specificity of the scan for abnormal arterialized flow in the major dural venous sinuses that run along the temporal bone; thus, care should be taken to place the venous saturation band so that normal venous inflow is suppressed. The radiologist should be familiar with the normal artifacts that his or her MRI scanner pro- duces in the dural sinuses on MRA so as to adjust the level of specificity accordingly when interpreting MRAs in patients with pulsatile tinnitus. On some current 3T units, for example, there is essentially no flow-related signal in the dural sinuses when the venous saturation bands are placed appropriately.

For postprocessing, the technologist must provide the entire source dataset to the radiologist for review. Many technologists are trained to MIP (to perform maximum intensity projections of ) only the circle of Willis—the internal carotid artery, middle cerebral artery (MCA), anterior cerebral artery (ACA), posterior cerebral artery (PCA), and basilar artery and to “cut out” the peripheral data. In the evaluation of tinnitus, however, the data on the edge of the scan are the principal data of interest.

Dural arteriovenous formations (AVFs) may occur on the edge of the dataset near the dural sinuses. The source data must be reviewed, and if MIPs are performed, they should include the entire dataset.

Safety Considerations

Standard MRI safety considerations obtain in the temporal bone, and the reader is referred to publications that list the safety of various prostheses.⁷ For those institutions with a busy otology service, the issue of MRI compatibility of stapes prostheses, total ossicular replacement prostheses (TORPs), and partial ossicular replacement prostheses (PORPs) may arise. These prostheses are listed as well in the standard MRI safety references. Most stapes prostheses do not deflect signif- icantly in a 1.5 T unit.

Plain Film Radiography

Plain film radiographs have limited application to imaging of the temporal bone. A plain radiograph in the Stenvers projection, however, may be used for intraoperative or postoperative confirmation of position of a cochlear implant lead. The patient’s head is placed in a 45-degree obliquity contralateral to the implanted ear that places the axis of the implanted temporal bone parallel to the film and then in a 15-degree Townes projection. For example, if the left ear has been implanted, the radiographer would turn the head 45 degrees to the right, with the film behind the head of the patient, and shoot a single radiograph with the beam tilted 15 degrees inferiorly toward the patient (Fig. 1.4). Familiarity with this projection and with the proper position of a cochlear implant is one of the few instances in current imaging where plain film radiography is critical, as the intraoperative assessment is often made while the patient is still anesthetized on the operat- ing room table.

Ultrasound

Ultrasound may be used for evaluation of periauricular cystic lesions such as first pharyngeal arch anomalies or ultrasound-guided biopsies of periauricular lesions.

Positron Emission Tomography (PET) or PET/CT

PET or PET/CT may be used for the assessment of temporal bone masses or nodal metastases.

Referrals and Imaging Strategies

In this part of the chapter, we address the major clinical indications for which a patient may be referred for temporal bone imaging. The indications are listed in alphabetical

(22)

order, and for each clinical indication, the following points are addressed.

1. How toprotocolthe case. You have the requisition in hand. What type of scan do you ask the technologist to do?

2. How tointerpretthe study. An interpretive strategy is provided that first considers the clinical background, then highlights the clinical questions most commonly asked by clinicians. Accordingly, a checklist approach to the images is presented.

3. How to reportthe findings. A report template is provided to help with the dictation.

Acute Otitis Media

Protocol

A routine CT scan of the temporal bone is the standard protocol.

If there is clinical concern for sinus thrombosis or coalescent mastoiditis, the study is performed with IV contrast.

MRI is a more sensitive modality for evaluation of some of the complications of acute otitis media (AOM), such as

suppurative labyrinthitis, facial nerve inflammation, and meningitis, but CT is overall the preferred initial study.

Interpretation Clinical Background

Commonly, patients with AOM present with a bulging ear drum and otalgia and appear sick with fever, malaise, and lethargy. Depending on the level of clinical concern for severity and sequelae of the AOM, the clinician will refer the patient for imaging to evaluate for intratemporal and extratemporal complications of AOM. Long-term antibiotic therapy may be indicated, for instance in the setting of coalescent mastoiditis.⁸

Clinical Questions

• Are there intratemporal (local auricular) complications of AOM, such as coalescent mastoiditis, osseous erosion, facial nerve involvement, or suppurative labyrinthitis?

