Page 1


FEATURES

Journal of Surgical Radiology

Operative surgery, which was formerly in the hands of a few privileged persons, is becoming more and more the common property of medical men. Theodor Kocher Textbook of Operative Surgery, 1903

104 | Portal Venous Gas COLUMN

Diana L. Diesen and Dan G. Blazer

10 | Social Networking 14 | Trauma Systems

CASES

A 61-year-old male with hematuria and a pancreatic tail mass.

102 | Stanford University Hospital

Stanley W. Ashley

A 61-year-old male with chest and back pain following blunt trauma.

106 | Transient Ischemic Attack REVIEW

J Surg Rad

SurgRad.com

James M. Hurst

The widespread availability if advanced imaging technology, in even small community hospitals, has resulted in significant delays in transport to definitive care facilities.

100 | Brigham and Women's Hospital

www.

Shahid Shah

The next big revolution in communication between patients and physicians. Why social networking via Facebook, Twitter, and more counts.

Ralph Greco

Robert B. Rutherford

Emeritus Professor of Surgery, University of Colorado School of Medicine, Denver, Colorado.

108 | May-Thurner Syndrome

Charles Y. Kim

Assistant Professor of Radiology, Duke University Medical Center, Durham, North Carolina. Contents

1


CONTENTS

COVER and Contents: MedicalRF

Advertisement 1 January 2011

| 58

Intraoperative Imaging and NeuroNavigation Mark Winder et al.

08 Editorial

Welcome to our world-wide classroom.

Cynthia Shortell

19 Technology The iPad in the Hospital and Operating Room

} 68

Unrecognized Spinal Instability

Jonathan Forbes et al.

A sample workflow for more efficient access to patient data in a clinical setting.

Felasfa Wodajo

| 76

24 Original Article Diagnosis of Upper GI Bleeding

Erin Moody et al.

A comprehensive review of upper GI and small bowel bleeding.

Occipitoatlantal Dislocation in Trauma

Leslie Kobayashi et al.

42 Original Article Marathon-Induced Colitis

} 80

Transected Horseshoe Kidney

Incidence, pathophysiology, diagnosis, and management of this previously unrecognized entity.

Ben Paxton et al.

David King and Laura Avery

| 86

Nipple Adenoma

J Surg Rad

Benzon Dy et al.

2

January 2011

50 Original Article Stents in Tents The nature and management of major vascular injuries sustained in the global war on terror.

Joshua O'Brien and Mitchell Cox

} 92

Molecular Breast Imaging

Dietlind WahnerRoedler et al.

110 Future Directions Custom Grafts 3D printers will allow surgeons to manufacture customized endografts in the OR.

Sapan Desai Contents

3


Advertisement

Journal of Surgical Radiology Editor-in-Chief

.com

Cynthia Shortell

Duke University Medical Center, Durham, North Carolina Duke University Medical Center, Durham, North Carolina

Hasan Alam Stanley Ashley David Bentrem Gavin Britz Alasdair Conn Mitchell Cox James Cusack Jr. Daniel Dent Celia Divino N. Joseph Espat Steven Evans Thomas Fahey III Kevin Foster Allan Friedman Richard Gray Ralph Greco Rajan Gupta Timothy Hall W. Scott Helton G. Chad Hughes Ronald C. Jones Charles Kim David King Christopher J. Kwolek Michael Leitman Matthew Menard Michael Miller Stephen Milner Eugene Moretti Theodore Pappas Jeff Pruitt Scott Pruitt Ranjan Sudan Gilbert Upchurch, Jr

Massachusetts General Hospital, Boston, Massachusetts Brigham and Women’s Hospital, Boston, Massachusetts Northwestern University, Chicago, Illinois Duke University Medical Center, Durham, North Carolina Massachusetts General Hospital, Boston, Massachusetts Duke University Medical Center, Durham, North Carolina Massachusetts General Hospital, Boston, Massachusetts University of Texas Health Sciences Center, San Antonio, Texas Mount Sinai School of Medicine, New York, New York Roger Williams Medical Center, Providence, Rhode Island Georgetown University Medical Center, Washington, DC Weill Cornell Medical College, New York City, New York Arizona Burn Center, Phoenix, Arizona Duke University Medical Center, Durham, North Carolina Mayo Clinic, Phoenix, Arizona Stanford University, Stanford, California Duke University Medical Center, Durham, North Carolina Stamford Hospital, Stamford, Connecticut Hospital of Saint Raphael, New Haven, Connecticut Duke University Medical Center, Durham, North Carolina Baylor University Medical Center, Dallas, Texas Duke University Medical Center, Durham, North Carolina Massachusetts General Hospital, Boston, Massachusetts Massachusetts General Hospital, Boston, Massachusetts Beth Israel Medical Center, New York City, New York Brigham and Women’s Hospital, Boston, Massachusetts Duke University Medical Center, Durham, North Carolina Johns Hopkins University, Baltimore, Maryland Duke University Medical Center, Durham, North Carolina Duke University Medical Center, Durham, North Carolina UT Southwestern Medical Center, Dallas, Texas Duke University Medical Center, Durham, North Carolina Duke University Medical Center, Durham, North Carolina University of Virginia, Charlottesville, Virginia

Diana Diesen Jeffrey Nienaber David Peterson Tammy Westmoreland

Duke University Medical Center, Durham, North Carolina Mayo Clinic, Rochester, Minnesota Stanford University, Stanford, California St. Jude’s Children’s Research Hospital, Memphis, Tennessee

Danny O. Jacobs

4

Managing Editor

Mark Shapiro

Journal Editors

Associate Editors

Advisor to the Journal

Duke University Medical Center, Durham, North Carolina

Executive Editor

Science Editor

Senior Staff Editor

Sapan Desai Surgisphere Corporation

Thomas Koenigsberger Surgisphere Corporation

Carol Fisher Surgisphere Corporation

Editors | January 2011

Mobile Medicine Commentary and News by Health Care Professionals

iMedicalApps is an independent online medical publication written by a team of physicians and medical students who provide commentary, news, and reviews of mobile medical technology and applications.

www.iMedicalApps.com

www.JSurgRad.com

5


Journal of Surgical Radiology

Advertisement

Published by the Surgisphere Corporation

Chief Executive Officer Chief Operations Officer Chief Financial Officer Managing Director Executive Vice-President Director of Sales & Marketing Science Editor Senior Staff Editor Associate Staff Editor Associate Staff Editor Assistant Staff Editor

Sapan S. Desai Niketa Desai Kishor Desai Sachin Desai Vishal Shirke Bernie Palmatier Thomas Koenigsberger Carol Fisher Julienne Au Michael Sakata Vivek Raj

President Vice-President and Director

Scott Vanderbilt Vance Allen

Indexing

The Journal of Surgical Radiology is indexed in EBSCOhost, Portico, SerialsSolutions, Genamics, CrossRef, and Google Scholar. ISSN: 2156-213X e-ISSN: 2156-4566 DOI: 10.5299/jsurgrad

Subscription Information

subscriptions@Surgisphere.com phone 888-837-8499 mail Surgisphere Corporation, 4706 Carmen Lane, Durham, NC 27707 customer service http://www.SurgRad.com or contact us using the information above. email

Contact Information

Editor@Surgisphere.com advertising Ads@Surgisphere.com online www.SurgRad.com Copyright Š2010 by the Surgisphere Corporation. All Rights Reserved. Reproduction without permission is prohibited. The Journal of Surgical Radiology is a publication of the Surgisphere Corporation in association with Catalyst Publishers. Printed in the USA. email

www.Surgisphere.com

6

www.SurgRad.com

J Surg Rad | January 2011

www.CatalystPublishers.com

www.ClinicalReview.com www.JSurgRad.com

7


EDITORIAL

H The Editor

J Surg Rad

i stor ic a l ly, c o n v e y i n g med ical k nowledge had been a one-way pro ce ss; st udent s passively acquired i n format ion f rom profe ssors v ia lect ure, tex tbook , or jour nal a r t icle. Our v ision for t he Jour nal of Surg ical R ad iolog y is to met amor phose t hat pa rad igm into a n excha nge of infor mat ion by adapti ng to development s in commu nicat ion technolog y. We have de sig ned a sy nerg ist ic ex per ience in which lear ners ca n interact w it h teachers, creat ing g reater oppor t unit ie s for u ndersta nd ing. Using t he lin k s prov ided w it h each a r t icle, readers of t he Jou r nal ca n commu nicate w it h aut hors t hroug h YouTube, Facebook a nd Tw itter.

8

Th i s rec ipro c at ive env i ron ment not on ly i mprove s comprehen sion of t he sc ient i f ic m ater i a l, but a l lows for feedback f rom t he re ader s to t he aut hor s t h at m ay be g i n to d i s sipate t he d i st i nc t ion bet ween te acher a nd st udent. Ac adem ic a nd sc ient i f ic conver sat ion s t h at once took mont h s to o cc u r v i a “L et ter s to t he E d itor” a nd t hei r subsequent “R e spon se s” c a n now o cc u r i n re a l t i me. L a st ly, t he prospec t for a l l re ader s to i nterac t elec t ron ic a l ly w it h a l l aut hor s ser ve s to bre a k dow n some of t he t rad it ion a l h iera rch ic a l ba rr ier s to le a r n i ng; st udent s or re sident s who nor m a l ly m ig ht not que st ion a sen ior physic i a n face -to -face m ay feel qu ite com for t able doi ng so on-l i ne. The Jou r n a l of Su r g ic a l R ad iolog y i s bre a k i ng dow n ot her t rad it ion a l ba r r ier s a s wel l. The web -ba sed, re a l t i me, i nterac t ive n at u re of ou r jou r n a l h a s at t rac ted you nger le a r n er s t h a n t he t rad it ion a l pr i nt-ba sed publ ic at ion s. The l ack of subsc r ipt ion fee a l so m a ke s t he Jou r n a l more re ad i ly ava i l able to st udent s a nd t ra i nee s. R e sident s ser ve a s a s si st a nt ed itor s, a llow i ng t hem to feel ow ner sh ip i n t he publ ic at ion pro ce s s. I n add it ion, ou r for m at h a s at t rac ted le a r ner s f rom d iver se back g rou nd s i nclud i ng physic i a n a s si st a nt s, nu r se prac t it ioner s, rad iolog y techn ic i a n s, a nd re g i stered nu r se s to whom t rad it ion a l med ic a l jou r n a l s a re i nt i m id at i ng. At t he sa me t i me, ed itor s, aut hor s, a nd m a ny re ader s h a i l f rom t he m idd le a nd sen ior ra n k s of med ic a l ac adem i a f rom some of t he f i ne st u n iver sit ie s i n t he cou nt r y. Th i s ad m i x t u re of re ader s, aut hor s a nd ed itor s f rom v i r t u a l ly a l l med ic a l wa l k s of l i fe h a s be g u n to d i s solve t he d i st i nc t ion s bet ween “ level s” of le a r ner s a nd a l lows new con t r ibut ion s f rom i nd iv idu a l s who e ach br i ng t hei r ow n u n ique per spec t ive to t he pro ce s s of sc ient i f ic e x plorat ion. R ecog n i z i ng t h at we h ave a d iver se g roup of re ader s, we h ave added a new educ at ion sec t ion t h at i nclude s boa rd- st yle prac t ice que st ion s i n su r ger y a nd rad iolog y, compac t c a se pre sent at ion s t h at i nv ite t he re ader to m a ke t he d i ag nosi s, a nd a n “ i m age of t he mont h ” t h at of fer s a poig n a nt look at a d i se a se pro ce s s f rom bot h a su r ger y a nd rad iolog y per spec t ive. Th i s sec t ion i s de sig ned for le a r ner s at a l l level s, i nclud i ng st udent s, re sident s st udy i ng for t he i nser v ice e x a m i n at ion, re sident s a nd fel lows st udy i ng for boa rd cer t i f ic at ion, a nd post- g radu ate le a r n i ng i nclud i ng recer t i f ic at ion. The succe s s of ou r jou r n a l i s a te st a ment to t he st reng t h of t h i s model. We c u r rent ly h ave over 32 , 0 0 0 subsc r iber s f rom over 10 0 cou nt r ie s; ou r Jou r n a l a nd it s a r t icle s h ave been fe at u red i n over a doz en newspaper s for it s u n ique on l i ne i nterac t ive de sig n, i nclud i ng U SA Tod ay, Wa l l St reet I n sider, a nd He a lt h Ti me s. The v i sion a nd ded ic at ion of ou r ed itor i a l st a f f, aut hor s, a nd re ader s cont i nue s to m a ke J Su r g R ad more succe s sf u l w it h e ach ed i t ion; t he Ja nu a r y 2 011 i s sue i s over 10 0 page s long a nd i nclude s a compre hen sive rev iew of GI bleed i ng, a n a r t icle on m a rat hon-i nduced col it i s, a nd a n i n spi r i ng de sc r ipt ion of endova sc u l a r t herapie s for sold ier s w it h va sc u l a r i nju r ie s du r i ng t he Gu l f Wa r. Welcome to ou r world-w ide cl a s sroom.

Cynthia Shortell, Editor-in-Chief

iStockPhoto

MD

9


COLUMN

iStockPhoto

Social Networking

J Surg Rad

Should you be connected?

10

Column | January 2011

www.JSurgRad.com

11


They are participating in Sermo and Ozmosis forums. You have received invitations to LinkedIn and are unsure just why you should be “linked in” at all. Surrounded by so much on-line activity you may be concerned that perhaps you are missing the next very big thing by not participating on these social media sites. Explore why these sites may be indeed useful to you and then decide how to utilize these social tools to your advantage.

Increase your personal credibility and establish a personal brand. You undoubtedly learned in medical school that authoring papers published in peer-reviewed journals promotes your credibility. Social media is publishing of sorts as well. By frequently writing on-line you work towards establishing a personal brand based upon the subject matter most important to you. Whether you blog, tweet, or simply post comments on other sites depends upon the type of credibility and branding you are trying to achieve; they are all helpful over time.

J Surg Rad

Improve marketing of your practice.

12

It is widely recognized that word of mouth or referral by a friend or family member inspires more trust than if a patient learns of you from an ad in the Yellow Pages or a listing in their health insurance directory. When you actively participate in social media it presents patients with an opportunity to read about you, your practice, and your clinical perspective as well as evaluate others’ reviews of their experiences with you. This multi-faceted marketing approach ultimately promotes your practice in ways that simple advertising cannot match. If you typically see your patients annually or irregularly, but would prefer to be engaged in their care more regularly, consistent participation in on-line forums maintains visibilColumn | January 2011

ity and hence, advances familiarity. This vastly increases the likelihood that your patients and colleagues will remember you when asked for a referral.

Influence peers and patients with your opinions. You may be an expert in your specialty but do not have the time to publish in peer-reviewed journals as often as you would like. Instead, consider writing short opinion pieces as a guest on other blogs, on sites such as Sermo or Ozmosis or, create your own blog. These take less time and allow you to influence your peers and educate your patients on topics that mainstream journals might not be currently publishing.

Provide patients with pertinent research or technological advances that are not easily available elsewhere. Feature relevant anecdotes and translate deep science into practical

Disclosure: The Journal of Surgical Radiology receives no compensation for featuring these columns. The opinions herein are strictly those of the author and are not endorsed by the Journal of Surgical Radiology, its editors, authors, publisher, advertisers, or affiliates. Twitter

COLUMN

Y

our colleagues are blogging, tweeting and creating Facebook fan pages, tabulating their fans and how many “like” them.

Your patients are not really your social “friends.” language patients and caregivers can understand. Participating in the same social spaces as patients allows them access to clear and concise information from an authoritative source. They can return to refer to your information saving you the time of redundant discussions with your patients in the office or by telephone.

dia. Do not be deterred by these risks, however, and consider dipping your toes into the on-line waters. If you are uncertain how to begin, start with the following steps to determine what topics you might approach:

While there are many good reasons to be engaged in social media, there are also risks to consider.

• Find one or more physician-run blogs that you like and comment on the stories and articles there. If you are unsure whose blogs with which to start, perform a quick search at Technorati.com, a blog-specific search engine.

Beware of patient privacy. This may seem obvious, however, because social media interactions are instantaneous and feel like cordial conversations between individuals, it is easy to inadvertently post protected health information (PHI).

Your patients are not really your social “friends.” Sites such as Facebook or LinkedIn impart the false impression of engaging in a confidential conversation between real friends. Remain mindful that you have no contractual relationship with these “friends” and you will be held legally liable for medical advice taken by on-line participants in your conversations; furthermore, it is all in writing. Unlike a quick conversation you might have when you encounter patient in a coffee shop, an on-line dialogue lives on forever and can be easily discovered by lawyers in the future. There are a number of benefits and a few risks associated with social me-

• Join Sermo.com, Ozmosis.com, or another physician-specific “trusted” social networking site where you can interact with fellow physicians in a secure area. Initially, you might consider simply following conversations without commenting or “lurking” and, later, participate by responding.

Use http://wefollow.com/twitter/physician to locate the top medical- and physician-related Twitter accounts. Follow a few of the most popular for an idea of what they are writing about. Once you become comfortable after participating in existing communities you will be ready to create your own community. To do so: • Introduce topics in which you are interested on Sermo, Ozmosis, or other similar sites and invite users to comment on what you would like, rather than the other way around. • Create your own Twitter handle and start tweeting your opinions on news, journal articles, and other healthcare content. • Create your own blog and post content that you believe your patients and other users should be aware.

Participating in existing communities is simple and although creating your own community of followers is a bit more labor-intensive it is most certainly worth the effort. Initially, begin by lurking and proceed with caution to find your comfort level. This will enable you to accelerate your social media interactions until you are proficient. Understanding the inherent risks is vital, but they should not prevent you from joining the on-line social revolution.

Shahid Shah

Chief Executive Officer Netspective Communications

www.JSurgRad.com

13


COLUMN

J Surg Rad

TRAUMA SYSTEMS The System Has Enemies

Wikipedia

defines system as follows: (from Latin systema) “whole compounded of several parts or members. It is a set of interacting or interdependent entities forming an integrated whole.� System characteristics include: 1) structure, 2) a set of behavioral norms, 3) interconnectivity and 4) units that func-

tion independently within the system. Structure typically consists of those states that have functions trauma systems codified by state law. Behavioral norms include pre-hospital triage guidelines, as well as clinical practice guidelines and performance improvement measures. Interconnectivity consists of multiple hospitals within the system, while independent functions consist of trauma care capability of each of the individual facilities, including rehabilitation.

iStockPhoto

14

Column | January 2011

www.JSurgRad.com

15


16

Historically, many systematic approaches to trauma care delivery have been driven by the needs of the military. Early adoption of those principles was evidenced by the West German trauma system in 1970, which resulted in a marked reduction in motor vehicle related crash mortality followed system implementation.1 In the U.S., the pioneering study by West in 1978, again demonstrated that a systematic approach to trauma care delivery resulted in a decrease in preventable deaths.2 This study has been repeated numerous times over the ensuing years. In 1995, the estimated cost of trauma care was $250 billion.3 This accounted for around 12 percent of health care expenditure. In addition to “up front” injury, more than ten times the number of patient face lifelong disability that has significant impact of patients, families and society. However, even today, the system has enemies that threaten its optimum performance. Two components that have a significant impact require remediation, sooner rather than later.

Column | January 2011

The widespread availability if advanced imaging technology, in even small community hospitals, has resulted in significant delays in transport to definitive care facilities. While the argument could be made that such imaging facilitates triage to the most appropriate care facility, the counter argument that the delay results in delayed care for the most severe injuries. Moreover, the virtual explosion of digital imaging technology has numerous proprietary imaging that may be virtually useless to destination facilities. Some images may not be able to be “loaded” or may be of poor technical quality. Some injuries require no imaging for triage/treatment (Figures 1 and 2). When images arrived at the trauma center, approximately one third of centers refuse to render an official interpretation, one third render an unofficial interpretation and one third are willing to render an official consultation and definitive interpretation (unpublished observation). This situation is likely multi-factorial. Images may be judged not suitable for interpreta-

tion, liability concerns and lastly reimbursement issues. Chwals et al evaluated the quality of computed tomography before transfer to a pediatric trauma center.4 Patients were divided into two groups. Group I was imaged prior to transfer, while Group II was imaged at the receiving trauma center. Ninety-one percent of patients imaged prior to transfer required duplicate images within four hours of transfer. This raised the question of exposure to additional radiation as well as costs. While exposure to additional radiation may not be as significant a risk, particularly in the elderly trauma population, administration of additional contrast may increase the morbidity associated with contrast nephropathy. Clearly, the risk of death from severe injury is lower in Level-I trauma centers. The goal, and challenge, is to triage the most severely injured patient to the most appropriate facility. Ideally, there should be no over or under triage. This is clearly not practical or functional. The Field Triage Decision Scheme of the American College of Sur-

Disclosure: The Journal of Surgical Radiology receives no compensation for featuring these columns. The opinions herein are strictly those of the author and are not endorsed by the Journal of Surgical Radiology, its editors, authors, publisher, advertisers, or affiliates.

COLUMN

J Surg Rad

Figures 1 (above) and 2 (right). Shown are images from a young woman who fell through a drywall ceiling panel, impaling her on an upright vacuum cleaner. Other injuries require precise imaging.