• Are there extratemporal complications of AOM, such as sigmoid sinus thrombosis, epidural abscess, or meningitis?

Approach

Evaluate for evidence of intratemporal complications of AOM.

• Inspect the osseous margins of the mastoid for evidence of demineralization (bone algorithm images) and for subperiosteal abscess (soft tissue algorithm images) that may reflect a coalescent mastoiditis. It is our experience that asymmetry of the central trabecu- lation of the mastoids is not specific for the diagnosis of coalescent mastoiditis and that demineralization along the external or cisternal walls of the sinus is a more reliable sign.

• Inspect the osseous margins of the facial nerve canal for demineralization that would suggest inflammatory dehiscence. This evaluation is problematic on CT, as the wall of the tympanic segment is naturally papyraceous.

An MRI would be a more sensitive modality for detection of neuritis.

• Evaluate for erosion of the walls of the membranous labyrinth that would suggest a suppurative labyrinthitis. Again, an MRI would be a more sensitive modality for evaluation for enhancement of the labyrinth.

• Evaluate for a middle ear cavity cholesteatoma.

Evaluate for extratemporal sequelae of AOM.

• Inspect the margins of the middle ear cavity (MEC) and mastoid for evidence of epidural or Bezold abscess.

9

Fig. 1.4 A Stenvers radiograph for cochlear implantation. If the Sten- vers radiograph is performed correctly, the cochlear implant lead (short white arrow) should lay coiled in the cochlea, anterior and inferior to the vestibule and lateral and superior semicircular canals (SSCs; summit of SSC marked by long white arrow).

(23)

• Evaluate the transverse and sigmoid sinus for evidence of thrombosis.

• Evaluate the meninges for evidence of meningitis. Here again, an MRI would be a more sensitive modality.

Report

On the right/left, the external auditory canal is severely opacified. The middle ear cavityis completely opacified, and severe periauricular inflammation is noted. There are no findings specific for a coalescent mastoiditis. No erosion is seen along the course of the facial nerve canal or walls of the inner ear. No lobular opacity is seen specific for a cholesteatoma. No extratemporal sequela of acute otitis mediais seen: no Bezold or epidural abscess, sinus thrombosis, or meningitis is seen.

Chronic Otitis Media, Chronic Otomastoiditis

Protocol

A noncontrast CT of the temporal bone is the preferred modality.

Otologists divide patients with chronic otitis media (COM) into three groups:

1. Active COM with or without cholesteatoma

2. Inactive COM with retraction pocket, perforation, ossicular resorption, or fixation

3. Inactive COM with frequent reactivation

The goal of imaging, therefore, is to guide the clinical management (including surgery) in each of these three groups and to evaluate for complications.

There are six questions to answer in evaluating a patient with COM. If the radiologist addresses these issues, what- ever the group of COM, the major clinical questions will have been answered.

• Is there a cholesteatoma, is it confined to the external attic, or has it spread beyond the attic?

• Are there any complications of COM, such as perilym- phatic or horizontal canal fistulas, tegmen dehiscence, or fallopian canal dehiscence?

• Is the mastoid healthy or diseased, and what is the size of the mastoid?

• What is the state of the ossicular chain?

• Is the external auditory canal (EAC) eroded?

• How well aerated is the MEC?

Active Chronic Otitis Media with Cholesteatoma It is important to keep in mind that the radiologist may not sometimes need to make the diagnosis of COM (or sometimes cholesteatoma) if such has been made by the otologist already.

For a patient wth COM and a cholesteatoma, the clinician in his or her consideration of surgical management confers with the radiologist to help answer the following three questions.

• Is the cholesteaoma limited to the attic?

• Is the mastoid well pneumatized and aerated (healthy) and what is the size of the mastoid?

• Is the EAC eroded?

The answers to these questions may lead to three dis- tinct surgical procedures.