Figures 3 (left) and 4 (right). These images are from a young woman involved in a frontal motor vehicle crash. She was triaged, according to pre-hospital transport protocol, to a Level-I trauma center. She was imaged based on mild hypotension and tachycardia. The subtle hepatic vein image is labeled. She underwent emergent laparotomy and made an uneventful recovery. These two cases raise important questions. What is the fate of images, if and when, they arrived at the trauma center? What is the value of, or potential hazards, of imaging prior to transport to a definitive care facility? Can patients be transported to appropriate facilities, based on triage guidelines alone? What is the impact of such triage, in terms of over and under triage?

geons has been revised four times since its implementation in 1986. Primary triage is based on pre-hospital destination criteria while secondary triage is based on interfacility transfer guidelines. In immature systems, secondary triage may result in an overwhelming burden on trauma centers, both in terms of resources and delay in definitive care. Ciesla et al reported that secondary triage of patients with ISS < 10, who did no require operation and were discharged within 48 hours occurred in 39% of their patients.5 Conversely, delay in triage to obtain images that may be less than useful and has the potential of delaying definitive care for seriously injured patients, particularly for the aging population. Patients over age 65 are at greatest risk for fatal injuries and are less likely to return to preinjury independence. Given the aging population, has trauma mortality improved? Dutton et al analyzed trauma mortality patterns at the R. Adams

Cowley Shock Trauma Center from 1997-2008.6 They concluded that survival after severe trauma and survival benchmarked against predicted risk improved over the past 12 years. They also noted that trauma care has kept pace with the aging population and increased severity of injury.

3. Committee on Injury Prevention and Control, Institute of Medicine. 1999. Reducing the Burden of Injury: Advancing Prevention and Treatment. Washington, DC: National Academy Press.

Today’s fiscal climate adds an additional layer of complexity and challenge to those who work in busy, urban trauma centers. This is a plea for attempts at standardization of digital imaging programs to decrease the workload in receiving centers. Likewise, attempts at refining secondary triage is long over due.

5. Ciesla DJ, Sava JA, Street JH et al: Secondary overtriage: a consequence of an immature trauma system. J Am Coll Surg 206: 131-137, 2008.

References

1. Wanek SS, Trunkey DD: Organization of trauma care. Scand J Surg 91: 7-10, 2002. 2. West JG, Cales RH, Gazzaniga AB: Impact of regionalization: The Orange County Experience. Arch Surg 118: 740-744, 1983.

4. Chwals WJ, Robinson AV, Sivit CJ et al: Computed tomography before transfer to a Level I pediatric trauma center risks duplication with associated increased radiation exposure. J Pediatr Surg 43: 2268-2272, 2008.

6. Dutton RP, Stansbury LG, Leone S et al: Trauma mortality in mature trauma systems: Are we doing better? An analysis of trauma mortality patterns, 1997-2008. J Trauma 69: 620-632, 2010.

James M. Hurst, MD, FACS

Director, Division of Acute Care Surgery, Trauma and Surgical Critical Care Beth Israel Deaconess Medical Center Harvard Medical School www.JSurgRad.com

17


Coronary Stent Endothelialization A new coronary artery stent has been developed by Genous. The stent differs from other stainless steel stents in that it is “covered with antibodies that specifically attract endothelial progentior cells.” The living endothelial lining cells eventually completely cover the stent. Healing can occur within one day, compared to weeks or longer with other stents. | www.OrbusNeich.com

Thomas Koenigsberger

J Surg Rad

Science Editor

18

Technology | January 2011

Disclosure: The Journal of Surgical Review receives no compensation for featuring these technological developments.

The first potentially implantable artificial kidney has been developed by a consortium of 10 research teams. This device, powered by the patient’s own blood pressure, could save millions of lives and tremendously lower health care costs. The new device utilizes silicon nanotechnology and “engineered renal tubule cells.” The new device would be utilized to replace renal dialysis and kidney transplants. | www.ucsf.edu

Company

TECHNOLOGY

Implantable Artificial Kidney

The iPad in the Hospital and Operating Room

†

www.JSurgRad.com

19


TECHNOLOGY Introduction

J Surg Rad

T

20

he iPad has received a significant amount of attention in the health care arena since its introduction only eight months ago. The attraction is fairly obvious; it is a portable, lightweight, powerful computing device with an intuitive interface and a large library of built-in applications. In fact, major medical schools such as Stanford and University of California, Irvine have made decisions to provide iPads to all incoming medical students this year.1,2 While predicting the future of medical technology is always precarious, here are a few things we have learned in the months since the iPad was introduced.

iPad in the Operating Room A short time ago, at IMedicalApps. com, we published a brief entry

Technology | January 2011

describing the potential use of the iPad in the operating room.3 We found that a simple X ray cassette sterile bag, ubiquitous in the OR, holds an iPad comfortably. Once the iPad is inserted into the plastic bag by the circulating nurse, the top can be cut off, folded back and clamped with a hemostat (see figure above) allowing the iPad to be introduced safely into the sterile field. Notably, we found that the iPad touch screen works quite well through the plastic bag, even while wearing gloves. Somehow, the touch of the plastic bag itself against the glass screen registers as a valid touch. There was hardly any problem navigating between and inside apps, or with gestures such as pinch and zoom. This was somewhat of a surprise since, as many people have noticed, using an iPhone touch screen with gloves is difficult at best and impossible if one is double-gloved.4 What is the use of an iPad in the

OR? The reasons may actually be myriad but, generally speaking, the same features which make the iPad great for surfing the web, such as looking at images and viewing video, nicely translate into the operating room. Thus far, the most obvious use for me has been as a convenient way to easily access previous patient imaging. Additional potential assets of utilizing the iPad in the OR include the ability to review relevant anatomy at the point of care and enhancement to resident teaching. It can also be useful in bypassing hospitals' restrictive networks to access remote files and office electronic medical records (EMRs) using the cellular 3G networks. Recently, there was a report of a Japanese surgeon using an imaging application on the iPad to plan surgery in the OR.5 Although it is not clear what application was being used, I suspect it was OsiriX (seen in the accompanying images). I am not aware of any currently available applications for the iOS platform

currently available which integrate with surgical devices such as laparoscopes, arthroscopes or computeraided navigation. It is fun though to speculate about a future iPad-like device which might use onboard gyroscopes to provide an "augmented reality" view of internal structures, so that tilting the device would show different portions of the body. This could even conceivably be integrated with views from internal cameras, navigation or robotically controlled surgical instruments.

CD ROM to the iPad? I have found a very useful method involving two terrific and freely available resources; OsiriX and Dropbox. The first, Osirix, may already be familiar to many readers of this journal.6 For those not yet acquainted,

OsiriX is a free, open source DICOM (digital imaging and communication in medicine) viewer written for the Macintosh. I find it easily superior to just about every built-in reader bundled with patients' CD-ROMs and routinely use it to view patients' DICOM data. It is an indispensable

Exporting Patient Imaging to the iPad As previously mentioned, one use of the iPad in the OR is to bring patient imaging studies to surgery. But, how does one copy computed tomography (CT) or magnetic resonance imaging (MRI) images from a patient's www.JSurgRad.com

21


22

Dropbox is a multi-platform filesyncing utility.8 This terrific service has a web component and native applications for Macintosh, Windows, Linux, iPhone and iPad. It is able to rapidly synchronize files across multiple devices through file-level trickery in which it determines only those portions of the files that are changed.

Technology | January 2011

Sample Workflow My workflow is as follows (see above screenshots): 1. Insert patient's CD ROM in computer and open with OsiriX 2. Identify key images of interest 3. Export images as JPEGs into a folder on computer 4. View images in Dropbox app in iPad I also use the Apple iPad camera adapter to quickly transfer intraoperative photos to my iPad.10 For the curious family member, these photos can really enliven the post-op waiting room conversation! One glaring omission on the iPad is a method to organize images into albums once they are uploaded that would be preserved when the iPad is synchronized with iPhoto. The apps available for organizing photos on the iPad, such as PhotoSort, appear to duplicate the images

rather than manipulate the native photo album.9 We can only hope Apple will provide this function in future updates of the iPad OS.

iPad on the Wards Dr. Henry Feldman recently shared with iMedicalApps.com his experience using the iPad as his primary computing device for a week as an attending at Beth Israel Deaconess Hospital in Boston.11 One instance where the iPad really shone, he wrote, was its "nearly seamless experience" in accessing his hospital's wireless networks. "The secure wireless network handoff was amazing. As I roved around it was seamless ... and the best example is that I would use the elevator ride to catch up on news/ tech websites, and every time the elevator doors would open it would reconnect and download some more prior to the door closing." The soft keyboard of the iPad, he found, was adequate for the bulk of his documentation needs on the

Disclosure: The Journal of Surgical Radiology receives no compensation for featuring these columns. The opinions herein are strictly those of the author and are not endorsed by the Journal of Surgical Radiology, its editors, authors, publisher, advertisers, or affiliates.

TECHNOLOGY

J Surg Rad

part of my practice. In addition to basic features, such as measuring distance and angle, modifying window and level, zooming and panning, OsiriX has some features usually available only on expensive imaging workstations, such as 2D multi-plane & 3D surface reconstructions, and fusing PET/ CT images. While the iPhone version of OsiriX will run on the iPad, I still find transferring DICOM files between the desktop and mobile applications unwieldy.7 Therefore, I will describe a simple method that uses Dropbox to export selected images to the iPad.

wards.12 He brought his Bluetooth keyboard and iPad stand to the hospital but found both "sat on the shelf for one week, unused." He used desktop computers for complex compositions such as admission notes and discharge summaries, but felt "this had less to do with the keyboard than being able to see multiple web pages at once.â&#x20AC;? Orders and discharge plans did not require an external keyboard although, "after one beefy paragraph the screen keyboard does feel a tad clunky." Despite this enthusiastic report of iPad usage from an experienced physician, we should remember that Beth Israel has a sophisticated web interface for all their major clini-

Conclusion The iPad clearly has the potential to be very useful in the hospital and in the operating theater. Medicine is, by its nature, a mobile occupation and a powerful and flexible computing device will almost certainly play some role in our future everyday practices. There is still much to be learned and undoubtedly we will see the introduction of other devices with different strengths and weaknesses. In my practice, using the resources described above, I have been able to maintain a portable image repository of my patients that has been very useful in the operating room, office, and

3. iMedicalApps. Test driving the iPad in the hospital Operating Room. Available at http://www.imedicalapps.com/2010/06/ ipad-hospital-operating-room/. Accessed 9/2010. 4. iMedicalApps. iPad could support "Handwriting Keyboard" - A requirement for medical point of care use in health care. Available at http://www.imedicalapps. com/2010/01/ipad-could-support-handwriting-keyboard-a-requirement-for-medical-point-of-care-use-in-health-care/. Accessed 9/2010. 5. medGadget. iPad Used in Japanese Operating Theater. Available at http://www. medgadget.com/archives/2010/06/ipad_ used_in_japanese_operating_theater.html. Accessed 9/2010. 6. OsiriX. Available at http://www.osirixviewer.com/Downloads.html. Accessed 9/2010.

The iPad clearly has the potential to be very useful in the hospital and in the operating theater. cal applications, making the transition to using an iPad fairly seamless. When asked about security concerns with the iPad, especially if one is left behind inadvertently, Dr. Feldman pointed out that as with everything web-based, nothing is stored on the device. The iPad (like an iPhone) can be "remote-wiped," meaning all the data and settings on the device can be erased remotely. Note that this requires a MobileMe subscription ($99 per year). Each of the applications in the hospital has a username/password, and network communication for most of the applications occurs over an encrypted (SSL) connection. These measures, Dr. Feldman suggested, already exceed the security that almost any institution places on paper charts. He also mentioned a strategy for the lock screen which I have since adopted. This involves replacing the iPadâ&#x20AC;&#x2122;s default background with an image including one's name, cell and pager numbers, for device return information if the iPad were to fall into friendly hands.

in casual hallway discussions with colleagues.

Felasfa M. Wodajo, MD Musculoskeletal Tumor Surgery Senior editor, iMedicalApps.com Assistant Professor, Orthopedic Surgery, VCU School of Medicine, Inova Campus Assistant Professor, Orthopedic Surgery, Georgetown University Hospital

References 1. iMedicalApps. Stanford School of Medicine is giving the iPad to all incoming medical students. Available at http://www.imedicalapps.com/2010/07/stanford-school-ofmedicine-ipad-incoming-class/. Accessed 9/2010.

7. iMedicalApps. OsiriX Medical App Puts a Small DICOM Workstation in Your Pocket [App Review]. Available at http://www. imedicalapps.com/2009/08/app-reviewosirix-iphone-app-puts-small/. Accessed 9/2010. 8. DropBox. Available at http://www.dropbox.com. Accessed 9/2010. 9. iTunes. Photo-Sort for iPad. Available at http://itunes.apple.com/us/app/photosort-for-ipad-organize/id369610590?mt=8. Accessed 9/2010. 10. Apple iPad Camera Connection Kit. Available at http://store.apple.com/us/product/ MC531ZM/A?fnode=MTc0MjU4NjE&mc o=MTcyMTgxODY. Accessed 9/2010. 11. iMedicalApps. The iPad makes the hospital rounds: report from a busy week on the wards - Part 1. Available at http://www. imedicalapps.com/2010/06/ipad-hospitalhealth-information-technology-ehr/. Accessed 9/2010. 12. iMedicalApps. The iPad Makes the Rounds: Part 2 - Portability, Security and Infection control on the Hospital Wards. Available at http://www.imedicalapps.com/2010/06/ ipad-hospital-wards-portability-securityinfection-control/. Accessed 9/2010.

2. iMedicalApps. UC Irvine School of Medicine joins the party - giving incoming med students iPads with their white coats. Available at http://www.imedicalapps. com/2010/08/uc-irvine-school-of-medicine-ipad/. Accessed 9/2010.

www.JSurgRad.com

23


ORIGINAL ARTICLE

A Comprehensive Review of Upper GI and Small Bowel Bleeding: The Role of Modern Imaging Technology and Advanced Endoscopy Leslie M. Kobayashi MD ∙ Ricardo Borsatto MD ∙ Raul Coimbra MD, PhD  Division of Trauma, Surgical Critical Care and Burns, Department of Surgery, University of California, San Diego  Gastrointestinal and Liver Disease Center, LLC, Banner Estrella Medical Center, Phoenix, Arizona

Abstract

Overview Bleeding from the upper gastrointestinal tract is a common cause of hospital admissions and a source of significant morbidity and mortality. Additionally, obscure bleeding, or bleeding not visualized on traditional upper and lower endoscopy remains a diagnostic dilemma, resulting in a significant expenditure of time and resources, as well as morbidity and mortality. Peptic ulcer disease and esophageal varices account for the majority of upper gastrointestinal bleeding, while small bowel pathology, particularly vascular anomalies and neoplasms account for the majority of obscure bleeding. Endoscopy, traditional as well as advanced techniques such as single and double balloon endoscopy, remain the mainstay of both diagnosis and treatment. However, when endoscopic techniques are not available, fail to localize bleeding, or are not tolerated by the patient a number of alternative diagnostic methods are available. Angiography is highly specific and can be both diagnostic and therapeutic, but because bleeding must be present and brisk at the time of diagnostic exams to be detected, it is becoming increasingly used for directed therapy after other methods have localized a bleeding source. Alternative tests include capsule endoscopy, Technitium 99m labeled red blood cell scans, and increasingly computed tomography and magnetic resonance imaging. Keywords gastrointestinal, peptic ulcer disease, esophageal varices, angiography, tagged RBC, CT

J Surg Rad

Introduction

24

Gastrointestinal bleeding (GIB) is a common medical problem, responsible for 1-2% of all acute inpatient hospitalizations.1,2 Upper GIB is defined as intra-luminal bleeding affecting the proximal gastrointestinal tract, including organs extending from the mouth to the Ligament of Treitz. This includes the esophagus, stomach, and duodenum. The incidence of upper GIB is approximately 100-300 cases/100,000 of the population per year in the United States.2 Citation Kobayashi LM, Borsatto R, Coimbra R. A comprehensive review of upper GI bleeding: the role of modern imaging technology and advanced endoscopy. J Surg Radiol. 2011 Jan 1;2(1). Correspondence Raul Coimbra, MD, PhD, FACS E-mail rcoimbra@ucsd.edu. Received August 31, 2010. Accepted September 14, 2010. Epub September 16, 2010.

Original Article | January 2011

Signs and symptoms of upper GIB include nausea, hematemesis, coffee ground emesis, melena, and hematochezia. Upper GIB can also be occult and present with a microcytic anemia, or hemoccult positive stools. In the intensive care unit patient upper GIB may also present with blood tinged, or coffee ground output from nasogastric, orogastric, or gastrostomy drains. Bleeding from the small bowel can present very similarly, but is not often visible on traditional endoscopy and can become a frustrating diagnostic dilemma. Cases where no bleeding source is visible on upper or lower endoscopy, are classified as obscure GIB (OGIB). The majority of OGIB has a small bowel source, consisting primarily of vascular and neoplastic lesions.3-6 This particular patient population tends to undergo more diagnostic procedures and require more blood transfusions during longer hospitalizations resulting in increased health care costs.7 Patients with lower GIB including the appendix, colon and rectum are beyond the scope of this review. The majority of upper GIB can be localized to the stom-

Original Article

Kobayashi et al. Diagnosis of Upper GI Bleeding

and intravenous access should be assessed for adequacy. Key medications include high dose intravenous proton pump inhibitors and in patients with variceal bleeding vasopressin and octreotide should be considered during the acute phase, while beta blockers should be given after the acute bleeding has resolved and the patient is no longer hypotensive.10-13 Additionally, in patients with peptic ulcer disease medications for Helicobacter pylori should be considered after the acute phase (Figure 3). Once the patient has been stabilized and adequately resuscitated diagnostic exams should be performed as soon as possible, ideally within the first 24-48 hours, as earlier examination results in a higher diagnostic yield.2,14 The majority of upper GIB resolves spontaneously in approximately 80-85% of patients. Success after therapeutic endoscopy is also very high, approximately 90% in most studies.2,15 Rates of recurrent bleeding will vary with etiology and increase in the presence of certain endoscopic stigmata such as active bleeding, adherent clot, and exposed vessels. Figure 1. Esophagogastroduodenoscopy with view of gastric Risk of recurrent bleeding is also increased by the presence of co-morbidities such as renal or hepatic insufficiency, and ulcer with adherent clot. coagulopathy. In cases of recurrent bleeding after therapeutic endoscopy, a second look should be attempted and results in ach and duodenum with gastritis and gastric or duodenal adequate hemostasis in 74% of patients.13,16 Surgical interven2,8,9 ulcers accounting for 40-55% of cases (Figures 1 and 2). tion is required in less than 20% of patients with UGIB. InEsophageal varices account for the majority of the remaining dications for surgery include refractory or recurrent bleeding cases. Less common causes of upper GIB include; esophagitis, with a known source and blood loss requiring transfusion of esophageal tears (Mallory-Weiss and Boerhaave’s syndrome), more than six units of blood. Additionally, known risk factors duodenal diverticulae, hiatal hernias, Dieulafoy’s ulcers, anfor surgery include; hypotension, visible vessel or ulcer greater giodysplasia, Meckle’s diverticulae, Crohn’s disease, hemobithan 2 cm at endoscopy, age greater than 60 years, and severe lia, aorto-enteric fistulae, and neoplasms. As stated previously co-morbidities.1,15 the majority of OGIB originates in the small intestine and the most common lesions include vascular ectasias, arteriovenous With improvements in resuscitation, critical care, and the abilmalformations, tumors and polyps, and inflammatory lesions ity to treat patients with minimally invasive endoscopic and associated with Crohn’s disease or celiac disease. angiographic therapies mortality associated with upper GIB has decreased 23% in the United States.13 Overall mortality Risk factors for upper GIB include alcohol use, smoking, liver associated with upper GIB is now approximately 5-14%.13,17 disease, non-steroidal anti-inflammatory drug use, anticoaguHowever, mortality increases sharply with increasing age. In lation, stress states including trauma, burns, head injury, and mechanical ventilation, steroid use, and splenic vein thrombosis. Additionally, post operative patients may develop marginal ulcers near intestinal suture lines, this is especially common after gastro-jejunal anastamoses. The treatment goals in patients with suspected upper GIB include resuscitation of the patient, diagnosis of the lesion, localization of bleeding, and treatment of the identified lesion with endoscopy, angiography or surgery. As with any acute process initial assessment and treatment of upper GIB should include evaluation of the airway, breathing and circulation. In some cases of severe acute upper GIB the airway can become compromised by vomiting, secretions and blood. In these cases consideration should be given to early or prophylactic intubation. Supplemental oxygen should be considered in patients not requiring intubation. Circulatory status should be continuously monitored and resuscitation should be performed with isotonic crystalloid solution and transfusion of blood and/or blood products if indicated. Additionally interventions should be performed to maintain normothermia and normotension. A nasogastric tube and Foley catheter should be considered,

Abbreviations GIB

Gastrointestinal bleeding

OGIB

Obscure gastrointestinal bleeding

EGD

Esophagogastroduodenoscopy

PE

Push enteroscopy

SBE

Single balloon enteroscopy

DBE

Double balloon enteroscopy

CE

Capsule endoscopy

Tc-99m

Technetium 99m

tPA

Tissue plasminogen activator

CT

Computed tomography

MRI

Magnetic resonance imaging

RBC

Red blood cell

www.JSurgRad.com

25


Original Article

Kobayashi et al. Diagnosis of Upper GI Bleeding

Kobayashi et al. Diagnosis of Upper GI Bleeding

a large population study from Italy patients over 70 had eight times, and patients over 80 sixteen times the mortality of younger cohorts aged 18-39.18 Additionally, patients with upper GIB associated with cirrhosis have a significantly increased mortality with ranges of 40-70%,1 nearly ten times the rate of their non-cirrhotic counterparts. Lastly, recurrent bleeding, and massive bleeding associated with hemodynamic instability or need for transfusion of more than four units of red blood cells have been associated with increased mortality rates.18-20 Conversely, patients undergoing endoscopy and receiving care in specialty gasteroenterology wards have decreased mortality.18 It is likely that this mortality benefit is due to increased access to early endoscopy and potentially more experienced endoscopists, more rapid delivery of medical treatments, and swifter recognition of re-bleeding.18

During EGD the patient is generally under sedation, but in some cases may require general anesthesia and intubation for comfort and/or airway control. The endoscope is passed via the mouth and direct visualization of the esophagus, stomach and duodenum is easily performed. A great benefit in suspected upper GIB is that no bowel preparation is required to perform an EGD. EGD is able to localize the source of bleeding in 8095% of cases.9,21-23 Endoscopy also has the added benefit of therapeutic applications. A complete review of endoscopic therapies for upper GIB is outside the scope of this review, but includes banding or sclerotherapy of esophageal varices (Figures 4 and 5), injection of epinephrine or sclerosants, thermocoagulation or electrocoagulation, argon plasma coagulation, and placement of hemoclips. Endoscopic therapies are generally successful with effective hemorrhage control in 80-90% of patients.24,25

Diagnosis Endoscopy Endoscopy is the primary diagnostic and therapeutic modality for upper GIB. It generally entails esophagogastroduodenosFigure 2. Esophagogastroduodenoscopy with view of duo- copy (EGD), but may also include more advanced techniques. denal ulcer.

Acute UGIB

HD Stable

HD Unstable ICU admission Central line/arterial line

IVF resuscitation

+/- Massive transfusion protocol

IV PPI

PPI infusion

Correct coagulopathy

Correct coagulopathy

Push Enteroscopy, Single and Double Balloon Enteroscopy and Capsule Endoscopy

Emergent Endoscopy

Urgent endoscopy

Source of bleeding identified -Treatment

Source of bleeding not identified

Angiography OR Operating Room

Figure 3. Treatment algorithm for upper gastrointestinal bleeding.

26

Original Article | January 2011

However, endoscopy requires specialized re- Figure 4. Esophagogastroduodenoscopy with view of esophageal varices sources and personnel and carries with it inher- with signs of recent bleeding. ent risks, including aspiration, loss of airway, carThe maximal depth of examination is variable but is approxidiac complications, side effects of sedation, and mately 160 cm in most patients.25 Bowel preparation is not re9 0.5-1% risk of gastrointestinal perforation. In patients with quired for PE. In cases of OGIB PE has a diagnostic yield of severe cardiac or pulmonary co-morbidities or in whom the 28-56% with a complication rate of less than 1%.4,27 risk of attendant sedation/anesthesia outweighs the benefits of endoscopy other diagnostic modalities may be considered. Traditionally if EGD and PE failed to localize bleeding in a patient with continued evidence of ongoing GIB, the gold Standard endoscopy can be non-diagnostic in up to 10% of standard for both diagnosis and treatment was surgery with 26 cases of upper GIB. When no source of bleeding is noted intraoperative enteroscopy. In this procedure the patient is at EGD and colonoscopy, small bowel lesions are the most taken to the operating room, under general anesthesia a lapacommon source of bleeding. The majority of these lesions are rotomy incision is created. An enteroscope, which can be invascular in nature, followed in frequency by neoplasms and serted trans-orally, trans-anally, or through an incision made 3-6 inflammatory lesions. The most common examinations performed in these cases are push enteroscopy, single or double balloon enteroscopy and capsule endoscopy. Table 1 includes a brief overview of the sensitivities and complications associated with endoscopy as well as other diagnostic modalities.

Secure airway

Telemetry/Floor admission

Original Article

Push enteroscopy (PE) and single (SBE) or double balloon enteroscopy (DBE) utilize small specialized endoscopes to visualize the small bowel distal to the duodenum, or proximal to the ileocecal valve. Similar to traditional EGD, the examinations can be both diagnostic and therapeutic, and can be done with direct visualization only, or in conjunction with fluoroscopy guidance. In PE a pediatric endoscope or specialized enteroscope Figure 5. Endoscopic band ligation of esophageal varices, seen on EGD is used to maneuver beyond the duodenum and through banding device. can be used to visualize the proximal jejunum. www.JSurgRad.com

27


Original Article

Kobayashi et al. Diagnosis of Upper GI Bleeding

Kobayashi et al. Diagnosis of Upper GI Bleeding

Original Article

Figure 6. A. Tagged RBC scan showing uptake of tracer in a spherical pattern in the left upper quadrant (above).

Figure 6. B. This area of increased uptake migrates medially over time, highly suggestive of a bleeding small bowel lesion.

in the bowel is guided through the small bowel. The diagnostic yield of intraoperative enteroscopy is 83-100%, and therapy can be performed surgically or endoscopically in real time.25 However, intraoperative enteroscopy is highly invasive, may not be tolerated in high risk patients, and has a very high rate of complications, ranging from 26-41.2%.25,28,29 Because of this intraoperative enteroscopy is infrequently used as a diagnostic modality today. Three less invasive endoscopic alternatives that also allow evaluation of the entire small bowel are SBE, DBE and CE.

sess the traditional territory of an EGD (esophagus, stomach, duodenum), and can also be used to assess the small bowel, if EGD is non-diagnostic consideration should be given to DBE as the next line in the diagnostic regimen.

DBE uses an endoscope with an inflatable balloon at the distal end in conjunction with a sheath over the scope which also contains a distal inflatable balloon. The small bowel is entered beyond the duodenum and serial inflation of the balloons allows the endoscopist to pull the small bowel lumen toward and over the end of the endoscope. This technique potentially

28

Original Article | January 2011

allows visualization of the entire length of the small bowel. However, the procedure is invasive, requiring conscious sedation or general anesthesia, and can be lengthy. In a study of 51 patients undergoing DBE the mean duration of the procedure was 179.8 minutes despite the use of fluoroscopic guidance in 95% of cases.28 DBE can be performed via a transoral or trans-anal route; in a study of 479 patients either route was successful in 61% of cases, with complete visualization of the entire small bowel possible in 26% of patients.5 Overall yield of DBE ranges from 55-76%.3-5,30,31 Diagnostic yield is increased in patients with overt bleeding or continuous fecal occult blood positivity if bleeding is not overt.5,32 Like EGD and PE, DBE does not require bowel preparation if performed via the trans-oral route, however, standard bowel preparation is required if the trans-anal route is used. Complications such as small bowel perforation, bleeding and pancreatitis can occur in approximately 1% of patients.4,5,31 Because DBE can as-

SBE is similar to DBE, but uses a single balloon to pull the small intestine over the enteroscope. SBE was created after DBE as a simpler, easier, faster alternative. SBE has the additional benefit of using a non-latex balloon, which can be of benefit in patients with latex allergy. Three large single center retrospective studies of SBE revealed a diagnostic yield was 58-60%, comparable to studies of DBE. Therapy was instituted in 21-48%.33-35 The average procedure length of 38-66 minutes was shorter than the average time for DBE, and no major complications were noted in any of the three studies. The rates of minor complications such as transient abdominal pain ranged from 0-6%. However, in very few patients was complete visualization of the small bowel successful or even

attempted.33-35 A large prospective randomized, multicenter study comparing SBE to DBE revealed an overall superiority of DBE. DBE had a higher rate of complete enteroscopy 66% vs. 22%, as well as a significantly higher diagnostic yield of 52% vs. 42%. Additionally the therapeutic yield of DBE was significantly greater at 72% vs. 48% when compared to SBE.36 One hypothesized benefit of SBE has been a potentially lower complication rate, specifically a lower rate of post procedure pancreatitis, associated with the double balloon technique in up to 1% of patients. A prospective study of SBE examining complications rates revealed a rate of major complications and post-procedure pancreatitis similar to DBE.37 Overall, SBE is an alternative to DBE, it is faster, and considered by many to be an easier technique to master however it appears to have a slightly lower diagnostic and therapeutic yield overall, and may be of limited utility if distal small bowel lesions are suspected. www.JSurgRad.com

29


Original Article

Kobayashi et al. Diagnosis of Upper GI Bleeding

Kobayashi et al. Diagnosis of Upper GI Bleeding

Original Article tive study of CE and DBE revealed that despite the fact that DBE visualized the entire length of the small bowel in only 5.5% of cases, it was diagnostic in 86% of patients with no statistical difference from CE.28 Several other studies have also demonstrated equivalent diagnostic accuracy of CE and DBE.28,30,32

Figure 7. Angiography demonstrating active extravasation of contrast within the bowel lumen. Another alternative to physician controlled endoscopy is capsule endoscopy (CE). CE involves the patient swallowing a small capsule capable of taking images along the entirety of the small bowel at specific time intervals as it is propelled by peristalsis. These images are then transmitted to recording devices in belts around the body. Transit can be augmented with the use of promotility or cathartic agents, although their use is controversial, as is the necessity and method of bowel preparation. Some studies advocate bowel preparation, others merely fasting, additionally the use of agents such as simethicone and polyethylene glycol to increase visibility is also actively debated in the literature.25,27,29,30,32,37-44 The main indication for CE is OGIB, accounting for 70-80% of exams.3 Benefits of CE include the non-invasive nature of the examination, repeatability and low risk of complications. CE can look at the entire length of the small bowel, and is highly sensitive, with an overall yield of 50-80%. The best results or CE are seen in patients with ongoing bleeding, and thus it is best done early, within 2 weeks of symptoms.3,25,27,30,32,38,40 Some limitations of CE include the need for bowel preparation or fasting prior to the exam and the potential for capsule retention. Retained capsules can occur in up to 1% or patients and is generally due to a pre-existing stricture;3 this rate of retention may increase

30

Original Article | January 2011

up to 5% in patients with Crohnâ&#x20AC;&#x2122;s disease.28 Because of this CE shouldnâ&#x20AC;&#x2122;t be used in patients with obstructive symptoms, or known stricture or fistula.3-5,17,28 CE is also contraindicated in patients with pacemakers or internal cardiac defibrillators.6 Additionally CE does not allow for therapeutic intervention as in physician controlled endoscopy (EGD, PE,SBE/DBE). Studies comparing diagnostic yield demonstrate an overall superiority of CE compared to traditional radiography and PE. Literature comparing CE and DBE is variable. In a multicenter study of 56 patients, CE was compared to PE, enteroclysis and angiography, CE outperformed all other diagnostic tests with a yield of 68% vs. 38% for all other exams.43 Two large meta-analyses of OGIB revealed a statistically significant increase in diagnostic yield of both pathologically and clinically significant lesions with CE over PE, small bowel radiography, CT, MRI, and angiography. The yield of CE and DBE were comparable.27,30 Similarly a study specifically of patients with Crohnâ&#x20AC;&#x2122;s disease revealed an overall superiority of CE compared to small bowel radiography, CT scan, MRI, and push enteroscopy. A prospective single institution study comparing CE and DBE revealed higher diagnostic yield of CE over DBE as well, 80% vs. 60%.41 However, another retrospec-

Because of the non-invasive nature of CE, lower risk of adverse events, and potentially higher diagnostic yield, most studies recommend CE as a first line examination in OGIB, followed by DBE if pathology is discovered on CE.4,5,30,39,41 Additionally, if CE is non-diagnostic, or OGIB is recurrent consideration should be given to repeating CE. A study of patients undergoing repeat CE revealed that 75% of repeated studies resulted in additional findings, of which 65.2% resulted in changes in patient management.42 The roles of CE and DBE, whether they are complimentary, serial, or interchangeable is still an area of ongoing research. Some data suggests that DBE may have a role in OGIB patients with negative CE. In a study of 51 patients, DBE significantly increased the diagnostic yield in patients who previously underwent CE regardless of whether CE was positive or negative.28 However, DBE missed lesions diagnosed on CE in 31 of 51 patients, suggesting that the two exams should be seen as complimentary rather than exclusionary. In cases of OGIB radiographic studies become increasingly important in diagnosis, localization and management. Adjuncts and alternatives to endoscopy include enteroclysis, angiography, tagged red blood cell scans, and more recently computed tomography (CT) and magnetic resonance imaging (MRI). Angiography can be used to both to diagnose and to treat GIB anywhere along the GI tract. The other options are solely diagnostic, Figure 8. Angiography demonstrating an area of active extravasation before (top) and but can help narrow differential diagnoses and aid tremendously in after (bottom) embolization. www.JSurgRad.com

31


Original Article

Kobayashi et al. Diagnosis of Upper GI Bleeding and 83%-100% respectively.48,49,54,55 Given the improved diagnostic yield and additional information as well as shorter examination time when CT technology is used, small bowel follow through and enteroclysis now have a very small role in diagnosis of GIB outside of patients with Crohn’s disease.3 One area where contrast radiography continues to be used is to rule out stricture in high risk patients preparing to undergo CE.25

Nuclear Scintigraphy Nuclear imaging includes a broad range of examinations performed with a variety of radio-labeled agents. The most commonly used examination in the diagnosis of GIB is scintigraphy with technetium 99m (Tc-99m) labeled erythrocytes. While scintigraphy can also be performed with Tc-99m labeled sulfur colloid, Tc-99m labeled RBC’s, or tagged RBC’s, have the benefit of a substantially longer half life when compared to sulfur colloid. Because of its short half life, scans with Tc-99m sulfur labeled colloid will only demonFigure 9. Angiogram after deployment of coils within the gastroduodenal artery. strate extravasation if bleeding occurs within 10-15 minutes of inpreoperative planning. They can also be used to identify lejection.56,57 To perform the tagged sions for directed endoscopy, SBE/DBE, or angiography. RBC scan autologous blood is withdrawn and labeled with Tc-99m, the blood is then re-injected and scans are performed, either intermittently or continuously with video for 1-3 hours. Contrast Radiography and Enteroclysis If desired, delayed images can be taken as late as the following Traditionally the only means to assess small bowel pathology day. Bleeding is diagnosed if tracer is noted to accumulate in was with contrast enhanced radiography. In these examinabowel lumen, if tracer is seen to migrate through the bowel tions contrast was used to coat the lumen of the small bowel to lumen, or the intensity of the tracer increases in the lumen elucidate lesions such as tumors, ulcers and polyps. Contrast over time (Figure 6a and b).3,4 Overall, tagged RBC scan has could be delivered via the trans-oral route (small bowel follow a very broad range of diagnostic yield, 15-100%, with an averthrough), or via a naso-jejunal tube placed under fluoroscopic age accurate localization of bleeding of 80%.56-64 The rate of guidance, known as enteroclysis. Naso-jejunal delivery has the false positives and inaccurate localization is particularly high benefit of continuous contrast flow with potentially improved in cases of upper GIB, and small bowel sources of OGIB.56,60,62luminal distension, improved patient tolerance and decreased 64 The concentration of tracer in the spleen and liver can esophago-gastric reflux. Both small bowel follow through and obscure upper GI sources of bleeding, particularly in the traditional enteroclysis are relatively insensitive for OGIB, with stomach and duodenum. Additionally splenules can obscure diagnostic yields of 6-10%,3,6,45 with maximal yields achieved sources of bleeding in the duodenum and small bowel, as well in patients with a diagnosis or suspicion of Crohn’s disease.32,46 as the splenic flexture. The small bowel is centrally located When compared to CT enterography/enteroclysis use of CT and can overlap with large blood vessels, again obscuring results in better visualization of intra-luminal pathology than bleeding sources. The small bowel is also highly mobile, with exams performed with fluoroscopy or plain radiography, and rapid antegrade and retrograde motion of the tracer, making can visualize extra-luminal pathology.27,47,48 The diagnostic localization very difficult. Advantages of tagged RBC scans yield of CT enterography and enteroclysis is higher for both include the ability to scan for up to 24 hours,4 relatively easy OGIB and active GIB, with diagnostic yields of 33-56%23,49-53 and non-invasive nature of the exam, and the low threshold

32

Original Article | January 2011

Kobayashi et al. Diagnosis of Upper GI Bleeding

Original Article

arterio-biliary fistula. However, the lower threshold for detection of bleeding is 0.5-1 mL/minute, and bleeding must be occurring at the time of the examination to be detected.1,3,66 The role of angiography is significantly different between upper and lower GIB. Angiography is primarily therapeutic in upper GIB, while it continues to be both diagnostic and therapeutic for lower GIB. EGD is more often definitive in upper GIB, even if treatment is unsuccessful or unfeasible. Therefore, patients with upper GIB rarely require angiography for diagnosis or localization. This is in contrast to lower GIB, where endoscopy is often limited by lack of bowel preparation and difficult visualization secondary to blood and blood clots if patients are actively bleeding (Figure 7).14,57,66,67 The exceptions to this rule are post-operative patients, in whom, surFigure 10A. CT scan of small bowel GIST tumor, coronal (above) and axial (next page) images. gically altered anatomy may make EGD for detection of bleeding (rate 0.1-0.5 mL/min) Disadvantages difficult or impossible. In these cases angiography continues include the lengthy nature of the exam, the potentially high to play a large role in diagnosis of upper GIB.66 Diagnostic false localization rate of 22%,65 which is increased for upper yield ranges widely from 20-87%, with yields increasing in heGI and small bowel hemorrhages to 33-85%,60,63 and the need modynamically unstable patients, and in upper rather than for specialized isotopes and detection equipment which may lower GIB.32,66,68,69 Given the importance of timing on diagnot be available or staffed at night or on weekends/holidays. nostic yield, and the high rate of spontaneous resolution of upper GIB, consideration should be given to pre-angiography Angiography screening with CT or tagged RBC scan especially if angiography is not immediately available. Screening with both exams Traditional angiography has become an important tool in the has been associated with increased diagnostic yield of angidiagnosis and treatment of GIB. Angiography requires arterial ography.26,66,69,70 If either scan is negative for bleeding the likeliaccess to be obtained, and injection of intra-arterial contrast hood of a positive angiography is low, as is the risk of continmaterial. Angiography is capable of assessing bleeding along ued or recurrent bleeding, and conservative management may the entirety of the GI tract, it has the added benefit of being be considered.56,57,66,69 able to diagnose and treat rare causes of upper GIB such as www.JSurgRad.com

33


Original Article

Kobayashi et al. Diagnosis of Upper GI Bleeding

Original Article

Kobayashi et al. Diagnosis of Upper GI Bleeding

OGIB Active

Occult

CT scan No Source

Capsule endoscopy Source identified

Source identified

Angiography

No source Active -DBE Occult-CE Figure 10B. Axial image from CT scan of small bowel GIST tumor. The main use for angiography in upper GIB however, is therapeutic, either first line, or in cases of failed endoscopic treatment. Therapeutic options include embolization with a variety of agents, or selective catheter directed vasopressin infusion. Because of high rates of recurrent bleeding after infusion was stopped and high rates of side effects associated with vasopressin, angioembolization is the preferred treatment for non-variceal upper GIB (Figure 8a, 8b, and 9).2 Rates of technical success with angioembolization range from 52-98%, with rates of clinical success ranging from 51-91%.66,69,71-73 Predictors of success include upper GI source, bleeding outside of the gastroduodenal artery territory, and reversal of coagulopathy prior to embolization.24,69 In the specific subset of patients failing endoscopic treatment, angiographic treatment appears to be equivalent to surgery, and can be considered either an adjunct to, or alternative for surgical treatment, especially in patients with high surgical risk.24,69,74 Risk factors for recurrent bleeding after angioembolization include post-procedural or post-operative bleeding, bleeding secondary to trauma, bleeding from a duodenal rather than a gastric source, presence of large or multiple ulcers at endoscopy, multisystem organ failure, co-morbidities, transfusion requirements greater than six units, shock, and coagulopathy.66,69,73,75 Interestingly the angiographic treatment technique,

34

Original Article | January 2011

embolization material used and blind vs. directed embolization were not consistent predictors of outcomes.2,66,69,73 Disadvantages of angiography include the risk of procedural related complications such as access site thrombosis or hemorrhage, extraluminal injection of contrast material, contrast reactions or induced nephropathy, injury to vascular structures, patient discomfort, and rarely, compartment syndrome. However procedure-related complications are rare in most studies, ranging from 0-26%, with higher rates associated with therapeutic rather than purely diagnostic angiography.66,69 Additionally, angiography requires specialized equipment and personnel with advanced training which may not be present in all centers.