If the cholesteatoma is confined to Prussak’s space, and the lateral attic and the mastoid are healthy (well pneumatized and aerated), then an anterior atticotomy may be performed, where the surgeon removes the scutum and lateral attic and leaves the mastoid intact.

If the cholesteatoma extends beyond the attic, and the mastoid is healthy, then a canal wall up (CWU) technique is preferred with a planned second-look procedure.

If the cholesteatoma is beyond the attic, and the mastoid is unhealthy, or if there is erosion of the EAC, then a canal wall down (CWD) technique is preferred.

Other clinical considerations may lead to a CWD technique such as surgical preference, single hearing diseased ear (each surgery carries a 1% risk of sensorineural hearing loss), poor patient compliance, or poor patient health, such that a CWU with second-look would be problematic.

Active Chronic Otitis Media without Cholesteatoma If the patient has active COM without cholesteatoma, then the surgical management depends on the health and size of the mastoid. Hence, the most important question to be addressed:

• Is the mastoid well pneumatized and aerated, and what is the size of the mastoid?

The larger the mastoid, the more problematic a CWD procedure becomes, as it may be difficult to manage a large mastoid cavity, status post CWD, postoperatively, while waiting for the mastoid bowl to reepithelialize.

(24)

On the other hand, the more severe the mastoid granulation tissue and inflammation (reflected by opacity on CT), the more a CWD procedure is indicated.

Thus, the decision between CWU and CWD is a balance between the severity of mastoid granulation tissue and the size of the mastoid. A patient with a severely opacified small sclerotic mastoid would be a good candidate for CWD. A patient with a large, well-aerated mastoid would be a good candidate for CWU.

Inactive Chronic Otitis Media There are two reasons to image in the setting of inactive COM: presurgical planning and to evaluate for complications of COM. In other words, inactive COM is not in itself an indication for surgery, but the sequela of prior COM such as ossicular erosion leading to conductive hearing loss (CHL) may be an indication.

Describing the intactness of the ossicular chain in this setting, for example, will help to guide the tympanoplasty (MEC and/or ossicular reconstruction).

Therefore, for inactive COM, there are two questions to be answered.

• What is the state of the ossicular chain?

• How well aerated is the MEC?

The state of the ossicular chain will guide the choice of ossicular reconstruction. The aeration of the MEC is important for determining the prognosis for hearing outcome after ossicular chain reconstruction.

Inactive Chronic Otitis Media with Frequent Reactivation Inactive COM with frequent reactivation can be managed with CWU or CWD mastoidectomy.

Approach

A lateral-to-medial approach may be used.

• Evaluate the EAC for erosion. If the EAC is eroded in the setting of active COM with or without cholesteatoma, a CWD procedure will be preferred to remove the diseased cells along the EAC.

• Address the issue of cholesteatoma. Are there findings that suggest or raise concern for a cholesteatoma? If so, describe its extent especially with regard to the attic.

• Describe the aeration or opacification of the MEC.

• Evaluate the ossicular chain. If there is CHL in the setting of inactive COM, then the reconstructive procedure (e.g., tympanoplasty, TORP, or PORP) will depend on the intactness of the ossicular chain.

• Evaluate the mastoid for size and pneumatization and for those imaging findings such as sclerosis and opacification that may suggest mastoid inflammation. Here, it is important to recognize the difference between a sclerotic mastoid and an opacified mastoid in other

words, air cells that are inflamed or filled with serous effusion versus those air cells that never developed. If there is active COM with cholesteatoma, then the surgery depends on the health of the mastoid, as above. If there is active COM without cholesteatoma, then a CWU technique is preferred if the mastoid is healthy, and a CWD is preferred if the mastoid is severely diseased.

• Report any surgical landmines. There are three surgical landmarks that must be assessed not only for COM surgery, but whenever a CWD or CWU technique is contemplated:

•• The position of the jugular bulb. As the surgeon approaches the mastoid and MEC from lateral to medial, a jugular bulb that ascends above the floor of the MEC can prove to be a surgical hazard and thus should be reported.