Provocative Angiography/Endoscopy The nature of GIB is sporadic, and there is a great deal of uncertainty, and difficulty in timing studies to episodes of active bleeding. Many studies with high sensitivity, such as angiography, require patients to be bleeding at the exact time of their examination. Because of this, provocative angiography or scintigraphy can be considered in patients in whom all studies fail to lead to a diagnosis and who continue to experience bleeding symptoms intermittently. Provocative imaging requires

Source

No Source

Tagged RBC scan

Directed SBE/DBE/Angio Vs. Surgery

Treatment

Figure 11. Diagnostic algorithm for OGIB. the administration of medications to incite bleeding during investigations. This can be accomplished with administration of vasodilators such as tolazoline, anticoagulants such as heparin, and thrombolytic agents such as tissue plasminogen activator (tPA) and urokinase. Provocative agents can be given locally with catheter directed infusion, systemically via intravenous infusion or bolus, or in combination. Additionally, medications can be used in combination with each other, and their dosages increased incrementally during the course of an investigation if no bleeding is noted initially. Only small nonrandomized case series exist, but the overall yield varies from 29-80%, with most studies reporting successful provocation in approximately 30%.4,76-81 After diagnosis lesions can either be treated angiographically or surgically depending on their nature and location. The use of provocative agents does increase the potential risk of bleeding complications both related to the initial GIB, and to extra-intestinal sources. However, most studies report a low rate of complications, less than 10%, primarily related to puncture site hematomas.76-80 In addition to angiography, provocative medications can be used in conjunction with scintigraphy.80,81 If provocative scintigraphy results in evidence of bleeding the patient can then undergo directed angiography for localization and treatment. In a study of nine

patients with OGIB, patients underwent scintigraphy while given heparin bolus followed by infusion, then bolus followed by infusion of urokinase in an escalating fashion, resulting in detectable GIB in 40% of patients on scintigraphy.80

Computed Tomography CT scan for diagnosis of GIB, and OGIB is a provocative new area of active ongoing research. There are a number of subtle variations in technique that can be used depending on center preference and suspected pathology. Multi-detector systems with 32 and 64 channels have led to improvements in temporal and spatial resolution, as well as tolerance of faster table speeds, resulting in improved detection of small rapidly resolving blushes on CT.52 Most investigators agree that protocols for OGIB need to use intravenous contrast with arterial, portal venous, and delayed phase scanning.23,52 Each phase is suited to detecting a specific subset of different bowel pathologies. The arterial phase is best used to detect early draining veins or vascular tufts, the portal venous phase is best for detection of bowel wall neoplasms (Figure 10a and b), and the delayed phase for identification of pooling contrast in cases of active extravasation.52 In cases of acute hemorrhage single or double www.JSurgRad.com

35


36

Perforation: 0.34-6.4% Pancreatitis: 0.3-1.7% 1) Overt OGIB 2) Treatment of SB known lesion Hemostasis: 65-77% Rebleed rate: 5-51%

Complete in 12-25%

Complete in 26-86%

58-60%

55-76%

Hemostasis: Adequate cessation of active hemorrhage on endoscopy or angiography, or resection of bleeding lesion at surgery.

Perforation: <1% Pancreatitis: <1% 1) OGIB 2) Treatment of known proximal SB lesion Hemostasis: 21-48% Rebleed rate: 19%

Complete in: 60-78% 50-80% Capsule endoscopy

NA

Diagnostic yield: Percentage of patients where lesion likely to explain hemorrhage is identified, or site of active hemorrhage is identified.

Retained capsule: <1% * 5-10% in Crohn’s Occult OGIB NA

50-80% 83-100% Intra-operative enteroscopy

DBE

Complications: Intra-operative: 3-6% Post-operative: 26-41.2% Ongoing overt OGIB with negative CE, CT, tagged RBC scan, angiography Hemostasis: 70-100% Rebleed rate: 17-52%

Max 50-160cm 19-56% PE

SBE

Perforation: 1% Treatment of proximal SB lesion identified on CT, CE, Tagged RBC scan Hemostasis: 14-92% Rebleed rate: 15-62%

NA 80-95% EGD

CT scan

Original Article | January 2011

Perforation: 0.5-1% 1) Active bleeding with hematemesis, coffee ground emesis, or melena 2) Recurrent UGIB Hemostasis: 80-94% Rebleed rate: 4-20% 2nd Treatment hemostasis: 74%

NA

1) Recurrent bleeding after endoscopic therapy 2) Treatment of bleeding source found on CT or Tagged RBC scan not accessible to endoscopy/ enteroscopy Hemostasis: 51-91% Rebleed rate: 7-34% 20-87%

CIN rate: 3.3-14.5% *increased to 14.8-62% in high risk patients Local complications: - Ischemia: 7-40% -Pseudoaneurysm: Diagnostic: 0.2-0.5% Therapeutic: 2-8% -Hematoma/Bleed: 3-17% -AVF: <0.1% -Dissection/thrombosis: <0.5% -Compartment syndrome: 0.004% Angiography

High false positive rate for upper GI and small bowel hemorrhage (33-85%) Occult obscure GIB with suspicion of distal small bowel or large bowel origin NA 15-100% Tagged RBC scan

Anaphylaxis: 0.0096% Nephrogenic systemic fibrosis: 2-5% Known diagnosis of Crohn’s disease, concern for extra-luminal pathology NA 100% 0-40% MRI

NA 100% Active bleed: 72-100% Occult bleeding: 33-56%

Anaphylaxis: 0.004-0.22% CIN rate: 1.2-2.6% *increased to 11-33% in high risk patients Local complications: -Infiltration: 0.2-0.9% 1) Localization of active bleeding if endoscopy not immediately available, or suspicion for extra-luminal source of bleeding. 2) Localization of ongoing bleeding if no source identified on endoscopy

Kobayashi et al. Diagnosis of Upper GI Bleeding

Therapy Small Bowel Visualization Diagnostic Yield

Table 1. Diagnostic modalities, sensitivity, indication, and complications.3-5,9,10,15,31,35,82,87-99

Indication

Complications

Original Article

Kobayashi et al. Diagnosis of Upper GI Bleeding phase scans may be adequate.26,55,82 In order to better visualize intra-luminal pathology, and as a neutral background for detection of active hemorrhage the small bowel can be distended with neutral contrast material. This material can be delivered via oral (CT enterography) or naso-jejunal (CT enteroclysis) routes. CT enterography uses oral contrast given in four doses every 20 minutes to achieve luminal distension, the last dose is typically given on table or immediately prior to scanning. In CT enteroclysis contrast is given at a constant rate of 60120 mL/minute via a naso-jejunal tube.3 Naso-jejunal tubes are generally placed prior to the examination under fluoroscopic guidance. There is some debate among investigators, but CT enteroclysis may be better than enterography due to the rapid rate and high volume of contrast delivery resulting in improved luminal distension.6 While the oral route may be preferable to a naso-jejunal tube in most ambulatory patients, because of the large volume of contrast material required, CT enteroclysis may be better tolerated by debilitated patients, those with altered mental status, severe gastroesophageal reflux, and intubated patients. CT enteroclysis may also help avoid vomiting and reflux.6 CT scanning can also be performed without intra-luminal contrast. Some investigators even suggest their use may dilute extravasated material decreasing diagnostic yield.48,55,83 If no enteral contrast is used, CT angiography protocols may be used to increase diagnostic yield. CT angiography requires no pre-exam preparation, and is rapid and easily performed.3 Both CT enteroclysis and enterography have better visualization of intra-luminal pathology than enteroclysis with fluoroscopy or plain radiography, and can additionally visualize extra-luminal pathology.27,47,48 Other benefits include repeatability and ease and speed of the exam; making it a useful tool in patients with brisk bleeding who may not tolerate a lengthy bowel preparation or exam. Alternatively CT angiography can be used to detect bleeding as slow as 0.3-0.5 mL/ minute, better than that seen with traditional angiography, and with better localization than tagged RBC scan.3,84 Unlike other diagnostic modalities for GIB (endoscopy, radiography, angiography, tagged RBC scan) CT scanning has the ability to identify extra-luminal sources of bleeding. Another benefit of CT scanning is that the results are presented in a format many physicians, including surgeons, internists, emergency medicine physicians, and radiologists are comfortable with. Lastly, CT scanners are present in most hospitals, CT scanning can be performed without specialty trained clinicians, and are generally available at night, on weekends and holidays. Some drawbacks include significant radiation exposure to patients with multiphase studies, possible allergic reactions or contrast induced nephropathy, and inability to perform therapeutic interventions. Some investigators have shown excellent sensitivity with single phase CT scans, and meta-analyses including studies with early generation single and double channel scans have also demonstrated good sensitivity suggesting that the dose of radiation might be reduced without sacrificing quality.26,55,82 Accuracy of CT scanning for detection of bleeding and cor-

Original Article rect diagnosis of the etiology of bleeding has varied greatly in the literature. Studies of patients with OGIB without evidence of active ongoing hemorrhage found diagnostic yields of 33-56%.23,49-53 However, other studies of acute active GIB, found much higher diagnostic yields of 83%-100%.48,49,54,55 As in endoscopic examinations the diagnostic yield of CT scanning increases in patients with evidence of active or continued bleeding.48,49,52,54 CT scanning may also have a higher diagnostic yield for patients with a raised or neoplastic lesion as the source of bleeding.32 Additionally, true positive results will be visible in multiple phases of the study, and evolve with time, hence triple phase scanning may decrease false positive rates.23,52 Overall diagnostic yield has improved with advanced technology and improved awareness of the importance of technique and timing of the examination. Two meta-analyses comparing CT with endoscopy, CE, angiography, tagged RBC scan and surgery revealed excellent overall results among patients with GIB.26,82 The more recent of the two meta-analyses by Chua and Ridley in 2008 examined eight studies including 129 patients with acute and OGIB, the overall sensitivity was 86% and specificity was 95% despite the use of early generation scanners (1-4 slice) in most studies.26 Overall, studies of CT in the diagnostic algorithm for acute upper GI and active OGIB are favorable, however they are limited by small sample sizes and retrospective data collection. It is likely that CT will become a part of the diagnostic algorithm for GIB, as it appears to be complimentary to traditional and advanced endoscopic techniques. While unlikely to replace endoscopy or CE, it is likely to become increasingly utilized over radiography, fluoroscopy, and may even replace tagged RBC scans in the algorithm particularly of upper GI and OGIB. Pre-procedural CT could allow endoscopic, angiographic, and surgical intervention to be more directed and more strictly therapeutic rather than diagnostic. In particular CT scanning appears to increase diagnostic and therapeutic yield of DBE and angiography, and should be considered a good screening exam prior to these complex and invasive interventions, especially in centers that do not have CE capabilities.

Magnetic Resonance Imaging Magnetic Resonance Imaging (MRI) has been acknowledged to be a significant advancement in imaging technology. In particular, it was a significant improvement over small bowel radiography for diagnosis and management of inflammatory bowel disease, where it very accurately identified ulcerations, bowel wall edema, augmented mesenteric vascularity and lymphadenopathy.38 However, its role in diagnosis and localization of GIB has not been well defined. Early studies in pig models gave promising results with sensitivity and specificity of contrast enhanced MRI of 100% in animals with surgically induced small and large bowel bleeding.85,86 Unfortunately findings on MRI of patients with OGIB have not been as encouraging as animal studies. A study of patients with mixed GI pathology, of which 14 had OGIB, found CE was far superior to MRI, with diagnostic yield of 36% in CE and zero in MRI. However, extra-intestinal findings were present on 28.6% of MRI’s in this patient population.44 A more recent www.JSurgRad.com

37


Original Article study comparing MRI and CE found comparable results with a diagnostic yield of 55% and 40% respectively for CE and MRI in patients with OGIB.38 Interestingly, the investigators of both studies noted that MRI tended to find more lesions in the distal small bowel than the proximal small bowel, and yields in this specific region were similar between MRI and CE. This may be secondary to improved sensitivity of MRI or decreased sensitivity of CE in the distal small bowel. This may possibly suggest MRI as an adjunct to SBE/DBE or CE if the entirety of the small bowel is not successfully visualized. However, these studies are small and non-randomized, and more investigation is needed to delineate where, and if, MRI has a role in the diagnostic algorithm of upper GIB and OGIB. Lastly, the addition of intravenous contrast and MR angiography may enlarge the role of MR imaging in GIB.

Summary Upper GIB and obscure GIB remain common problems, and are responsible for a significant number of hospitalizations as well as a great deal of morbidity and mortality. A multidisciplinary approach is required for timely and accurate diagnosis and treatment. Endoscopy remains a mainstay of both diagnosis and treatment with excellent diagnostic yield and therapeutic results in the vast majority of patients. In cases of recurrent bleeding repeat endoscopy should be the first intervention, followed by either angiography or surgery. In patients with OGIB, CE should be the first choice for diagnosis, but care should be taken to minimize the risk of capsule retention with pre-procedural contrast radiography or CT scan in patients with high risk of stricture. CE and scintigraphy are good screening exams prior to therapeutic angiography or DBE. CT is an alternative to CE and tagged RBC scans that is more accessible, easier and faster to perform, and provides more clinical information. Figure 11 provides a brief diagnostic algorithm for OGIB suggested by the authors of this review.

Disclosures The authors have no disclosures or conflicts of interest related to this manuscript.

References

1. Gracias, V. Reilly, P. McKenny, M. Velmahos, G (Ed.) Acute Care Surgery: A Guide for General Surgeons. : McGraw-Hill, 2009. 2. Burke, S. J., Golzarian, J., Weldon, D., and Sun, S. Nonvariceal upper gastrointestinal bleeding. Eur. Radiol. 17: 1714-1726, 2007. 3. Graca, B. M., Freire, P. A., Brito, J. B., Ilharco, J. M., Carvalheiro, V. M., and Caseiro-Alves, F. Gastroenterologic and radiologic approach to obscure gastrointestinal bleeding: how, why, and when? Radiographics 30: 235-252, 2010.

Kobayashi et al. Diagnosis of Upper GI Bleeding Matsui, T., Iida, M., Tanaka, S., Chiba, T., Sakamoto, C., Sugano, K., and Goto, H. Diagnosis and treatment of obscure GI bleeding at double balloon endoscopy. Gastrointest. Endosc. 66: S72-7, 2007.

24. Millward, S. F. ACR Appropriateness Criteria on treatment of acute nonvariceal gastrointestinal tract bleeding. J. Am. Coll. Radiol. 5: 550554, 2008.

6. Filippone, A., Cianci, R., Milano, A., Valeriano, S., Di Mizio, V., and Storto, M. L. Obscure gastrointestinal bleeding and small bowel pathology: comparison between wireless capsule endoscopy and multidetector-row CT enteroclysis. Abdom. Imaging 33: 398-406, 2008.

25. Carey, E. J., and Fleischer, D. E. Investigation of the small bowel in gastrointestinal bleeding--enteroscopy and capsule endoscopy. Gastroenterol. Clin. North Am. 34: 719-734, 2005.

7. Prakash, C., and Zuckerman, G. R. Acute small bowel bleeding: a distinct entity with significantly different economic implications compared with GI bleeding from other locations. Gastrointest. Endosc. 58: 330-335, 2003. 8. Van Dam, J., and Brugge, W. R. Endoscopy of the upper gastrointestinal tract. N. Engl. J. Med. 341: 1738-1748, 1999. 9. Cappell, M. S., and Medscape Therapeutic endoscopy for acute upper gastrointestinal bleeding. Nat. Rev. Gastroenterol. Hepatol. 7: 214-229, 2010. 10. Cappell, M. S., and Friedel, D. Initial management of acute upper gastrointestinal bleeding: from initial evaluation up to gastrointestinal endoscopy. Med. Clin. North Am. 92: 491-509, xi, 2008. 11. Dohler, K. D., and Meyer, M. Vasopressin analogues in the treatment of hepatorenal syndrome and gastrointestinal haemorrhage. Best Pract. Res. Clin. Anaesthesiol. 22: 335-350, 2008. 12. Abid, S., Jafri, W., Hamid, S., Salih, M., Azam, Z., Mumtaz, K., Shah, H. A., and Abbas, Z. Terlipressin vs. octreotide in bleeding esophageal varices as an adjuvant therapy with endoscopic band ligation: a randomized double-blind placebo-controlled trial. Am. J. Gastroenterol. 104: 617-623, 2009. 13. Barkun, A. N., Bardou, M., Kuipers, E. J., Sung, J., Hunt, R. H., Martel, M., Sinclair, P., and International Consensus Upper Gastrointestinal Bleeding Conference Group International consensus recommendations on the management of patients with nonvariceal upper gastrointestinal bleeding. Ann. Intern. Med. 152: 101-113, 2010. 14. Concha, R., Amaro, R., and Barkin, J. S. Obscure gastrointestinal bleeding: diagnostic and therapeutic approach. J. Clin. Gastroenterol. 41: 242-251, 2007. 15. Schoenberg, M. H. Surgical therapy for peptic ulcer and nonvariceal bleeding. Langenbecks Arch. Surg. 386: 98-103, 2001. 16. Lau, J. Y., Sung, J. J., Lam, Y. H., Chan, A. C., Ng, E. K., Lee, D. W., Chan, F. K., Suen, R. C., and Chung, S. C. Endoscopic retreatment compared with surgery in patients with recurrent bleeding after initial endoscopic control of bleeding ulcers. N. Engl. J. Med. 340: 751-756, 1999. 17. Rubin, M., Hussain, S. A., Shalomov, A., Cortes, R. A., Smith, M. S., and Kim, S. H. Live View Video Capsule Endoscopy Enables Risk Stratification of Patients with Acute Upper GI Bleeding in the Emergency Room: A Pilot Study. Dig. Dis. Sci., 2010. 18. Kohn, A., Ancona, C., Belleudi, V., Davoli, M., Giglio, L., Fusco, D., Andreoli, A., Perucci, C., and Prantera, C. The impact of endoscopy and specialist care on 30-day mortality among patients with acute nonvariceal upper gastrointestinal hemorrhage: An Italian populationbased study. Dig. Liver Dis. 42: 629-634, 2010. 19. Leitman, I. M., Paull, D. E., and Shires, G. T.,3rd Evaluation and management of massive lower gastrointestinal hemorrhage. Ann. Surg. 209: 175-180, 1989. 20. Walsh, R. M., Anain, P., Geisinger, M., Vogt, D., Mayes, J., GrundfestBroniatowski, S., and Henderson, J. M. Role of angiography and embolization for massive gastroduodenal hemorrhage. J. Gastrointest. Surg. 3: 61-5; discussion 66, 1999. 21. Cappell, M. S., and Friedel, D. Acute nonvariceal upper gastrointestinal bleeding: endoscopic diagnosis and therapy. Med. Clin. North Am. 92: 511-50, vii-viii, 2008.

4. Singh, V., and Alexander, J. A. The evaluation and management of obscure and occult gastrointestinal bleeding. Abdom. Imaging 34: 311319, 2009.

22. Chak, A., Cooper, G. S., Lloyd, L. E., Kolz, C. S., Barnhart, B. A., and Wong, R. C. Effectiveness of endoscopy in patients admitted to the intensive care unit with upper GI hemorrhage. Gastrointest. Endosc. 53: 6-13, 2001.

5. Ohmiya, N., Yano, T., Yamamoto, H., Arakawa, D., Nakamura, M., Honda, W., Itoh, A., Hirooka, Y., Niwa, Y., Maeda, O., Ando, T., Yao, T.,

23. Huprich, J. E. Multi-phase CT enterography in obscure GI bleeding. Abdom. Imaging 34: 303-309, 2009.

38

Original Article | January 2011

Kobayashi et al. Diagnosis of Upper GI Bleeding

Original Article obscure gastrointestinal bleeding. Digestion 70: 201-206, 2004. 41. Hadithi, M., Heine, G. D., Jacobs, M. A., van Bodegraven, A. A., and Mulder, C. J. A prospective study comparing video capsule endoscopy with double-balloon enteroscopy in patients with obscure gastrointestinal bleeding. Am. J. Gastroenterol. 101: 52-57, 2006.

26. Chua, A. E., and Ridley, L. J. Diagnostic accuracy of CT angiography in acute gastrointestinal bleeding. J. Med. Imaging Radiat. Oncol. 52: 333-338, 2008.

42. Jones, B. H., Fleischer, D. E., Sharma, V. K., Heigh, R. I., Shiff, A. D., Hernandez, J. L., and Leighton, J. A. Yield of repeat wireless video capsule endoscopy in patients with obscure gastrointestinal bleeding. Am. J. Gastroenterol. 100: 1058-1064, 2005.

27. Triester, S. L., Leighton, J. A., Leontiadis, G. I., Fleischer, D. E., Hara, A. K., Heigh, R. I., Shiff, A. D., and Sharma, V. K. A meta-analysis of the yield of capsule endoscopy compared to other diagnostic modalities in patients with obscure gastrointestinal bleeding. Am. J. Gastroenterol. 100: 2407-2418, 2005.

43. Neu, B., Ell, C., May, A., Schmid, E., Riemann, J. F., Hagenmuller, F., Keuchel, M., Soehendra, N., Seitz, U., Meining, A., and Rosch, T. Capsule endoscopy versus standard tests in influencing management of obscure digestive bleeding: results from a German multicenter trial. Am. J. Gastroenterol. 100: 1736-1742, 2005.

28. Kamalaporn, P., Cho, S., Basset, N., Cirocco, M., May, G., Kortan, P., Kandel, G., and Marcon, N. Double-balloon enteroscopy following capsule endoscopy in the management of obscure gastrointestinal bleeding: outcome of a combined approach. Can. J. Gastroenterol. 22: 491-495, 2008.

44. Golder, S. K., Schreyer, A. G., Endlicher, E., Feuerbach, S., Scholmerich, J., Kullmann, F., Seitz, J., Rogler, G., and Herfarth, H. Comparison of capsule endoscopy and magnetic resonance (MR) enteroclysis in suspected small bowel disease. Int. J. Colorectal Dis. 21: 97-104, 2006.

29. Hartmann, D., Schmidt, H., Bolz, G., Schilling, D., Kinzel, F., Eickhoff, A., Huschner, W., Moller, K., Jakobs, R., Reitzig, P., Weickert, U., Gellert, K., Schultz, H., Guenther, K., Hollerbuhl, H., Schoenleben, K., Schulz, H. J., and Riemann, J. F. A prospective two-center study comparing wireless capsule endoscopy with intraoperative enteroscopy in patients with obscure GI bleeding. Gastrointest. Endosc. 61: 826-832, 2005.

45. Zuckerman, G. R., Prakash, C., Askin, M. P., and Lewis, B. S. AGA technical review on the evaluation and management of occult and obscure gastrointestinal bleeding. Gastroenterology 118: 201-221, 2000. 46. Dionisio, P. M., Gurudu, S. R., Leighton, J. A., Leontiadis, G. I., Fleischer, D. E., Hara, A. K., Heigh, R. I., Shiff, A. D., and Sharma, V. K. Capsule endoscopy has a significantly higher diagnostic yield in patients with suspected and established small-bowel Crohn's disease: a meta-analysis. Am. J. Gastroenterol. 105: 1240-8; quiz 1249, 2010.

30. Pasha, S. F., Leighton, J. A., Das, A., Harrison, M. E., Decker, G. A., Fleischer, D. E., and Sharma, V. K. Double-balloon enteroscopy and capsule endoscopy have comparable diagnostic yield in small-bowel disease: a meta-analysis. Clin. Gastroenterol. Hepatol. 6: 671-676, 2008.

47. Liangpunsakul, S., Chadalawada, V., Rex, D. K., Maglinte, D., and Lappas, J. Wireless capsule endoscopy detects small bowel ulcers in patients with normal results from state of the art enteroclysis. Am. J. Gastroenterol. 98: 1295-1298, 2003.

31. Yamamoto, H., Kita, H., Sunada, K., Hayashi, Y., Sato, H., Yano, T., Iwamoto, M., Sekine, Y., Miyata, T., Kuno, A., Ajibe, H., Ido, K., and Sugano, K. Clinical outcomes of double-balloon endoscopy for the diagnosis and treatment of small-intestinal diseases. Clin. Gastroenterol. Hepatol. 2: 1010-1016, 2004.

48. Scheffel, H., Pfammatter, T., Wildi, S., Bauerfeind, P., Marincek, B., and Alkadhi, H. Acute gastrointestinal bleeding: detection of source and etiology with multi-detector-row CT. Eur. Radiol. 17: 1555-1565, 2007.

32. Arakawa, D., Ohmiya, N., Nakamura, M., Honda, W., Shirai, O., Itoh, A., Hirooka, Y., Niwa, Y., Maeda, O., Ando, T., and Goto, H. Outcome after enteroscopy for patients with obscure GI bleeding: diagnostic comparison between double-balloon endoscopy and videocapsule endoscopy. Gastrointest. Endosc. 69: 866-874, 2009. 33. Ramchandani, M., Reddy, D. N., Gupta, R., Lakhtakia, S., Tandan, M., Rao, G. V., and Darisetty, S. Diagnostic yield and therapeutic impact of single-balloon enteroscopy: series of 106 cases. J. Gastroenterol. Hepatol. 24: 1631-1638, 2009. 34. Upchurch, B. R., Sanaka, M. R., Lopez, A. R., and Vargo, J. J. The clinical utility of single-balloon enteroscopy: a single-center experience of 172 procedures. Gastrointest. Endosc. 71: 1218-1223, 2010. 35. Frantz, D. J., Dellon, E. S., Grimm, I. S., and Morgan, D. R. Single-balloon enteroscopy: results from an initial experience at a U.S. tertiarycare center. Gastrointest. Endosc. 72: 422-426, 2010. 36. May, A., Farber, M., Aschmoneit, I., Pohl, J., Manner, H., Lotterer, E., Moschler, O., Kunz, J., Gossner, L., Monkemuller, K., and Ell, C. Prospective multicenter trial comparing push-and-pull enteroscopy with the single- and double-balloon techniques in patients with small-bowel disorders. Am. J. Gastroenterol. 105: 575-581, 2010. 37. Aktas, H., de Ridder, L., Haringsma, J., Kuipers, E. J., and Mensink, P. B. Complications of single-balloon enteroscopy: a prospective evaluation of 166 procedures. Endoscopy 42: 365-368, 2010. 38. Bocker, U., Dinter, D., Litterer, C., Hummel, F., Knebel, P., Franke, A., Weiss, C., Singer, M. V., and Lohr, J. M. Comparison of magnetic resonance imaging and video capsule enteroscopy in diagnosing smallbowel pathology: localization-dependent diagnostic yield. Scand. J. Gastroenterol. 45: 490-500, 2010. 39. Cellier, C. Obscure gastrointestinal bleeding: role of videocapsule and double-balloon enteroscopy. Best Pract. Res. Clin. Gastroenterol. 22: 329-340, 2008. 40. Fireman, Z., and Friedman, S. Diagnostic yield of capsule endoscopy in

49. Jaeckle, T., Stuber, G., Hoffmann, M. H., Jeltsch, M., Schmitz, B. L., and Aschoff, A. J. Detection and localization of acute upper and lower gastrointestinal (GI) bleeding with arterial phase multi-detector row helical CT. Eur. Radiol. 18: 1406-1413, 2008. 50. Ko, H. S., Tesdal, K., Dominguez, E., Kaehler, G., Sadick, M., Duber, C., and Diehl, S. Localization of bleeding using 4-row detector-CT in patients with clinical signs of acute gastrointestinal hemorrhage. Rofo 177: 1649-1654, 2005. 51. Onoda, H., Shimizu, K., Washida, Y., Matsunaga, N., Higaki, S., Hashimoto, S., Matsunaga, T., and Sakaida, I. Usefulness of computed tomography enteroclysis in the intestinal tract. Jpn. J. Radiol. 28: 193-198, 2010. 52. Huprich, J. E., Fletcher, J. G., Alexander, J. A., Fidler, J. L., Burton, S. S., and McCullough, C. H. Obscure gastrointestinal bleeding: evaluation with 64-section multiphase CT enterography--initial experience. Radiology 246: 562-571, 2008. 53. Hara, A. K., Walker, F. B., Silva, A. C., and Leighton, J. A. Preliminary estimate of triphasic CT enterography performance in hemodynamically stable patients with suspected gastrointestinal bleeding. AJR Am. J. Roentgenol. 193: 1252-1260, 2009. 54. Frattaroli, F. M., Casciani, E., Spoletini, D., Polettini, E., Nunziale, A., Bertini, L., Vestri, A., Gualdi, G., and Pappalardo, G. Prospective study comparing multi-detector row CT and endoscopy in acute gastrointestinal bleeding. World J. Surg. 33: 2209-2217, 2009. 55. Yoon, W., Jeong, Y. Y., Shin, S. S., Lim, H. S., Song, S. G., Jang, N. G., Kim, J. K., and Kang, H. K. Acute massive gastrointestinal bleeding: detection and localization with arterial phase multi-detector row helical CT. Radiology 239: 160-167, 2006. 56. Jacobson, A. F., and Cerqueira, M. D. Prognostic significance of late imaging results in technetium-99m-labeled red blood cell gastrointestinal bleeding studies with early negative images. J. Nucl. Med. 33: 202-207, 1992. 57. Howarth, D. M. The role of nuclear medicine in the detection of acute gastrointestinal bleeding. Semin. Nucl. Med. 36: 133-146, 2006.

www.JSurgRad.com

39


Original Article 58. Gupta, S., Luna, E., Kingsley, S., Prince, M., and Herrera, N. Detection of gastrointestinal bleeding by radionuclide scintigraphy. Am. J. Gastroenterol. 79: 26-31, 1984. 59. Abdel-Dayem, H. M., Mahajan, K. K., Ericsson, S., Nawaz, K., Owunwanne, A., Kouris, K., Higazy, E., and Awdeh, M. Evaluation of technetium-99m DTPA for localization of site of acute upper gastrointestinal bleeding. Clin. Nucl. Med. 11: 788-791, 1986. 60. Voeller, G. R., Bunch, G., and Britt, L. G. Use of technetium-labeled red blood cell scintigraphy in the detection and management of gastrointestinal hemorrhage. Surgery 110: 799-804, 1991. 61. Kurosawa, S., Kuwata, H., Kushibiki, K., Akimoto, K., Hashimoto, T., and Kojima, T. The value of RI scintigraphy and angiography in small intestinal bleeding--report of eight cases. Gastroenterol. Jpn. 26 Suppl 3: 129-132, 1991.

Kobayashi et al. Diagnosis of Upper GI Bleeding 77. Kim, C. Y., Suhocki, P. V., Miller, M. J.,Jr, Khan, M., Janus, G., and Smith, T. P. Provocative mesenteric angiography for lower gastrointestinal hemorrhage: results from a single-institution study. J. Vasc. Interv. Radiol. 21: 477-483, 2010. 78. Johnston, C., Tuite, D., Pritchard, R., Reynolds, J., McEniff, N., and Ryan, J. M. Use of provocative angiography to localize site in recurrent gastrointestinal bleeding. Cardiovasc. Intervent. Radiol. 30: 1042-1046, 2007. 79. Bloomfeld, R. S., Smith, T. P., Schneider, A. M., and Rockey, D. C. Provocative angiography in patients with gastrointestinal hemorrhage of obscure origin. Am. J. Gastroenterol. 95: 2807-2812, 2000. 80. Malden, E. S., Hicks, M. E., Royal, H. D., Aliperti, G., Allen, B. T., and Picus, D. Recurrent gastrointestinal bleeding: use of thrombolysis with anticoagulation in diagnosis. Radiology 207: 147-151, 1998.

62. Dusold, R., Burke, K., Carpentier, W., and Dyck, W. P. The accuracy of technetium-99m-labeled red cell scintigraphy in localizing gastrointestinal bleeding. Am. J. Gastroenterol. 89: 345-348, 1994.

81. Bakalar, R. S., Tourigny, P. R., Silverman, E. D., and Schub, R. O. Provocative red blood cell scintiscan in occult chronic gastrointestinal hemorrhage. Clin. Nucl. Med. 19: 945-948, 1994.

63. Howarth, D. M., Tang, K., and Lees, W. The clinical utility of nuclear medicine imaging for the detection of occult gastrointestinal haemorrhage. Nucl. Med. Commun. 23: 591-594, 2002.

82. Anthony, S., Milburn, S., and Uberoi, R. Multi-detector CT: review of its use in acute GI haemorrhage. Clin. Radiol. 62: 938-949, 2007.

64. O'Neill, B. B., Gosnell, J. E., Lull, R. J., Schecter, W. P., Koch, J., Halvorsen, R. A., and Harris, H. W. Cinematic nuclear scintigraphy reliably directs surgical intervention for patients with gastrointestinal bleeding. Arch. Surg. 135: 1076-81; discussion 1081-2, 2000. 65. Zuckerman, G. R., and Prakash, C. Acute lower intestinal bleeding: part I: clinical presentation and diagnosis. Gastrointest. Endosc. 48: 606-617, 1998. 66. Miller, M.,Jr, and Smith, T. P. Angiographic diagnosis and endovascular management of nonvariceal gastrointestinal hemorrhage. Gastroenterol. Clin. North Am. 34: 735-752, 2005. 67. Lim, J. K., and Ahmed, A. Endoscopic approach to the treatment of gastrointestinal bleeding. Tech. Vasc. Interv. Radiol. 7: 123-129, 2004. 68. Kim, J. H., Shin, J. H., Yoon, H. K., Chae, E. Y., Myung, S. J., Ko, G. Y., Gwon, D. I., and Sung, K. B. Angiographically negative acute arterial upper and lower gastrointestinal bleeding: incidence, predictive factors, and clinical outcomes. Korean J. Radiol. 10: 384-390, 2009. 69. Loffroy, R., Rao, P., Ota, S., De Lin, M., Kwak, B. K., and Geschwind, J. F. Embolization of Acute Nonvariceal Upper Gastrointestinal Hemorrhage Resistant to Endoscopic Treatment: Results and Predictors of Recurrent Bleeding. Cardiovasc. Intervent. Radiol., 2010. 70. Gunderman, R., Leef, J., Ong, K., Reba, R., and Metz, C. Scintigraphic screening prior to visceral arteriography in acute lower gastrointestinal bleeding. J. Nucl. Med. 39: 1081-1083, 1998. 71. Duvnjak, S., and Andersen, P. E. The effect of transcatheter arterial embolisation for nonvariceal upper gastrointestinal bleeding. Dan. Med. Bull. 57: A4138, 2010. 72. Charbonnet, P., Toman, J., Buhler, L., Vermeulen, B., Morel, P., Becker, C. D., and Terrier, F. Treatment of gastrointestinal hemorrhage. Abdom. Imaging 30: 719-726, 2005. 73. Poultsides, G. A., Kim, C. J., Orlando, R.,3rd, Peros, G., Hallisey, M. J., and Vignati, P. V. Angiographic embolization for gastroduodenal hemorrhage: safety, efficacy, and predictors of outcome. Arch. Surg. 143: 457-461, 2008. 74. Ripoll, C., Banares, R., Beceiro, I., Menchen, P., Catalina, M. V., Echenagusia, A., and Turegano, F. Comparison of transcatheter arterial embolization and surgery for treatment of bleeding peptic ulcer after endoscopic treatment failure. J. Vasc. Interv. Radiol. 15: 447-450, 2004.

97. Tizon-Marcos, H., and Barbeau, G. R. Incidence of compartment syndrome of the arm in a large series of transradial approach for coronary procedures. J. Interv. Cardiol. 21: 380-384, 2008. 98. Fruhwirth, J., Pascher, O., Hauser, H., and Amann, W. Local vascular complications after iatrogenic femoral artery puncture. Wien. Klin. Wochenschr. 108: 196-200, 1996.

99. Meyerson, S. L., Feldman, T., Desai, T. R., Leef, J., Schwartz, L. B., and McKinsey, J. F. Angiographic access site complications in the era of arterial closure devices. Vasc. Endovascular Surg. 36: 137-144, 2002.

Visit www.SurgRad.com for high resolution images, interactive discussions, and a PDF version of this article.

84. Kuhle, W. G., and Sheiman, R. G. Detection of active colonic hemorrhage with use of helical CT: findings in a swine model. Radiology 228: 743-752, 2003. 85. Hilfiker, P. R., Weishaupt, D., Kacl, G. M., Hetzer, F. H., Griff, M. D., Ruehm, S. G., and Debatin, J. F. Comparison of three dimensional magnetic resonance imaging in conjunction with a blood pool contrast agent and nuclear scintigraphy for the detection of experimentally induced gastrointestinal bleeding. Gut 45: 581-587, 1999. 86. Hilfiker, P. R., Zimmermann-Paul, G. G., Schmidt, M., Klotz, H. P., Kacl, G. M., and Debatin, J. F. Intestinal and peritoneal bleeding: detection with an intravascular contrast agent and fast three-dimensional MR imaging--preliminary experience from an experimental study. Radiology 209: 769-774, 1998. 87. Morcos, S. K. Review article: Acute serious and fatal reactions to contrast media: our current understanding. Br. J. Radiol. 78: 686-693, 2005. 88. Haveman, J. W., Gansevoort, R. T., Bongaerts, A. H., and Nijsten, M. W. Low incidence of nephropathy in surgical ICU patients receiving intravenous contrast: a retrospective analysis. Intensive Care Med. 32: 1199-1205, 2006. 89. Lencioni, R., Fattori, R., Morana, G., and Stacul, F. Contrast-induced nephropathy in patients undergoing computed tomography (CONNECT) - a clinical problem in daily practice? A multicenter observational study. Acta Radiol. 51: 741-750, 2010. 90. Rashid, A. H., Brieva, J. L., and Stokes, B. Incidence of contrast-induced nephropathy in intensive care patients undergoing computerised tomography and prevalence of risk factors. Anaesth. Intensive Care 37: 968-975, 2009. 91. Pucelikova, T., Dangas, G., and Mehran, R. Contrast-induced nephropathy. Catheter. Cardiovasc. Interv. 71: 62-72, 2008. 92. Martin, D. R., Semelka, R. C., Chapman, A., Peters, H., Finn, P. J., Kalb, B., and Thomsen, H. Nephrogenic systemic fibrosis versus contrastinduced nephropathy: risks and benefits of contrast-enhanced MR and CT in renally impaired patients. J. Magn. Reson. Imaging 30: 13501356, 2009. 93. Tsetis, D. Endovascular treatment of complications of femoral arterial access. Cardiovasc. Intervent. Radiol. 33: 457-468, 2010.

76. Ryan, J. M., Key, S. M., Dumbleton, S. A., and Smith, T. P. Nonlocalized lower gastrointestinal bleeding: provocative bleeding studies with intraarterial tPA, heparin, and tolazoline. J. Vasc. Interv. Radiol. 12: 12731277, 2001.

95. Wang, C. L., Cohan, R. H., Ellis, J. H., Adusumilli, S., and Dunnick, N. R. Frequency, management, and outcome of extravasation of nonionic iodinated contrast medium in 69,657 intravenous injections. Radiology 243: 80-87, 2007.

Original Article | January 2011

96. Bellin, M. F., Jakobsen, J. A., Tomassin, I., Thomsen, H. S., Morcos, S. K., Thomsen, H. S., Morcos, S. K., Almen, T., Aspelin, P., Bellin, M. F., Clauss, W., Flaten, H., Grenier, N., Idee, J. M., Jakobsen, J. A., Krestin, G. P., Stacul, F., Webb, J. A., and Contrast Media Safety Committee Of The European Society Of Urogenital Radiology Contrast medium extravasation injury: guidelines for prevention and management. Eur. Radiol. 12: 2807-2812, 2002.

83. Stuber, T., Hoffmann, M. H., Stuber, G., Klass, O., Feuerlein, S., and Aschoff, A. J. Pitfalls in detection of acute gastrointestinal bleeding with multi-detector row helical CT. Abdom. Imaging 34: 476-482, 2009.

75. Schenker, M. P., Duszak, R.,Jr, Soulen, M. C., Smith, K. P., Baum, R. A., Cope, C., Freiman, D. B., Roberts, D. A., and Shlansky-Goldberg, R. D. Upper gastrointestinal hemorrhage and transcatheter embolotherapy: clinical and technical factors impacting success and survival. J. Vasc. Interv. Radiol. 12: 1263-1271, 2001.

40

Original Article

Kobayashi et al. Diagnosis of Upper GI Bleeding

94. Li, A., Wong, C. S., Wong, M. K., Lee, C. M., and Au Yeung, M. C. Acute adverse reactions to magnetic resonance contrast media--gadolinium chelates. Br. J. Radiol. 79: 368-371, 2006.

Editorial Commentary

I

n this issue of the Journal of Surgical Radiology, Kobayashi and colleagues review the latest approaches to diagnosis and management of upper gastrointestinal bleeding. As surgeons, radiologists, and gastroenterologists, we have all spent many sleepless nights taking care of patients with upper gastrointestinal bleeding, so this fresh look at this extremely common problem will be of great interest to most of you. There is no question that the diagnosis and treatment of this condition has evolved a great deal since the days of emergent exploratory laparotomies, frequent gastric resections, surgical vagotomies, and the oversewing of gastroduodenal artery bleeds. The advent of reliable chemical gastric acid suppression and advances in interventional endoscopy and radiology have made the surgical treatment of these problems quite rare.  Perhaps most exciting is the literature reviewed in this paper highlighting the use of computed tomoangiography (CTA) in the diagnosis of the source of bleeding in upper GI bleeds.  A shift to the use of this ubiquitous and user-independent technology as a first-line diagnostic test in these patients would certainly decrease the invasiveness of diagnosis as well as decrease the manpower resources necessary in the diagnosis phase of patients presenting with signs and symptoms of upper gastrointestinal bleeding.

Eric J. Hanly, MD Assistant Clinical Professor of Surgery, University of Colorado Laparoscopic, Bariatric, and General Surgery General Surgeons of Western Colorado, PC www.JSurgRad.com

41


ORIGINAL ARTICLE

Marathon-Induced Colitis David R. King, MD ∙ Laura Avery, MD‚  Division of Trauma, Emergency Surgery, and Surgical Critical Care, Department of Surgery, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts. ‚ Division of Emergency Radiology, Department of Radiology, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts.

Abstract

Overview Marathon runners are often thought to represent the prototype of exceptional human health. Although the vast majority of these athletes improve their health by marathoning, a rare few will experience a variety of abdominal maladies including cramping, vague pain, diarrhea, nausea, vomiting, hematochezia, and peritonitis from colonic ischemia. Marathoners have long joked about post-race loose stools; however with laboratory testing and axial imaging we now have a better understanding of this unfortunate disease process that afflicts some of the healthiest highperformance athletes on the planet. The pathophysiology, indications for imaging, radiographic findings, and management are discussed. Keywords colitis, marathon, runner, hematochezia, diarrhea

Introduction

J Surg Rad

A marathoner is a unique athlete. Ever since Pheidippides’ original 24.85 mile run from Marathon to Athens in 490 B.C. (and his subsequent immediate collapse and death upon arrival in Athens while delivering the victory message), the world has recognized the extraordinary human endurance required to complete the race. In the 1908 Olympic Games in London, the historic original marathon distance was lengthened to 26.22 miles in order that the finish line be placed exactly in front of the royal family’s viewing box in White Castle stadium. Since then, numerous studies have examined the impact of marathoning on physiology, cardiac function, electrolyte abnormalities, and even absolute risk of death associated with the event.

42

Despite Pheidippides’ death after his famous run, the risk of death from a sudden cardiovascular event while running a marathon has been estimated at 1 in 50,000.1 This represents 1/100th the risk of a normal American.1 Age of marathoners appears to not be associated with any particular increased risk of death either.2-5 Serum troponin levels are routinely elevated after completion of a marathon; however no particular health risk has been attributed to these findings with unclear cardiovascular significance.6-9 Troponin elevations are not ischemic in origin, are not Citation King DR, Avery L. Marathon-induced colitis. J Surg Radiol. 2011 Jan 1;2(1). Correspondence David R. King, MD E-mail dking3@partners.org. Received August 12, 2010. Accepted August 26, 2010. Epub September 12, 2010.