•• The position of the sigmoid sinus with respect to the posterior wall of the EAC. This distance helps to determine the operative window (in the sagittal plane) that may be safely opened in the mastoid.

•• The position of the mastoid tegmen with respect to the roof of the MEC and a low-lying tympanic tegmen.

A low-lying mastoid tegmen can prove to be a surgical hazard and should be reported.

In addition to these considerations, other surgical hazards such as a dehiscent CN VII or horizontal canal fistula should be reported.

Report

On the right/left, the EAC is well aerated, and no erosion is seen. The tympanic membrane (TM) appears thick, which may reflect myringosclerosis. There are no findings specific for cholesteatoma. The middle ear cavity is moderately opacified, which likely reflects granulation tissue in the setting of the reported chronic otitis media. The ossicles are intact.

The mastoid appears small, sclerotic, and completely opaci- fied. The jugular bulb is not high-riding. The sigmoid notch is 15 mm posterior to the posterior wall of the external auditory canal. The mastoid tegmen is in normal position.

CN VII describes a normal course. No inner ear dysplasia is seen. The internal auditory canal appears normal.

In this case, the imaging findings are most consistent with chronic otitis media without cholesteatoma, and the patient would most likely undergo a canal wall down mastoidectomy in light of the severely diseased mastoid.

Cochlear Implantation

Protocol

A standard CT of the temporal bone is prescribed without IV contrast.

11

(25)

If there is clinical concern or CT findings that raise concern for hypoplasia of CN VIII (e.g., stenosis of the cochlear fossette or internal auditory canal [IAC]), a noncontrast MRI of the temporal bone that may be limited to 3D FIESTA, CISS, or DRIVE (driven equilibrium) sequence should be considered for evaluation of CN VIII.

Candidates for cochlear implantation have severe bilateral sensorineural hearing loss (SNHL).⁹ The otologic work-up and audiogram have already identified a defect that involves the hair cells in the organ of Corti within the cochlea, and a decision has been made to consider the patient for an implant. The focus is thus less on diagnosis of the cause of the SNHL and more on the preoperative planning. In general, successful implantation depends on a patent cochlea, an intact CN VIII, and adequate mastoid aeration for access to the facial recess.

• Are there any potential hazards based on the patient’s anatomy that the surgeon may encounter while placing the implant?

Approach

The approach is thus to inspect the temporal bone moving from superficial to deep, as if you were the surgeon placing the cochlear implant electrode and threading it into position, keeping in mind the potential surgical pitfalls.

• Inspect the thickness of the skull 4 to 5 cm posterior and superior to the spine of Henle (posterosuperior margin of the EAC at the osteocartilaginous junction). The implant device consists of a receiver stimulator package and a multichannel electrode. The receiver–stimulator package is placed in a depression drilled on the external table of the skull in this region.⁹If the bone is thin, this finding should be reported; the implant receiver–stimulator may need to be placed against dura, and the dura may need to be exposed intraoperatively.

• Report the pneumatization of the mastoid. If the mastoid is underdeveloped, the drilling will require more effort, and if one ear is better pneumatized than the other, the more pneumatized side may be preferred.

• Evaluate the cisternal face of the petrous temporal bone for arachnoid granulations and for a lateral position of the sigmoid notch that may prove to be surgical obsta- cles. The position of the sigmoid notch may be reported with respect to the posterior wall of the EAC.

• Inspect the facial recess, where the electrode is usually threaded from the mastoid into the tympanic cavity, and evaluate the pneumatization of the recess and describe the position of the facial nerve with respect to the recess. An aberrant or dehiscent nerve or a laterally positioned posterior genu may lead to injury and require an alternate approach to the MEC.

• Inspect the MEC and round window niche, at the site of cochleostomy, for aeration.

• Evaluate the inner ear for evidence of malformation that may reduce the efficacy of the implant or make insertion more challenging. Cochlear malformation may require modification of electrode insertion techniques.

Inner ear malformations may raise the risk of meningitis.

Labyrinthitis ossificans (LO) that may be a sequela of prior meningitis may limit the ability of the surgeon to advance the electrode.^10,11If there is evidence of LO, an MRI should be performed to determine better the caliber of the membranous labyrinth.

• Evaluate the cochlear fossette (bony canal for CN VIII at the base of the cochlea) for evidence of stenosis that may herald a hypoplastic nerve or dysplasia that may increase the risk for a cerebrospinal fluid (CSF) leak.

• Check the otic capsule for evidence of otospongiosis or, for example, Paget disease that may increase the risk of facial nerve stimulation from the electrode.¹²

• Describe the IAC. If the canal appears stenotic, an MRI should be considered to evaluate the intactness of CN VIII, as IAC stenosis may be associated with CN VIII hypoplasia or aplasia, a contraindication to implantation.

• Evaluate the temporal bone for evidence of COM. Active COM is a contraindication to implantation.

• Report any surgical landmines. There are three surgical landmarks that must be assessed:

•• The position of the jugular bulb. As the surgeon approaches the mastoid and MEC from lateral to medial, a jugular bulb that ascends above the floor of the MEC can prove to be a surgical hazard and thus should be reported.

•• The position of the sigmoid sinus with respect to the posterior wall of the EAC. This distance helps to determine the operative window (in the sagittal plane) that may be safely opened in the mastoid.

•• The position of the mastoid tegmen with respect to the roof of the MEC and a low-lying tympanic tegmen.

A low-lying mastoid tegmen can prove to be a surgical hazard and should be reported.

Report

The report summarizes this interpretive superficial-to-deep approach.

(26)

On the right/left, the mastoids are well pneumatized and aerated. The sigmoid notch is in normal position 15 mm posterior to the posterior wall of the external auditory canal. The retrotympanum and facial recess are well aer- ated. The CN VII describes a normal course; there is no evidence of dehiscence along the facial recess. The middle ear cavity is well aerated, including at the standard site of cochleostomy. There are no findings specific for labyrinthitis ossificans. No otospongiosis is seen. No inner ear dysplasia is seen. The internal auditory canal appears normal. The external auditory canal, tympanic membrane, and ossicular chain appear normal.

Impression: Temporal bone anatomy, as detailed above.

The EAC, TM, and ossicular chain are mentioned at the end of the report as they are not the primary considerations for implantation.

Congenital Aural Atresia, External Auditory Canal Stenosis

Protocol

The routine CT temporal bone will answer most questions with regard to preoperative planning for atresiaplasty, which is the most common indication in this group of patients.

There are two protocol decisions, however, that you need to make when you have the requisition in hand, before you check off the routine noncontrast CT temporal bone box.

First, is the patient the correct age for imaging? Radio- logic evaluation is usually deferred until at least 5 years of age because microtia surgery, which is done to coincide with the atresia surgery, usually involves harvesting of the costal cartilage that takes until 5 years of age to mature sufficiently for surgical use.¹³

Second, if it is elected to study the craniofacial bones as well as the temporal bone (see Clinical Background section directly below), the protocol may be modified to include the abnormal bones in question; for a child with mandibulofacial dysostosis, for example, the scan should be carried from the frontal sinuses (or calvaria) through the hyoid bone to include the mandible using a 1 mm (instead of the standard 0.6 mm) collimator. The resolution of the temporal bone images will be less than that of the standard 0.6 mm collimation technique, but allows for adequate evaluation of the temporal bone anatomy, reduces the radiation dose, and allows for 3D modeling of the craniofacial structures.

In patients with bilateral atresia, multistage surgery commences around 6 years of age. In unilateral atresia, if the atresia plate is thin and the oval window and

stapes are normal, surgery may be considered with parental consent. Otherwise, surgery for unilateral conductive hearing loss is usually delayed until the patient can consent.¹³In cases of unilateral hearing loss, surgery carries a risk of SNHL. As long as the opposite ear provides the child with sufficient hearing for educational needs, surgery may be left until the patient is of the age of consent.