Original Article | January 2011

accompanied with pump failure, and may simply represent myocardial inflammation.10 These elevations are not age-dependent, and seem to decrease with greater marathon experience,9 suggesting an adaptive response to marathoning. A significant risk of illness after a marathon, however, is associated with severe hyponatremia. Hyponatremia may be present in a proportion of marathoners, and severe life-threatening hyponatremia present in 0.6% of racers.12 None of these findings, however, are remotely as common as gastrointestinal-associated complaints after marathoning, with diarrhea as the leading complaint in up to a third of all marathoners.13 Diarrhea is so common, in fact, that the vernacular of “Runner’s Trots” has been coined to describe the ubiquitous presence of loose stool following marathons.14 This diarrhea has been attributed to post-marathon ischemic colitis, likely representing the most common medical problem marathoners face.15

King and Avery. Marathon-Induced Colitis

Original Article Figure 1 A-B. 37 year old female who ran the Boston Marathon. Panels A (top) & B (bottom): low (fluid) attenuation tracing along both sides of the enhanced portal veins (black arrows). which a comparison can be made,21-23 the magnitude of which is as great as that seen with pulmonary artery catheterization, 30% total body surface area thermal injury, and even experimental loss of 60% blood volume combined with mild traumatic brain injury.24-26 This hypercoagulable state may cause intravascular sludging and microvascular thrombotic events within the delicate capillary network within the wall of the colon. Upon completing the race, reperfusion results in normal fibrinolysis and an acute inflammatory milieu is generated, resulting in colitis. Ischemia is prolonged and worsened by dehydration and volume contraction, problems exceedingly common among marathoners, even those with apparently appropriate volume and electrolyte replacement during the race.10,12

Pathophysiology

There is even evidence that the vasculature undergoes dramatic changes during racing, resulting is significant alterations of pressure and flow within the abdominal aorta.27 Adaptive changes to damaged mitochondria and cellular suprastructure are evident at the DNA level, in that marathoners appear to upregulate anti-apoptotic genes and mediators following a race, possibly allowing compromised tissues a chance at recovery instead of undergoing the predetermined cell death due to oxidative stress generated from the insult of the race.28,29 A leukemoid response ensues and quickly resolves, suggesting an initial response that becomes blunted, or even directly inhibited.30 A delicate interaction between pro-apoptotic and anti-apoptotic genes, involving these activated leukocytes, takes place during which time a cumulative anti-apoptotic environment is favored.31

Up to 20% of marathoners will have biochemical evidence of microscopic blood in their stools upon post-race fecal occult blood testing (“heme-positive”).16 Multiple cases and analyses of these cases suggest ischemia and reperfusion as the dominant mechanism for inducing colitis after a marathon.17-20 Shunting and redirection of cardiac output to the skeletal muscles presumably leaves the gastrointestinal tract in a state of acutely diminished blood flow for the duration of the marathon.

All of these physiologic and cellular level changes are likely adaptive, since it appears that the more experienced marathoners have far fewer medical complaints after the race than do first-time marathoners.32 In fact, virgin marathoners are far more likely to experience lower gastrointestinal problems than are experienced racers.2,5 Colitis remains the most common enemy.15

Additionally, marathoning appears to induce a systemic hypercoagulable state, more so than any other sport to www.JSurgRad.com

43


Original Article

King and Avery. Marathon-Induced Colitis

King and Avery. Marathon-Induced Colitis

Original Article

Figure 1 C-E. 37 year old female who ran the Boston Marathon. Panel C (top): bowel wall thickening of the distal transverse colon extending to the splenic flexure (white arrows). Panels D (bottom left) & E (bottom right): bowel wall thickening of the descending colon (white arrow head). Magnified image (E) demonstrates enhancement of the mucosal (white arrow) and adventitial (black arrow) layers in the descending colon.

Indications For Imaging Clearly, the prerequisite history of recent marathoning is required to consider the diagnosis. The diagnosis might also be considered in other serious endurance sports as well, such as triathlon, â&#x20AC;&#x153;head-distanceâ&#x20AC;? rowing, and long-distance bicycling, although there are no specific reports in the literature of colonic ischemia associated with these sports. Patients commonly present with crampy, poorly localized, vague abdominal pain of visceral origin.33-35 Diarrhea is common and most patients presenting to the hospital will test positive for fecal occult blood.16,33 Metabolic abnormalities may be present on laboratory testing, including (in some cases) profound electrolyte abnormalities.33-35 The electrolyte abnormalities are varied and largely related to the athleteâ&#x20AC;&#x2122;s fluid and electrolyte consumption during

Figure 2. 40 year old marathoner. Panels A (top left) & B (top right): bowel wall thickening of the distal transverse colon at the level of the splenic flexure. Magnification view demonstrates wall thickening and mild pericolonic fat stranding (arrowhead). Panels C (bottom left) & D (bottom right): demonstrates circumferential low-attenuation bowel wall thickening consistent with edematous submucosa (black arrows). the race. Most marathoners are only mildly dehydrated since the introduction of balanced electrolyte replacement on nearly every fully-supported marathon course in the world.5,12 The total white blood cell count is routinely elevated.12,30 Some degree of acidosis is also commonly present.33-35 On exam, findings will vary from non-tender to peritonitis, depending on the degree of ischemia. Certainly focal perito-

44

Original Article | January 2011

neal findings are concerning and should prompt a surgical consultation immediately. Patients with any abdominal tenderness should be managed as inpatients by a surgical team familiar with this disease process. Once the diagnosis is suspected based upon the appropriate history, imaging may be considered based upon the clinical picture. Patients with no tenderness or only mild tenderness www.JSurgRad.com

45


Original Article

King and Avery. Marathon-Induced Colitis

Original Article

King and Avery. Marathon-Induced Colitis

Figure 3 A-B. 73 year old man after a race. Panel A (top): demonstrates bowel wall thickening of the ascending colon (white arrow). Note peri colonic fat stranding (white arrowhead). Panel B (bottom): circumferential nature of the low attenuatin bowel wall thickening. Note relative hyperenhancement of the bowel mucosa (white arrowhead). do not require plain film or cross sectional imaging. Computed tomography (CT) for the sole purpose of demonstrating colonic inflammation is unnecessary and not indicated. In these patients, even the presence of extensive severe colitis seen by CT would not alter management. Patients with significant abdominal tenderness and localizing signs, however, should undergo CT imaging to exclude perforation (contained or free). These findings will dictate important alterations in surgical management. Oral contrast is not required for the study and may even interfere with the diagnostic value of the study by obscuring colonic mucosal changes related to ischemia. Intravenous contrast is recommended and should be administered if there is no evidence of acute pre-renal contraction or chronic renal failure. Rectal contrast is unnecessary. It is interesting to note that, after reviewing all the cases described in the literature, no cases of marathon-induced colitis have ever been reported in subthree hour marathoners. Generally speaking, marathoners who run below three hours are considered elite runners. The lack of colitis in this population suggests an adaptive, perhaps protective, effect of training to become an elite marathoner and is consistent with our understanding of the pathophysiology of the condition. Any patient whose diagnosis is uncertain and symptoms warrant imaging should undergo CT evaluation. This is especially true in older marathoners with non-classic symptoms and concern for additional abdominal processes such as aortic aneurysms or dissection.

Typical Radiographic Findings Figure 1 is a contrast enhanced CT scan of a 37 year old woman who completed the Boston Marathon. She presented in the classical nature as described above. Panels A and B are axial images from her CT at the level of the liver. These images demonstrate periportal edema related to fluid resuscitation. Note the low (fluid) attenuation tracing along both sides of the enhanced portal veins (black arrows). The circumferential nature of the fluid is consistent

46

Original Article | January 2011

Figure 3 C-D. 73 year old man after a race. Panel C (top): demonstrates bowel wall thickening at the ascending colon. Panel D (bottom): relative hyperenhancement of thickened ascending colon (white arrowhead). with retention of fluid in the periportal interstitial space, as is frequently seen in over-resuscitated patients. The circumferential nature of the fluid attenuation distinguishes the cause from bile duct dilatation which would travel on one side of the vein. Panel C is a coronal reformatted image demonstrating bowel wall thickening most pronounced at the level of the distal transverse colon extending to the splenic flexure (white arrows). Panels D and E are axial images with magnification inset demonstrating bowel wall thickening of the descending colon (white arrow head). Magnified image (E) demonstrates slight enhancement of the mucosal (white arrow) and adventitial (black arrow) layers in the descending colon. There is moderate circumferential low-attenuation bowel wall thickening consistent with edematous submucosa Figure 2 is a contrast enhanced CT scan from a 40 year old woman who also completed the Boston Marathon and developed prototypical symptoms. Panels A and B are coronal reformatted images with magnification inset. This demonstrates bowel wall thickening of the distal transverse colon at the level of the splenic flexure. Magnification view demonstrates wall thickening and mild pericolonic fat stranding (arrowhead). Note intraluminal contrast obscures visualization of the colonic mucosa. Panels C and D and axial images through the level of the splenic flexure demonstrates circumferential lowattenuation bowel wall thickening consistent with edematous submucosa (black arrows). Figure 3 is a contrast enhanced CT scan of a 73 year-old man with a similar presentation as the previous patients. Panel A is a coronal reformatted image demonstrating bowel wall thickening of the ascending colon (white arrows). Note pericolonic fat stranding (white arrowhead). Panel B is an axial image demonstrating the circumferential nature of the low attenuation bowel wall thickening. Note relative hyperenhancement of the bowel mucosa (White arrowhead)

As is demonstrated in the cases above, colonic ischemia may occur at any point in the colon, and is not limited to “watershed” areas. It is not surprising that any area of the colon may be affected since superior mesenteric blood flow may be reduced as much as 70% in humans during intense running (36). Consequently, clinicians should maintain vigilance that any portion of the colon may be affected.

Management Reversible ischemia is the rule, with nearly all patients recovering after appropriate rehydration, electrolyte replacement, and rest.13-20,37,38 For patients with impressive exams and CT findings, serial exams are indicated since even the most severely inflamed colons will recover without surgery. The role of antibiotics in these patients is unclear and is left to individual clinician judgment, although extrapolation from other colonic disease states would suggest no utility in antimicrobial coverage. For an extremely small subset of patients, colitis may progress to frank necrosis or perforation requiring operation.34,36,39 This will be detected clinically in a patient undergoing serial exams, and repeat CT imaging is not indicated. However, an unfortunate trip to the operating room will be required. Patients with evidence of perforation on their initial imaging should proceed to immediate laparotomy after adequate fluid and electrolyte resuscitation. The need for appropriate preoperative resuscitation is particularly important since proper fluid resuscitation may lessen the distribution of ischemic involvement of the colon, thus minimizing the resultant length of resection. When possible a laparoscopic approach should be considered in all patients.

www.JSurgRad.com

47


Original Article

Conclusion Why Pheidippides died will remain a mystery, however his death articulates the unusual physiologic insult that accompanies marathoning. The modern marathoner is generally well hydrated (even at completion of the race), has exceptional cardiovascular fitness, and appears to be protected from the general population-based risks for acute coronary events. There remain, however, some racers who, through a combination of dehydration, a hypercoagulable vasculature, blood shunting, and under-training, will develop marathon induced colitis. The diagnosis is made based upon history and examination. Axial imaging will reveal colitis that may involve any portion of the colon, even the cecum and right colon, which many believe to be “protected” from ischemic insults by its substantial blood supply. Other important radiographic markers of the disease include periportal edema (seen after aggressive fluid resuscitation), mucosal enhancement, and low-attenuation of the submucosa. Most patients will recover with non-operative management.

Disclosures

King and Avery. Marathon-Induced Colitis 11. Fortescue EB, Shin AY, Greenes DS, Mannix RC, Agarwal S, Feldman BJ, Shah MI, Rifai N, Landzberg MJ, Newburger JW, Almond CS. Cardiac troponin increases among runners in the Boston Marathon. Ann Emerg Med. 2007 Feb;49(2):137-43, 143.e1. 12. Almond CS, Shin AY, Fortescue EB, Mannix RC, Wypij D, Binstadt BA, Duncan CN, Olson DP, Salerno AE, Newburger JW, Greenes DS. Hyponatremia among runners in the Boston Marathon. N Engl J Med. 2005 Apr 14;352(15):1550-6. 13. Cantwell J (1980) Gastrointestinal disorders in runners. JAMA 246:1404–1405 14. Fogoros R (1980) ‘‘Runner’s trots’’ Gastrointestinal disturbances in runners. JAMA 243:1743–1744. 15. Ho GW. Lower gastrointestinal distress in endurance athletes. Curr Sports Med Rep. 2009 Mar-Apr;8(2):85-91.

18. Moses F, Brewer T, Peura D (1988) Running associated proximal hemorrhagic colitis. Ann Intern Med 108:385–386 19. Heer M, Repond F, Hany A, et al. (1987) Acute ischaemic colitis in a female long distance runner. Gut 7:896–899 20. Schaub N, Spichtin HP, et al. (1985) Ischemic colitis as a cause for intestinal bleeding after marathon running. Schweiz Med Wochenschr.115:454–457

References

22. Weiss C, Welsch B, Albert M, Friedmann B, Strobel G, Jost J, Nawroth P, Bärtsch P. Coagulation and thrombomodulin in response to exercise of different type and duration. Med Sci Sports Exerc. 1998 Aug;30(8):1205-10.

3. Holtzhausen LM, Noakes TD, Kroning B, de Klerk M, Roberts M, Emsley R. Clinical and biochemical characteristics of collapsed ultramarathon runners. Med Sci Sports Exerc. 1994 Sep;26(9):1095-101. 4. Roberts WO, Maron BJ. Evidence for decreasing occurrence of sudden cardiac death associated with the marathon. J Am Coll Cardiol. 2005 Oct 4;46(7):1373-4. 5. Roberts WO. A 12-yr profile of medical injury and illness for the Twin Cities Marathon. Med Sci Sports Exerc. 2000 Sep;32(9):1549-55. 6. Regwan S, Hulten EA, Martinho S, Slim J, Villines TC, Mitchell J, Slim AM. Marathon Running as a Cause of Troponin Elevation: A Systematic Review and Meta-Analysis. J Interv Cardiol. 2010 Jul 20. 7. Mingels A, Jacobs L, Michielsen E, Swaanenburg J, Wodzig W, van Dieijen-Visser M. Reference population and marathon runner sera assessed by highly sensitive cardiac troponin T and commercial cardiac troponin T and I assays. Clin Chem. 2009 Jan;55(1):101-8. Epub 2008 Nov 6. 8. Jassal DS, Moffat D, Krahn J, Ahmadie R, Fang T, Eschun G, Sharma S. Cardiac injury markers in non-elite marathon runners. Int J Sports Med. 2009 Feb;30(2):75-9. 9. Michielsen EC, Wodzig WK, Van Dieijen-Visser MP. Cardiac troponin T release after prolonged strenuous exercise. Sports Med. 2008;38(5):425-35. 10. O'Hanlon R, Wilson M, Wage R, Smith G, Alpendurada FD, Wong J, Dahl A, Oxborough D, Godfrey R, Sharma S, Roughton M, George K, Pennell DJ, Whyte G, Prasad SK. Troponin release following endurance exercise: is inflammation the cause? a cardiovascular magnetic resonance study. J Cardiovasc Magn Reson. 2010 Jul 2;12:38.

48

Original Article | January 2011

33. Cohen DC, Winstanley A, Engledow A, Windsor AC, Skipworth JR. Marathon-induced ischemic colitis: why running is not always good for you. Am J Emerg Med. 2009 Feb;27(2):255.e5-7.

37. Lucas W, Schroy PC 3rd. Reversible ischemic colitis in a high endurance athlete. Am J Gastroenterol. 1998 Nov;93(11):2231-4. 38. Sanchez LD, Tracy JA, Berkoff D, Pedrosa I. Ischemic colitis in marathon runners: a case-based review. J Emerg Med. 2006 Apr;30(3):321-6. 39. Beaumont AC, Teare JP (1991) Subtotal colectomy following marathon running in a female patient. R Soc Med 84:39–40.

34. Egloff BP, Dombrowsky J, McKinnon H, O'Connor FG. Abdominal pain in a marathon runner. Curr Sports Med Rep. 2009 MarApr;8(2):49-51. 35. Viola TA. Evaluation of the athlete with exertional abdominal pain. Curr Sports Med Rep. 2010 Mar-Apr;9(2):106-10. 36. Heer M, Repond F, Hany A, Sulser H, Kehl O, Jäger K. Acute ischaemic colitis in a female long distance runner. Gut. 1987 Jul;28(7):896-9.

Visit www.SurgRad.com for high resolution images, interactive discussions, and a PDF version of this article.

17. Lucas W, Schroy P (1998) Reversible ischemic colitis in a high endurance athlete. Am J Gastroenterol 93:2231–2234

The authors have no disclosures or conflicts of interest related to this manuscript.

2. Roberts WO, Nicholson WG. Youth marathon runners and race day medical risk over 26 years. Clin J Sport Med. 2010 Jul;20(4):318-21.

ar cells. Eur J Appl Physiol. 2008 Nov;104(4):711-7 32. Nguyen RB, Milsten AM, Cushman JT. Injury patterns and levels of care at a marathon. Prehosp Disaster Med. 2008 Nov-Dec;23(6):519-25.

16. Mcmahon LF, Ryan MJ, Larson DL, Fischer RL (1984) Occult gastrointestinal blood loss in marathon runners. Ann Intern Med 100:846–847.

21. Hanke AA, Staib A, Görlinger K, Perrey M, Dirkmann D, Kienbaum P. Whole blood coagulation and platelet activation in the athlete: a comparison of marathon, triathlon and long distance cycling. Eur J Med Res. 2010 Feb 26;15(2):59-65.

1. Maron BJ, Poliac LC, Roberts WO. Risk for sudden cardiac death associated with marathon running. J Am Coll Cardiol. 1996 Aug;28(2):42831.

Original Article

King and Avery. Marathon-Induced Colitis

23. Sucker C, Zotz RB, Senft B, Scharf RE, Kroger K, Mohlenkamp S. Exercise-Induced Hemostatic Alterations Are Detectable by Rotation Thrombelastography (ROTEM): A Marathon Study. Clin Appl Thromb Hemost. 2009 Aug 19. 24. King DR, Cohn SM, Feinstein AJ, Proctor KG. Systemic coagulation changes caused by pulmonary artery catheters: laboratory findings and clinical correlation. J Trauma. 2005 Oct;59(4):853-7; discussion 857-9. 25. King DR, Namias N, Andrews DM. Coagulation abnormalities following thermal injury. Blood Coagul Fibrinolysis. 2010 Aug 4 26. King DR, Cohn SM, Proctor KG. Changes in intracranial pressure, coagulation, and neurologic outcome after resuscitation from experimental traumatic brain injury with hetastarch. Surgery. 2004 Aug;136(2):355-63. 27. Vlachopoulos C, Kardara D, Anastasakis A, Baou K, Terentes-Printzios D, Tousoulis D, Stefanadis C. Arterial Stiffness and Wave Reflections in Marathon Runners. Am J Hypertens. 2010 May 20 28. Marfe G, Tafani M, Pucci B, Di Stefano C, Indelicato M, Andreoli A, Russo MA, Sinibaldi-Salimei P, Manzi V. The effect of marathon on mRNA expression of anti-apoptotic and pro-apoptotic proteins and sirtuins family in male recreational long-distance runners. BMC Physiol. 2010 May 12;10:7. 29. Marfe G, Tafani M, Pucci B, Di Stefano C, Indelicato M, Andreoli A, Russo MA, Sinibaldi-Salimei P, Manzi V. The effect of marathon on mRNA expression of anti-apoptotic and pro-apoptotic proteins and sirtuins family in male recreational long-distance runners. BMC Physiol. 2010 May 12;10:7. 30. Lippi G, Banfi G, Montagnana M, Salvagno GL, Schena F, Guidi GC. Acute variation of leucocytes counts following a half-marathon run. Int J Lab Hematol. 2010 Feb;32(1 Pt 2):117-21 31. Atamaniuk J, Stuhlmeier KM, Vidotto C, Tschan H, Dossenbach-Glaninger A, Mueller MM. Effects of ultra-marathon on circulating DNA and mRNA expression of pro- and anti-apoptotic genes in mononucle-

Editorial Commentary

F

ew will argue of the concept of the marathon runner as the athlete-elite. Nonetheless, these veritable athletes did not acheive this status from the day they conceived the notion that they were going to train for this glorified and somewhat punitive competition to the day they achieved nonpareil status. To that point, the marathon race is simply the culmination of hours upon hour of unparalleled, disciplined, often boring and occasionally deleterious training and roadwork. One may argue as to what makes the marathon runner different from the wrestler, from the rower, from the soccer player or the cyclist? Furthermore, those of us who have competed at an elite level, who has not suffered accelerated gastrointestinal transit so thoughtfully coined “the Trots,” just prior to or during competition? The authors bring to us an important piece of literature and history that few authors (Sanchez, Corwell, Berkoff; Cohen, Winstall, et al; Babic, Papa, et al, and select others) have recognized. Through their efforts, we are encouraged as treating physicians the importance of recognizing this unique condition that if approached in ignorance may lead to complications of overtreatment and excessive and wasteful expense. Ischemic colitis is often thought of in terms of the aged and the vasculopathic patient.; however, we must recognize the constellation of signs of crampy abdominal pain, loose stools, and hematochezia, as ischemic colitis in this select group of athletes who present to their physicians. Nonessential CT scans with hypertonic contrast in the setting of low flow states –especially in watershed areas- and unsupported use of antibiotics for extended periods, laparotomy, and other diagnostics may do more harm than benefit. These are the important educational points one must etch in their consciousness when

searching for a reasonable cause and ultimately a diagnosis, when examining an athlete during or just after vigorous, yet subjectively masochistic, exercise. Still, one must question this entity and why it is a relatively uncommon diagnosis. The dearth of literature on this subject is not limited to Marathon runners, but competitive rowers, rugby players, and cyclists. Clearly, the endurance and the literal pounding and low financial cost of running in comparison to the other sports may lead to the abundance of representation in marathon runners, but are we simply missing the diagnosis in others? In the Sanchez paper, one athlete experienced all the pertinent symptoms within 5 miles. Others, were symptomatic within 10 miles or slightly more. There was no mention of what the pregame (run) meal was or what they did preceding the competition. Carbohydrate loading? Salicylates? NSAIDS? All may give one similar symptoms and yet, there was no mention of this variably practiced ritual for some athletes. In summary, Marathon-Induced Colitis is a distinct yet perhaps more common entity than has been recognized. In fact, it may not be isolated to Marathon running, but competitive exercise in general. The author does a remarkable job of educating the readers in the presentation, diagnosis, and the specifics of management in what appears to be post-exercise ischemic colitis in otherwise healthy, competitive, athletes.