Although most cases of congenital aural dysplasia (CAD) are isolated, there are also syndromic cases that may be associated with other craniofacial malformations that may require imaging. At least 30 distinctive hereditary syndromes with hearing loss and abnormalities of the external ear have been described.¹⁴Such syndromes may involve (1) the mandible, such as the mandibulofacial dysostoses, oculo–auriculo–vertebral spectrum, or del22q (DiGeorge syndrome); (2) facial clefts, such as the HMC syndrome (hypertelorism–

microtia–clefting); (3) cervical cysts or fistulas, such as the branchio–oto–renal syndrome; (4) craniosynostoses, such as Apert or Crouzon syndrome; (5) the airway, such as the CHARGE association (a nonrandom pattern of congenital anomalies comprising colobomata, heart defect, choanal atresia, retarded growth and development, genital hypoplasia, and ear abnormalities and/or deafness), or (6) less commonly, anomalies of the lacrimal drainage system, such as the lacrimo–

auriculo–dento–digital (LADD) syndrome. If time per- mits, inspection of the patient or focused interview of the patient or parent may suggest a syndrome, and if so, coordination of the imaging with the oromaxillofacial surgeon, pediatric ear–nose–throat (ENT) physician, or oculoplastic surgeon may save time and spare the patient radiation and further anesthesia.

The surgical questions you will need to answer when patients with CAD are sent for imaging are as follows.

The atresiaplasty surgeon would like to know if the patient’s anatomy is such that the external auditory canal and middle ear can be reconstructed and how difficult the atresiaplasty may prove to be.¹³

• How pneumatized and large are the mastoid and middle ear cavity (i.e., will it be hard to drill the mastoid)?

• Is the facial nerve at risk for injury?

• Are the oval window and stapes normal?

• Are the inner ear structures normal (i.e., what is the risk of acoustic injury to the inner ear during surgery)?

• What is the status of the ossicular chain?

There is also a genetics question associated with imaging a CAD patient.

13

(27)

• Is there an associated syndrome or evidence of other extraauricular malformation that may affect the patient’s prognosis?

The clinical goal is to use the CT to identify radiographic evidence of any associated finding that may aid in syndromic diagnosis. Although often a geneticist will have seen the patient, and the syndromic diagnosis will have been made, a brief organized review of the images may unveil unexpected findings that may make or confirm the diagnosis.

Approach

To answer the surgical questions, the images are reviewed from lateral to medial.

• Describe the pneumatization of the mastoid and middle ear cavity.

• Describe the course of CN VII, paying particular attention to segments that may be aberrant. Also, report the anteroposterior (AP) distance from the posterior margin of the glenoid fossa to the descending portion of the facial nerve because this space is where the surgi- cally created external auditory canal will pass.

• Describe the oval window and stapes.

• Describe the appearance of the inner ear and the IAC. If the cochlear fossette or the IAC is stenotic, an MRI should be considered with a 3D FIESTA, CISS, or DRIVE sequence for better delineation of the nerve.

• Describe any other findings associated with CAD, such as ossicular dysplasias. The malleus and incus are commonly fused, and in nonsyndromic CAD, the stapes is usually normal.

• Describe the thickness of the skull. The option to atresiaplasty is a bone-anchored hearing aid (BAHA), which is placed typically 5 cm behind the atretic canal.

Next, the genetic question is addressed, by doing a review of the extraauricular components of the scan performed from lateral to medial.

• Study the visualized cranial sutures to evaluate for craniosynostosis (e.g., Apert or Crouzon syndrome).

• Evaluate the periauricular soft tissues to search for fistulas or cysts that may be seen in BOR syndrome.

• Inspect the visualized zygomas; zygomatic deficiency is a common finding in first pharyngeal arch syndromes, such as the mandibulofacial dysostoses and the OAV (oculo–auriculo–vertebral) spectrum.

• Inspect the visualized mandible to pick up those syndromic cases that may be associated with mandibular dysplasia, such as the OAV spectrum or mandibulofacial dysostosis.

• Evaluate the orbits for ocular malformations, such as coloboma, that may be seen in CHARGE association or lacrimal drainage system abnormalities that may be seen, for example, in BOR, OAV, or LADD syndromes.