Mark L. Shapiro, MD

Associate Professor of Surgery Division of Trauma and Critical Care Department of Surgery Duke University Medical Center Durham, North Carolina

www.JSurgRad.com

49


ORIGINAL ARTICLE

Stents in Tents: Endovascular Therapy on the Battlefields of the Global War on Terror

O'Brien and Cox. Stents in Tents

Equipment in Combat Environments The initial phases of the conflicts in Afghanistan and Iraq were characterized by rapid, large-scale troop movements as territory was taken, and required mobile, minimally equipped forward surgical teams (FST’s) that could advance with combat forces.3 However, by 2004, the conflicts, particularly in Iraq, had settled into a more static phase as US and coalition troops confronted an ongoing violent insurgency. At this point, a true trauma system was instituted, with relatively robust medical capacity at designated combat support hospitals (CSH) placed at strategic locations in-theater—the rough equivalent of civilian level I trauma centers.4 Despite being housed in primitive-appearing structures, these facilities offered modern ICU care, 3-4 operating rooms, inpatient wards, advanced imaging, and multiple surgical and medical subspecialists (Figure 3).

P. Joshua O'Brien, MD ∙ Mitchell W. Cox, MD Division of Vascular Surgery, Department of Surgery, Duke University Medical Center, Durham, North Carolina.

Abstract

Overview Since late 2001, the United States military has been engaged in a series of conflicts known collectively as the Global War on Terror (GWOT). Medical capabilities have changed significantly since our last major conflict in the early 1970’s, and the field of vascular surgery in particular has been radically altered, with improved imaging and a profusion of new endovascular devices. We will discuss how modern military vascular surgeons are now bringing the latest in endovascular technology to bear on horrifically injured troops in the field, very close to the point of injury. Keywords vascular surgery, military, trauma, gunshot wounds, CTA, imaging

Introduction

J Surg Rad

Since late 2001, the United States military has been engaged in a series of conflicts known collectively as the Global War on Terror (GWOT). Unfortunately, we have also seen the largest number of battlefield casualties since the Vietnam War during this period. Medical capabilities have changed significantly since our last major conflict in the early 1970’s, and the field of vascular surgery in particular has been radically altered, with improved imaging and a profusion of new endovascular devices. Employing these peacetime advances effectively on the modern battlefield has been the task of the military vascular surgeon.

50

The popular concept of combat medicine is of surgeons working feverishly in makeshift hospitals with limited, low technology equipment, and minimal subspecialty availability. While this may sometimes be the case, trauma care in the field increasingly resembles trauma care at level 1 trauma centers in the United States, with the availability of advanced imaging and high-level subspecialty support. More specifically, modern military vascular surgeons are now bringing the latest in endovascular techCitation O'Brien PJ, Cox MW. Stents in tents: endovascular therapy on the battlefields of the global war on terror. J Surg Radiol. 2011 Jan 1;2(1). Correspondence Mitchell W. Cox, MD E-mail mitchell.cox2@duke.edu. Received August 21, 2010. Accepted September 13, 2010. Epub September 14, 2010.

Original Article | January 2011

nology to bear on horrifically injured troops in the field, very close to the point of injury.

Nature of Military Trauma Injuries on the battlefield are often different from civilian trauma with respect to both the mechanism and extent of injury. While blunt trauma from motor vehicle accidents and low-velocity penetrating injuries dominate most civilian series, high-velocity gun-shot wounds (GSW) and blast injuries are characteristic of the military experience.1 In the GWOT, 25% of vascular injuries have been due to high-velocity gunshot wounds, while 64% result from explosive devices such as mortars, land mines, and improvised explosive devices (IED’s).2 With blast injuries in particular, there are typically innumerable penetrating fragments which involve multiple organ systems and may produce extensive soft-tissue wounds and complex fractures. Figure 1 illustrates a cervical wound with multiple small fragments spread across all three of the traditional zones of the neck. Distribution of wartime vascular injuries follows a characteristic pattern which is also somewhat different from the civilian experience. As documented by Fox et al, 97% of diagnosed vascular injuries involved the extremities or the neck. The surprising lack of truncal vascular injuries is likely accounted for by the use of body armor which protects the chest and abdomen (Figure 2).2 It may also be the case that truncal vascular injuries in the setting of blast or high velocity GSW were highly lethal at the point of injury and thus unlikely to present to medical facilities.

Original Article

Multi-slice CT scanners capable of CT angiography (CTA) were available at these CSH’s as well as portable fluoroscopy equipment capable of digital subtraction angiography, and portable duplex ultrasound. Two Figure 1. Multiple fragments spread across all three conventional of these hospitals in Iraq, at Balad and Bagdad, were continuously staffed with vascular surgeons capable of zones of the neck. both traditional surgical procedures and endovascular intervention. Endovascular capabilities were continuously updated at the insistence of the vascular surgery staff, and eventually included an OEC 9800 C-Arm (GE Healthcare, Waukesha, WI), a dedicated imaging table, and a full range of wires, catheters, stent-grafts, and embolization coils. The imaging suite in use during early 2006 is illustrated in Figure 4.

Approach to Military Vascular Injuries

Figure 2. Interceptor Body Armor System. The iconic injury of the war may be the blast-related mangled extremity in a patient who ultimately survives due to protective body armor.

The principles for management of vascular trauma are no different on the battlefield then in a university medical center. It is, however, often challenging to prioritize multiple injuries in patients with complex injury patterns that can encompass nearly the entire body (Figure 5). Blast injuries in particular, can produce thousands of tiny projectiles, any one of which might result in a significant underlying vascular injury with only minimal external evidence of trauma. In such a situation, the evaluation is necessarily image-intensive, relying heavily on the so-called “total-body CT” to identify the positions of foreign bodies in the torso, head, and neck, and plan father treatment, whether laparotomy, craniotomy, or vascular repair. Particularly in the case of a hemodynamically stable patient with trauma to the head and neck, CTA can be a critical screening test to identify vascular injuries that need to be adwww.JSurgRad.com

51


Original Article

Figure 3. Combat Support Hospitals.

O'Brien and Cox. Stents in Tents In the case of extremity vascular injuries, clinical exam and plain films are usually adequate to guide the initial evaluation and therapy. If suspicion is high for a vascular injury, a conventional catheter-based arteriogram is often most helpful. We have found that CTA quality is often not adequate for clinical decision-making in the field, and a mangled extremity with multiple embedded fragments, complex fractures and questionable perfusion is best evaluated with an ontable arteriogram (Figure 7). Tourniquet use, metallic fragments, external fixation devices, and vasospasm make reliable interpretation of CTAâ&#x20AC;&#x2122;s impossible, and it is usually safer to evaluate the patient in the operating room where hemorrhage can be immediately controlled. Fracture stabilization, arteriogram, wound debridement and arterial repair can all take place in the same setting.

Spectrum of Endovascular Interventions

Figure 4. Imaging suite near Balad, Iraq in early 2006. dressed in conjunction with neurosurgical or ENT procedures. It is simply impossible to perform an adequate diagnostic neck exploration in patients with hundreds of projectiles scattered across all three of the classically described zones of the neck (Figure 6). Patients with an equivocal CTA can proceed to a conventional arteriogram, and clear-cut injuries can be addressed with open vascular reconstruction or endovascular intervention.

52

Original Article | January 2011

The role of endovascular intervention has been in flux since the beginning of the conflict. Current military doctrine regarding casualty evacuation emphasizes early evacuation of US casualties to tertiary care centers within the continental United States, often with 48-72 hours. With this paradigm of continuous rapid transfer to the next level of care, the role of advanced imaging and endovascular procedures in-theater initially seemed limited to diagnostic arteriograms. However, the utility of more complex interventions soon became apparent since endovascular interventions offer a markedly less morbid approach to select vascular injuries. By early 2006, imaging and interventional capabilities were in place at the CSH in Balad, Iraq and catheter-based procedures

O'Brien and Cox. Stents in Tents

Original Article

Figure 5. A. Plain film demonstrating multiple missile fragments (left). B. Numerous shrapnel wounds from an Improvised Explosive Device (IED) (right).

Figure 6. Lingual artery pseudoaneurysm Figure 7. Severe lower extremity injury secondary to explosion. from multiple penetrating fragments. were being performed in isolated cases, including inferior vena cava (IVC) filter placement, arterial embolization, and stent-grafting of arterial injuries. In mid-2008, capability was robust enough that even stent-grafting of blunt aortic injury could be performed successfully.5 Placement of IVC filters was one of the first endovascular interventions performed in field hospitals. The need for pulmonary embolus (PE) prophylaxis was immediately apparent early in the conflict as severely injured patients with multiple extremity injuries and penetrating head wounds were at exceedingly high risk of DVT/PE. Enoxaparin was considered standard for DVT prophylaxis, but patients felt to be at particularly high risk, especially those with severe intracranial injury, often received a removable IVC filter prior to evacuation to the US. While the indications for IVC filter placement in combat casualties are controversial and have been in flux

throughout the conflict, the placement of IVC filters in a select group of high-risk patients was an early demonstration of our capability for endovascular intervention. Once the success of IVC filter placements was demonstrated, the next obvious application of catheter-based therapy was arterial embolization. Transcatheter control of arterial bleeding that is difficult to access surgically can be invaluable, and in many cases can be done quickly with minimal supplies. Figure 8 illustrates a patient presenting after GSW through the mouth with massive pulsatile bleeding coming from the posterior pharynx. With an unclear bleeding source in an inaccessible location at the level of the skull base, angiography was clearly indicated, and a single coil deployed in a branch of the external carotid definitively addressed the problem. Similarly, the patient illustrated in Figure 9 sustained a high-velocity GSW to the pelvis and had pelvic bleeding that could not be conwww.JSurgRad.com

53


Original Article

O'Brien and Cox. Stents in Tents

Original Article

O'Brien and Cox. Stents in Tents

Figure 8. A. Angiogram showing contrast extravasation from external carotid artery branches (top). B. Completion angiogram after coil embolization of the affected artery (bottom). trolled at laparotomy, even with aggressive packing. The bleeding point was selected and embolized with GelfoamÂŽ(Pfizer, New York, NY), and the patient promptly stabilized. Transcatheter embolization is a proven technique that has been around for decades, but stent-grafts in appropriate sizes for repair of peripheral vessels have only been available for the last few years. Their use in civilian trauma centers has been expanding exponentially, and in certain military settings the immediate benefits of a minimally invasive procedure are obvious. Achieving rapid surgical control of the vessels in the thoracic outlet, for instance, can be a challenge, and may involve rather morbid incisions. Stent-grafting of the proximal common carotid, subclavian and axillary arteries, in contrast, is relatively straightforward, with minimal short-term morbidity. In Figure 10 we see a patient who sustained a GSW to the shoulder with injury to the proximal axillary artery. Delivery of a stent-graft via the brachial artery required only a small cut-down in the antecubital fossa and quickly treated the problem with what amounted to an outpatient procedure. In cases of major vascular injury that cannot be easily or rapidly accessed with a surgical exposure, stent-grafting is rapidly becoming the procedure of choice. While we still await long-term data regarding late stent infections and long-term patency, anecdotally, we have seen no reports at tertiary-care military hospitals in the US of stent-graft infections or thrombosis with ischemia.

Recommendations Combat-related vascular injuries are still most often treated with traditional surgical approaches; however, in some situations transcatheter diagnosis and therapy can be a critical adjunct in addressing a battlefield injury. The ideal candidate for endovascular repair is the hemodynamically stable patient with a vascular injury

54

Original Article | January 2011

Figure 9. A. Angiogram demonstrating contrast extravasation involving the hypogastric artery (top left). B. Placement of a microcatheter in the affected vessel (right). C. Completion angiogram (bottom left).

that is difficult to access with a standard open approach. If appropriate imaging equipment and trained personnel are available, endovascular interventions are feasible, even in a relatively austere environment. While a fixed imaging suite may be preferable, a portable fluoroscopy unit with the capacity for digital subtraction is certainly adequate for any of the simple endovascular interventions necessary for trauma. Since a good-quality C-Arm is necessary for most orthopedic trauma cases, the additional expenditure for vascular capability is minimal. A power-injector is unnecessary, since formal aortography is rarely indicated, and simple disposable hand-operated devices can be used in the occasional instances where high-pressure contrast injection is required.

The most daunting task can be assembling the necessary disposable supplies such as catheters, wires, coils and stentgrafts. Typical endovascular stock for an interventional suite in the United States can include hundreds of different devices in multiple lengths and diameters and may require an initial outlay of several hundred thousand dollars. Fortunately, a military surgeon can ruthlessly pare down the necessary inventory, and stock only a few items that will be frequently used. In a young trauma patient with no atherosclerosis, virtually any named artery can be accessed with one or two simple catheters and a long angled GlidewireÂŽ (Terumo, Tokyo, Japan). One need not stock any bare metal stents, and stent-grafts in 6-8Â mm diameters and 4-6 cm lengths will treat the vast majority of arterial injuries encountered. Non-compliant angioplasty balloons will almost never be used, and a compliant occlusion balloon can both help with hemostasis and postdilate a stent-graft if necessary. Coil stock can be equally minimalist, and a few 3-8 mm coils can take out most bleeders. All of the truly necessary supplies can be carried in a backpack and stored in a small cabinet.

Summary The ongoing GWOT has tested military medical capability in ways not seen since Vietnam. Fortunately, vascular surgeons have new tools at our disposal which can dramatically improve outcomes for injured soldiers. While endovascular therapies currently occupy a small niche in the treatment of select wartime vascular injuries, they may be increasingly applicable as www.JSurgRad.com

55


Original Article

O'Brien and Cox. Stents in Tents

O'Brien and Cox. Stents in Tents

Original Article

longer-term efficacy data becomes available. As the latest imaging and endovascular techniques have been pushed from civilian medical centers on the battlefield, our surgeons in the field have the best chance to save lives and salvage limbs.

Editorial Commentary

doing so, as appropriate), a variety of modalities may become reasonable, such as on-table angiography and endovascular procedures, particularly embolization and stent-grafting. The ideal candidates for endovascular therapies on the battlefield, with rare exception, remain those patients with diagnostic quandaries and hemodynamic stability.

T

his is a well written article summarizing the experience of combat surgeons with endovascular diagnostics and therapies in Operation Iraqi Freedom. Clearly, this conflict has demonstrated that endovascular interventions are possible within theater; however the fundamental question that remains is selecting the appropriate patients. As the authors point out, a multiple blast-fragmentation wound to all zones of the neck is nearly impossible to explore adequately. Imaging and possible endovascular interventions are becoming the favored approach, in both the combat and civilian environments.

Disclosures The authors have no disclosures or conflicts of interest related to this manuscript.

I would caution the reader, however, to not stray too far from out “combat roots” and previous lessons learned from earlier military conflicts. The hypotensive penetrating trauma patient belongs in the operating room, not the CT scanner or in the angio suite. Once in the operating room with the ability to rapidly obtain appropriate surgical hemorrhage control (and

References

1. Schreiber MA, Zink K, Underwood S, Sullenberger L, Kelly M, Holcomb JB. A Comparison Between Patients Treated At a Combat Support Hospital in Iraq and a Level I Trauma Center in the United States. J Trauma 2008;64:S118-S122.

This paper reminds us, as combat surgeons, to continuously push to bring the best available therapies to our injuries soldiers, marines, and airmen, even when those therapies seem impossible on the battlefield. The authors demonstrate that not only are advanced technology endovascular therapies possible in a developed theater of operations, they may even be the preferred approach in appropriately selected patients.

David R. King, MD, FACS, MAJ, MC, USAR

Massachusetts General Hospital & Harvard Medical School Division of Trauma, Emergency Surgery, & Surgical Critical Care

Boston, Massachusetts

3. Starnes BW, Arthurs ZM. Endovascular Management of Vascular Trauma. Perspect Vasc Surg Endovasc Ther 2006;18:114-129. 4. Eastridge BJ, Jenkins D, Flaherty S, et al. Trauma System Development in a Theater of War: Experiences From Operation Iraqi Freedom and Operation Enduring Freedom. J Trauma 2006;61:1366-1373. 5. Propper BW, Alley JB, Gifford SM, et al. Endovascular Treatment of a Blunt Aortic Injury in Iraq: Extension of Innovative Endovascular Capabilities to the Modern Battlefield. Ann Vasc Surg 2009;23:687. e19-687.e22.

Public Domain

2. Fox CJ, Gillespie DL, O’Donnell SD, et al. Contemporary management of wartime vascular trauma. J Vasc Surg 2005;41:638-44.

Figure 10. A. Transection of the proximal axillary artery secondary to gunshot (top). B. Completion angiogram after placement of stent graft (bottom).

56

Original Article | January 2011

www.JSurgRad.com

57


ORIGINAL ARTICLE

The Evolution of Intraoperative Imaging and Neuro-Navigation in Transsphenoidal Surgery

Winder et al. Intraoperative Imaging and Neuro-Navigation

proven useful. More recently, Jane et al. described TSS using a stereotactic navigation system based upon lateral fluoroscopy, enabling a reference for intra-operative surgical instruments. This fluoroscopic frameless stereotaxy, despite its limitations, confirms its utility.14 The advent of CT and subsequent CT guided stereotaxy for cranial surgery has expanded the viable options in the management of TSS.

CT Based Frameless Stereotactic Guidance

Mark J. Winder, MD ∙ Justin Spooler, MD ∙ Marc R. Mayberg, MD  St Vincent’s Hospital, Sydney, Australia  Swedish Neuroscience Institute, Seattle, Washington, USA  Loma Linda University Medical Center, Loma Linda, California, USA

Abstract

Overview Transsphenoidal surgery (TSS) for pituitary tumors and other lesions of the anterior skull base has been in practice for over one hundred years. The advent of technological advancements in imaging combined with improvement in surgical skills has seen a marked evolution in transsphenoidal approaches. We describe the history of neuro-navigation and intraoperative imaging for TSS, from lateral radiographs and pneumoencephalograms to intra-operative MRI (iMRI) and electromagnetic navigation systems (EMS).

J Surg Rad

Keywords navigation, pituitary, surgery, intra-operative imaging, CT, MRI, stereotactic

58

Introduction

History of Neuro-Navigation

Transsphenoidal surgery (TSS) has been widely used for surgical treatment of lesions related to the sellar and parasellar region. Recent advances in technology have enabled the application of this approach to a broader scope of lesions affecting the anterior skull base. The development of TSS has benefited from the utilization of high-resolution three-dimensional images incorporated into navigation systems providing excellent detail of both bony and soft tissue anatomy, neural and vascular structures. This has enabled operative morbidity to be significantly reduced whilst improving clinical outcomes. The origins of intra-operative imaging and neuro-navigation for TSS date to the early twentieth century, but have significantly progressed in recent years due to technological advances in frameless stereotactic navigation and portable intraoperative computed tomography (iCT) and magnetic resonance imaging (iMRI).

TSS was first attempted over one hundred years ago by Herman Schoffler in 19079, 16, 18, 20, 30 and was subtly modified by von Eiselsberg and Hochenegg.20 In 1909, Kocher performed the approach with a submucosal resection of the septum. The following year, Hirsch described the endonasal trans-septal TSS, which led to Halstead’s description of the sublabial approach in 1909.9, 16, 18, 20 Cushing began utilizing the approach for pituitary tumors in 1909 performing over 300 procedures based on Schoffler’s initial description.20, 30 Unfortunately, mixed results combined with advocates for trans-cranial surgery, including Cushing, subsequently led the TSS approach to be largely abandoned until the late 1950’s. Ironically it was the pupils of Cushing, who reignited the interest in TSS. A key factor in the revitalization of TSS was the advent of intraoperative imaging and navigation using intra-operative fluoroscopy.4, 16, 20, 30 Norman Dott, who was trained by Cushing, continued practicing the TSS in Edinburgh, whilst Hirsch, having immigrated to Boston, attempted to preserve the endonasal TSS with the help of Hamlin.9, 16, 30 The persistence of Dott to maintain the TSS along with his development of a lighted speculum retractor improved the operation significantly. He introduced the notion to Guiot of France, who subsequently performed over 1000 cases using the TSS approach, along with the use of intraoperative radiofluoroscopy to help define the sellar anatomy.9, 16, 20, 30 Although the surgical approaches

Citation Winder MJ, Spooler A, Mayberg MR. The evolution of intraoperative imaging and neuro-navigation in transsphenoidal surgery. J Surg Radiol. 2011 Jan 1;2(1). Correspondence Marc Mayberg, MD E-mail marc.mayberg@swedish.org. Received August 25, 2010. Accepted September 22, 2010. Epub September 23, 2010.