• Evaluate the maxillae and palate for evidence of clefting disorder.

• Check the choanae for choanal stenosis that may be seen, for example, in the CHARGE association.

Report

The report should answer the surgical questions for which the patient was most likely referred in anatomical order from lateral to medial and then address the question of associated syndrome, if there is any evidence thereof. An example of an isolated case of CAD follows.

On the right/left, the mastoid air cells and middle ear cavity are well pneumatized and aerated. The facial nerve describes an anomalous course: the labyrinthine and proxi- mal tympanic segments of CN VII appear normal, but the nerve appears dehiscent along a 5 mm segment as it courses above the oval window and turns laterally into the meso- tympanum. The anteroposterior distance between the mas- toid segment of the facial nerve and the posterior margin of the glenoid measures 7 mm. The mastoid segment is well covered but exits into the posterior aspect of the deficient glenoid fossa. The oval window appears normal. The malleus appears fused to the atresia plate along the lateral wall of the middle ear cavity. The incus is hypoplastic. The stapes appears normal. The middle ear cavity is mildly hypoplastic but well aerated. No inner ear dysplasia is seen.

The internal auditory canal appears normal. The external auditory canal is atretic. The tympanic portion of the temporal bone is deficient. The osseous component of the atresia plate measures 6 mm at the level of the oval window niche.

No other malformation is seen.

Impression: Congential aural dysplasia, detailed above, with well-pneumatized mastoid and middle ear cavity, aberrant facial nerve, and normal-appearing inner ear

Otitis Externa Protocol

A routine noncontrast CT scan is the modality of choice for evaluation for otitis externa (OE) in both immunocom- petent and immunocompromised patients. Most of the relevant clinical questions (see the Interpretation section below) can be answered without IV contrast. Because evaluation of cartilaginous involvement is important in OE, 2 mm coronal reformats in soft tissue algorithm are added to the protocol. Contrast-enhanced MRI may be performed in cases recalcitrant to initial therapy or with

(28)

neurologic symptoms where intracranial infection becomes a consideration.¹⁵

Inflammation of the external auditory canal is usually apparent clinically before the patient is sent to imaging.

A key clinical datum for the clinician and radiologist is a history of immunocompromise because necrotizing external otitis (NEO), otherwise known as malignant otitis externa (MOE), occurs classically in this group of patients (e.g., elderly diabetic patients or patients with HIV).¹⁶ Clinical Questions

• Is there evidence of NEO/MOE or otitis externa? For example, is there bony involvement or inflammation that extends into the soft tissues inferior to the EAC that would suggest NEO?

• What is the extent and severity of infection?

• Are there any sequelae of infection?

• Is there an identifiable nidus or cause of infection?

One sound interpretive strategy is to begin with the EAC and to move from lateral to medial, to assess the ear for spread of inflammation.

Approach

• Inspect the auricle and periauricular soft tissues for chondritis or cellulitis. Inflammation extending inferior to the EAC may indicate NEO.

• Assess the severity of opacification of the EAC. The coronal soft tissue images are key in evaluating for cartilage involvement that is the site of origin of NEO. In external otitis, the canal may stenose, and evaluation of the aerated lumen, following treatment of the acute infection, may be a preoperative question for canalplasty.

The canal is inspected as well for a foreign body that may occasionally be at the origin of the inflammation.¹⁷

• Describe the degree of opacification of the middle ear cavity (MEC) and mastoid, as these may be inflamed as well. It is important to keep in mind that it is not unusual to have opacification of some middle ear air cells in the setting of EO and that otomastoid opacification alone does not indicate NEO.

• Scrutinize the bone for erosion, especially the tympanic and mastoid components of the temporal bone. Bony erosion suggests NEO and may indicate the need for long-term parenteral antibiotics. If there is evidence of skull base erosion and osteomyelitis, an MRI should be considered to evaluate for intracranial involvement, for example, an abscess or sinus thrombosis.¹⁵

Report

The report should answer the pertinent clinical questions with regard to severity and extent of infection and any identifiable cause.

On the right/left, the external auditory canal is nearly completely opacified, consistent with the reported otitis externa. The cartilaginous margin of the external auditory canal appears blurred. There is no evidence of erosion of the external auditory canal or mastoid. The middle ear cavity and mastoid are only mildly opacified. No auricular inflam- mation is defined. There is mild reticulation of the periauric- ular soft tissues, consistent with cellulitis. No osseous erosion is seen. The ossicles appear normal. CN VII describes a normal course. The inner ear and internal auditory canal appear normal.

No abnormal-appearing lymph nodes are seen.

Impression: Findings consistent with right otitis externa with severe opacification of the external auditory canal and mild cellulitis

Facial Nerve Disorders

Protocol

For atypical Bell’s palsy, multiple cranial neuropathies, or facial nerve paresis of uncertain etiology, MRI of the facial nerve is the preferred modality. Atypical Bell’s palsy is a facial nerve palsy that has either slow onset (progresses over weeks or months rather than days), incomplete recovery (failure to recover in 6 months), or recurs. A high-resolution coronal postgadolinium T1-weighted sequence through the course of the facial nerve and pregadolinium axial T1- and T2-weighted sequences through the parotid gland should be added to the standard temporal bone MRI protocol. MRI provides superior imaging of the facial nerve nucleus and cisternal segments of the nerve that are not well imaged by CT (Fig. 1.5).

In the setting of trauma, a noncontrast standard CT scan of the temporal bone is the preferred modality.

Most patients sent for imaging of the facial nerve fall into two categories: atypical “Bell’s palsy” or posttraumatic facial nerve paresis.

Bell’s palsy accounts for 80% of peripheral facial nerve paresis. Certain clinical features of CN VII paresis, however, may prompt the clinician to question a diagnosis of Bell’s palsy, such as recurrent episodes of paresis, slow progressive course, symptoms referable to other cranial nerves,

15

(29)

twitching or spasm that may suggest vascular compression, or failure of return of function within 6 months.¹⁸

In the posttraumatic setting, the differential diagnosis includes facial nerve interruption and posttraumatic facial nerve swelling with compression. The onset of paresis can help differentiate between these two, and interruption should lead to immediate paresis, whereas facial nerve swelling is more subacute. Coexisting injuries may pre- vent a timely examination in the trauma patient, and the

clinician may turn to CT for diagnosis and localization of the suspected facial nerve injury.

The primary clinical question in the case of atypical Bell’s palsy is straightforward.

• Is there a lesion along the course of CN VII?

Fig. 1.5 The normal appearance of CN VII on DRIVE (driven equilibrium) sequence versus hypoplasia of CN VII. (A)An axial CISS (constructive interference in steady state) image shows the normal-appearing CN VII in the right internal auditory canal (IAC; short arrow), but CSF signal within the superoanterior quadrant of the left IAC. (B)A sagittal oblique reformat obtained in a plane perpendicular to the right IAC that shows the facial nerve (short white arrow), vestibular, and cochlear nerves in cross section. (C)A sagittal oblique reformat along the left IAC where the facial nerve is not defined, consistent with aplasia or severe hypoplasia.

A

C

B

Thieme: Imaging of the Temporal Bone

Imaging of the Temporal Bone

Fourth Edition

Imaging of the Temporal Bone

Fourth Edition

Thieme

New York

Stuttgart

Contents

Preface

Contributors

Imaging Modalities and Technical Parameters

Computed Tomography

1 Temporal Bone Imaging Technique

1

3

Radiation Dose Reduction Techniques and Considerations for Pediatric Patients

5

Magnetic Resonance Imaging

7

Plain Film Radiography

Ultrasound

Positron Emission Tomography (PET) or PET/CT

Referrals and Imaging Strategies

Acute Otitis Media

9

Chronic Otitis Media, Chronic Otomastoiditis

Cochlear Implantation

11

Congenital Aural Atresia, External Auditory Canal Stenosis

13

Facial Nerve Disorders

15