Original Article | January 2011

Original Article

The development of systems for frameless stereotaxy represented a significant advance in intra-operative navigation. The initial prototype for a stereotactic frame was developed by Sir Victor Horsley in 1908, but was only used in experimental studies.7 Following Walter Dandy’s invention of ventriculography as a tool for navigation, the first human stereotactic operation, a percutaneous trigeminal rhizotomy, was performed by Kirschner in 1933. Spiegel and Wycis were the first to introduce a three dimensional stereotactic tool termed the “steFigure 1. Lateral skull fluoroscopy as utilized for trans-sphe- reoencephalatome” in 1947, using it perform a thalamotomy. noidal surgery. Note the absence of soft tissue detail and In 1949, Lars Leksell devised his own arc-centered, fixed skull right-left-midline orientation for fluoroscopic navigation. frame stereotactic device which employed polar coordinates, which later became used for radiosurgery. Over the followof the TSS were relatively well formulated, the approach was ing 15 years several other systems were developed leading to hampered by illumination and subsequent visualization. Jules the development of a widely used stereotactic frame by Todd Hardy of Montreal, having been exposed to the approach as a and Wells. This paved the way, with the advent of CT, for the fellow of Guiot, continued using fluoroscopy but also added Brown-Roberts-Wells (BRW) stereotactic system, developed in pre-operative angiography and intra-operative air encepha1979 and still used throughout the world, which allowed stelography. In 1967 Hardy first introduced the operating microreotactic navigation incorporated with CT data. scope to high resolution TSS, allowing stereoscopic visualizaThe next major progression for neuro-navigation was the tion in conjunction with intra-operative fluoroscopy.4, 9, 13, 30 move from frame-based to frameless stereotaxy. Roberts and This laid the cornerstone foundation for the development of colleagues are credited with developing the first frameless modern TSS. neuro-navigation employing a sonic digitizer, correlating tar-

Intra-Operative Fluoroscopy The lateral radiograph has been a standard in TSS since its inception by Guiot and Hardy, and for the past 40 years has been used to augment visualization of the sella turcica as well as helping outline tumor morphology in a sagittal plane (Figure 1). Unfortunately lateral fluoroscopy has several limitations; firstly, it provides only two-dimensional information (anterior-posterior and rostral-caudal) with an absence of critical information regarding midline and right-left lateral anatomical relationships. Secondly, soft tissue is poorly delineated and prior surgery or bony destruction by tumor may limit accurate identification of location. Thirdly, accuracy is dependent upon a “true” lateral projection and movement of the patient head requires repositioning of the fluoroscope. Lastly, the patient, surgeon and operating room personnel are exposed to ionizing radiation, often necessitating the use of cumbersome lead aprons throughout the case. Despite these limitations, intra-operative fluoroscopy has been able to offer some degree of operative navigation, which has

Figure 2. Fiducial registration using CT images for frameless stereotactic navigation. Adhesive fiducial markers are attached to the scalp prior to CT scan, then used as reference points for registration to the frameless navigation system. www.JSurgRad.com

59


Original Article

Winder et al. Intraoperative Imaging and Neuro-Navigation

Winder et al. Intraoperative Imaging and Neuro-Navigation

Original Article

Figure 3. Surface Matched Registration using CT images. The preoperative CT scan without fiducials is reconstructed with a 3-dimensional representation of the scalp and facial features. These are traced and the navigation system computes registration according to surface matching algorithms.

Figure 4 (facing page). A. Intra-operative CT scanner (top right). This portable device (Ceretom速; NeuroLogica, Danvers, MA.) has a gantry which can accommodate the head and upper cervical spine. B. Illustration of intraoperative CT scan (bottom right). A sterile cover is placed over the head during the image acquisition.

60

Original Article | January 2011

www.JSurgRad.com

61


Original Article

Winder et al. Intraoperative Imaging and Neuro-Navigation

Original Article

Winder et al. Intraoperative Imaging and Neuro-Navigation

get points with either CT or MRI.44 The systems subsequently progressed from stereotactic arm-based systems to frameless, armless navigation utilizing ultrasound, providing intra-operative three dimensional localization.7, 45-47 In 1993, the first report of neuro-navigation employing infrared optical measurements was presented,48 and has since evolved in both accuracy and usability. The introduction of the multislice CT scanner augmented by the evolution of navigation systems allowed three dimensional information to be accessed intra-operatively. The stereotactic navigation systems are based on pre-operative imaging with localization dependent upon patient-positioned fiducial markers, which are then registered, or calibrated with the specific reference system at the time of surgery (Figure 2). The software algorithms are able to translate three dimensional special points based on the pre-loaded imaging referenced directly from a fixed reference point, either on or near the patient. A useful option in navigation has been the development of surface matched registration, which obviates the need for fiducials or reference markers to be directly attached to the patient. In these cases the reference frame is placed separate to the patient and a navigation reference probe is used to trace the surface of the patientâ&#x20AC;&#x2122;s face and is correlated with a three dimensional model created on the pre-operative imaging (Figure 3). This method of registration has become feasible due to the three dimensional image quality that is now attainable with current radiology and the computer software systems. CT-based frameless stereotactic navigation has been routinely used in cranial and spinal neurosurgical procedures for over 25 years and has been successfully applied to TSS.26 Early experiences with frameless stereotactic systems for TSS have been described, utilizing ACUSTAR, Brainlab, Stealth Station systems and Instatrak.6, 49, 50 Elias et al reported on their experience in 37 patients undergoing TSS, noting improvements in intra-operative localization and trajectory planning with minimal additional OR time, or cost.6 In a separate series of 176 patients, Jagannathan et al. compared the use of MRI or CT based frameless stereotactic navigation to fluoroscopy in patients undergoing TSS.51 In their series, CT and MR navigation provided a significant advantage over fluoroscopy, particularly in cases of recurrent tumor where traditional landmarks could not be utilized. The benefits of CT based navigation include low cost, short acquisition time, and excellent detail of Figure 5. A. Intra-operative MRI scanner (top left). This portable device (PolestarÂŽ; Medtronics Surgical Navigation Technologies, Louisville, CO.) enables intraoperative low-field strength (0.2 Tesla) images to be acquired in the operating room, and integration of the updated MRI images into the navigation system. B. Illustration of intraoperative portable MRI in use (bottom left). Note that ferromagnetic objects such as microscope and surgical trays can be placed in relatively close proximity to the scanner. C. Typical intraoperative MRI showing pituitary tumor (facing page).

62

Original Article | January 2011

sinus and bony anatomy. Conversely, CT requires exposure to ionizing radiation, and is poor at demonstrating soft tissues and critical neurovascular structures. The use of preoperative MRI scans in conjunction with optical based navigation provides excellent detail of soft tissue and neurovascular structures, and allows for accurate intraoperative localization. Multiple reports describe the use of this modality in TSS.22, 27, 37, 51 Registration of the system may be rapidly accomplished using either implanted fiducial markers or surface registration. Shamir et al evaluated the accuracy of surface based registration at serial depths from the facial surface in 12 patients undergoing cranial surgery with intraoperative navigation. They found the average error following registration on the surface to be less than 1mm, while at increasing depths this error increased to a maximum of 4.5 mm at 150 mm from the surface.52 Navigation using MRI is inadequate for delineating bony anatomy and requires longer acquisition times with increased cost compared to CT. Optical tracking systems may also interfere with the positioning of the operating microscope or surgical instruments, particularly in transsphenoidal procedures.9, 12, 13, 20, 22

While neuro-navigation utilizing preoperative CT or MRI is useful, the concept of intra-operative scanning was considered by many to represent the gold standard, providing real time navigation and assessment of tumor resection (Figure 4a,b). Intra-operative CT scanning was first introduced by Shalit et al. in 1979, with several reports to follow in the 1980â&#x20AC;&#x2122;s.8, 26, 36, 53, 54 More recently Uhl et al reported their experience using a multislice intra-operative CT scanner including 45 patients undergoing pituitary tumor surgery. Out of the 45 patients they noted 5 with accessible residual tumor identified on the intra-operative scan.38, 42 Although there is some evidence to suggest intra-operative imaging using CT improves surgical outcome, there is no data comparing clinical results to Fluoroscopy or MRI.

Intra-Operative MRI The technological progression of MRI systems has allowed intra-operative MRI (iMRI) to augment neuro-navigation (Figure 5a,b,c) using either low field strength1, 2, 10, 15, 21, 24, 28, 33, 34, 41, www.JSurgRad.com

63


Original Article

Winder et al. Intraoperative Imaging and Neuro-Navigation

or high field strength scanners.24, 58, 59 Low field iMRI systems carry the advantages of lower cost, and the ability to use standard surgical instruments, as the surgical field is past the 5-Gauss line of the magnet (Figure 5). The patient or the scanner must be moved each time for image acquisition, which significantly increases the operative time. In a series of thirty patients with pituitary microadenomas undergoing transsphenoidal microscopic resection, Bohinski et al reported detection and subsequent resection of residual tumor in 66% of the patients using low field iMRI.23 Additionally, they detected a significant hematoma in one patient which led to conversion of the procedure to open craniotomy. In a separate series of forty-four patients undergoing transsphenoidal resection of pituitary macro-adenomas, Falbusch et al. reported that 34% of patients underwent resection for residual tumor detected on low field iMRI.24 While a significant percentage of residual tumor was detected in their series, the images were inconclu55-62

sive in 27% of patients, with 20% having false positive exams, casting some doubt on the utility of low field iMRI. High field iMRI systems provide significant advantages over their low field counterpart with regards to image acquisition and quality (Figure 6). For most sequences the acquisition time is shorter, and high field strengths allow for more advanced sequences to be performed, such as diffusion tensor and functional imaging, though this may be less useful in transsphenoidal surgery. As with low field iMRI, surgery is performed outside the magnetic field necessitating movement of the patient for imaging. Although this can be partially circumnavigated with the use of MRI compatible instruments, allowing surgery to be performed inside the magnetic field, the costs can be somewhat prohibitive. Nimsky et al., using high field strength iMRI, noted superior image quality and decreased acquisition time of some sequences compared to low field iMRI, with a 31% increase in complete resection of

Figure 6. Electromagnetic tracking system with surface matched registration. Note registration reference marker attached to forehead and magnetic field reference frame on left. In this system, the head does not require rigid fixation and the electromagnetic signal eliminates line-of-sight issues which may be present for optical tracking systems.

64

Original Article | January 2011

Winder et al. Intraoperative Imaging and Neuro-Navigation pituitary tumors.31 They also reported false positive exams for 20% patients using T1 sequences, which resulted in T2 sequences being adopted for assessment of residual tumor. Whilst iMRI is able to provide superior soft tissue detail, it is burdened by the necessity of specialized operating suites, instrumentation, and prolonged anesthesia/OR time which increases the overall cost of using this modality.10 Low field MRI leads to a diminution of image quality, with a limitation of specific sequences, which may impact the interpretation of pathology.11, 33, 61 Many authors however feel that despite these shortcomings, the ability of iMRI to detect occult residual tumor affords it a significant advantage over other forms of navigation during TSS. Visual technology through the use of microscopes, and now endoscopes for TSS, has been augmented by intra-operative navigation using iMRI. In a series of fifteen patients of endoscopic assisted TSS, Schwartz et al detected post-operative residual tumor on iMRI in three patients.34, 35 They noted lower detection of residual tumor with iMRI with the use of the endoscope, compared to historical controls where microscopic resection was performed. In a similar comparison, Theodosopoulos et al reported their results using endoscopic resection of pituitary macroadenomas in conjunction with iMRI in twenty-seven patients,62 in which three patients underwent additional resection based on detection of residual tumor. Given the paucity of studies, the small numbers in each, and the lack of data regarding clinical effect, it is difficult to make recommendations regarding the true advantages of the endoscopic assisted technology. However, it is likely that as image quality and endoscopic technology improve, matched by the push towards minimally invasive approaches, we will hopefully see more data with which to make a definitive comparison.

Original Article

EM navigation systems offer the ability of direct image fusion from previous MRI scans with either pre-operative or intraoperative CT (iCT) scans. Portable iCT, if available within the OR, may expedite workflow, minimizing additional costs.64 Magnetic tracking systems possess several essential advantages over the more traditional optical tracking neuro-navigation systems. Firstly, magnetic tracking obviates the need for rigid three-point skull fixation. Positioning the head on a horseshoe device is not only less traumatic to the patient, but it affords the surgeon the added benefit of increased comfort and better visualization by not having to rely on a fixed angle of approach throughout the case. Minor positional readjustments by changing the degrees of flexion or extension of the patientâ&#x20AC;&#x2122;s head as well as lateral tilt can be beneficial in approaching some skull base lesions. Alternatively, when optical based systems are utilized with rigid skull fixation, any accidental displacement of the head in relation to the reference fiducials may necessitate re-registration of the navigational device. In contrast, our experience with EM navigation systems for TSS where the reference marker is secured to the scalp usually with an adhesive, has been extremely effective ensuring high level accuracy in conjunction with the surgical freedom to move the patients head as required.64

Electromagnetic Navigation Systems

Secondly, magnetic based devices preclude the need for a direct line of sight. One of the problems associated with optical systems is the interruption of the infrared pathways between camera and reference fiducials or probe by the operative microscope, sterile drapes, or a variety of obstacles present in the operating room. This is especially true in TS cases due to the close proximity of the operative microscope and surgeon to both the head and the operative field. EM navigation systems solve this problem by replacing the infrared tracing device with a magnetic field generator placed under the sterile drapes and close to the patientâ&#x20AC;&#x2122;s head. Although there have been no definitive comparative trials of EM and optical based systems in TS surgery, our experience suggests the accuracy of EM systems is at least comparable.

Neuro-navigation systems offering three dimensional tracking most commonly employ either optical or electromagnetic (EM) systems. EM systems have been in development since the late 1970s with utilization in surgical clinical applications over the past ten years.63 The concept of EM navigation is based on the generation of a low energy cubital magnetic field that encompasses the head using a single transmitter coil array. A pointer that is either rigid or flexible, can be digitally defined in space through triangulation using magnetic field voluming algorithms referenced from two distally placed probe markers. A reference frame, usually applied to the scalp, is required enabling surface registration within the cubital magnetic field. Initial EM systems relied upon radiofrequency transmitters mounted to the headsets, requiring a CT with the headset in place.40 As EM systems and software algorithms evolved, headsets and their associated risk of image drift have been augmented through the utilization of matched surface registration,64 negating the requirements of CT scans with fiducial markers (Figure 6).

One of the concerns regarding EM navigation is the distortion of image accuracy due to other electromagnetic devices or ferromagnetic objects impacting the low energy cubital magnetic field. It is known that three types of sources may interfere with the normal operation of EM tracking systems: background noise from ambient electrical devices (wiring, lighting etc.), generated EM fields (electrical equipment) and ferromagnetic behavior of metallic instrumentation.65, 66 Although it is often difficult to detect the source and strength of many potential distortions (in contrast to optical systems where line of sight issues are directly apparent), inference affecting the accuracy of EM systems in a standard operating room has been shown to be minimal with most systems, maintaining millimeter accuracy.31, 32, 65, 67 Suggestions to maintain accuracy and avoid distortion include operating rooms distant to intra-operative functioning MRIs, and the avoidance of ferromagnetic metal within the EM field. In our experience, distortion by other EM devices in the room or ferromagnetic objects has not been apparent. The only alteration to our usual TS procewww.JSurgRad.com

65


Original Article

Winder et al. Intraoperative Imaging and Neuro-Navigation

dures has been the substitution to an MRI compatible (nonferromagnetic) nasal speculum, which enables tracking along the barrel of the device. A further advantage of EM tracking is the ability to combine the probe with instruments in use (e.g. curette or suction), and to accurately navigate in a non-linear trajectory by bending the probe tip to interrogate anatomical points behind other structures or out of direct line of sight view. Likely these attributes will be increasingly important as flexible endoscopic devices are developed and become more utilized.

Conclusion TSS in the modern era has been practiced for over one hundred years, undergoing gradual refinement in concordance with the development of new technologies for real-time intraoperative imaging and navigation. The evolution of neuroimaging and subsequent navigation tools in TSS has included the progression from lateral fluoroscopy and pneumatography to frameless stereotactic navigation using intra-operative optical and electromagnetic tracking systems. This evolution has improved surgical morbidity, operative success and improved post-operative recovery. It is likely that the combination of intra-operative imaging and frameless navigation will be applied to an increasing spectrum of disorders throughout the body. As these systems continue to evolve, surgeons should recognize the benefits and limitations of the varied technologies and select the appropriate neuro-navigation tool for individual cases.

8. Engle DJ, Lunsford LD. Brain tumor resection guided by intraoperative computed tomography. J Neurooncol. 1987;4(4):361-70. 9. Gandhi CD, Christiano LD, Eloy JA, Prestigiacomo CJ, Post KD. The historical evolution of transsphenoidal surgery: facilitation by technological advances. Neurosurg Focus. 2009 Sep;27(3):E8. 10. Gerlach R, du Mesnil de Rochemont R, Gasser T, et al. Feasibility of Polestar N20, an ultra-low-field intraoperative magnetic resonance imaging system in resection control of pituitary macroadenomas: lessons learned from the first 40 cases. Neurosurgery. 2008 Aug;63(2):272-84; discussion 84-5. 11. Hadani M, Spiegelman R, Feldman Z, Berkenstadt H, Ram Z. Novel, compact, intraoperative magnetic resonance imaging-guided system for conventional neurosurgical operating rooms. Neurosurgery. 2001 Apr;48(4):799-807; discussion -9. 12. Hardy J. Neuronavigation in pituitary surgery. Surg Neurol. 1999 Dec;52(6):648-9. 13. Hardy J, Wigser SM. Trans-sphenoidal surgery of pituitary fossa tumors with televised radiofluoroscopic control. J Neurosurg. 1965 Dec;23(6):612-9. 14. Jane JA, Jr., Thapar K, Alden TD, Laws ER, Jr. Fluoroscopic frameless stereotaxy for transsphenoidal surgery. Neurosurgery. 2001 Jun;48(6):1302-7; discussion 7-8. 15. Kanner AA, Vogelbaum MA, Mayberg MR, Weisenberger JP, Barnett GH. Intracranial navigation by using low-field intraoperative magnetic resonance imaging: preliminary experience. J Neurosurg. 2002 Nov;97(5):1115-24. 16. Kanter AS, Dumont AS, Asthagiri AR, Oskouian RJ, Jane JA, Jr., Laws ER, Jr. The transsphenoidal approach. A historical perspective. Neurosurg Focus. 2005 Apr 15;18(4):e6.

Winder et al. Intraoperative Imaging and Neuro-Navigation 29. Quinones-Hinojosa A, Ware ML, Sanai N, McDermott MW. Assessment of image guided accuracy in a skull model: comparison of frameless stereotaxy techniques vs. frame-based localization. J Neurooncol. 2006 Jan;76(1):65-70.

Original Article

50. Sandeman D, Moufid A. Interactive image-guided pituitary surgery. An experience of 101 procedures. Neurochirurgie. 1998 Dec;44(5):331-8.

30. Rosegay H. Cushing's legacy to transsphenoidal surgery. J Neurosurg. 1981 Apr;54(4):448-54.

51. Jagannathan J, Prevedello DM, Ayer VS, Dumont AS, Jane JA, Jr., Laws ER. Computer-assisted frameless stereotaxy in transsphenoidal surgery at a single institution: review of 176 cases. Neurosurg Focus. 2006;20(2):E9.

31. Rosenow JM, Sootsman WK. Application accuracy of an electromagnetic field-based image-guided navigation system. Stereotact Funct Neurosurg. 2007;85(2-3):75-81.

52. Shamir RR, Freiman M, Joskowicz L, Spektor S, Shoshan Y. Surfacebased facial scan registration in neuronavigation procedures: a clinical study. J Neurosurg. 2009 Dec;111(6):1201-6.

32. Sangra M, Clark S, Hayhurst C, Mallucci C. Electromagnetic-guided neuroendoscopy in the pediatric population. J Neurosurg Pediatr. 2009 Apr;3(4):325-30.

53. Shalit MN, Israeli Y, Matz S, Cohen ML. Intra-operative computerized axial tomography. Surg Neurol. 1979 May;11(5):382-4.

33. Schulder M, Sernas TJ, Carmel PW. Cranial surgery and navigation with a compact intraoperative MRI system. Acta Neurochir Suppl. 2003;85:79-86.

54. Lunsford LD, Parrish R, Albright L. Intraoperative imaging with a therapeutic computed tomographic scanner. Neurosurgery. 1984 Oct;15(4):559-61.

34. Schwartz RB, Hsu L, Wong TZ, et al. Intraoperative MR imaging guidance for intracranial neurosurgery: experience with the first 200 cases. Radiology. 1999 May;211(2):477-88.

55. Bohinski RJ, Warnick RE, Gaskill-Shipley MF, et al. Intraoperative magnetic resonance imaging to determine the extent of resection of pituitary macroadenomas during transsphenoidal microsurgery. Neurosurgery. 2001 Nov;49(5):1133-43; discussion 43-4.

35. Schwartz TH, Stieg PE, Anand VK. Endoscopic transsphenoidal pituitary surgery with intraoperative magnetic resonance imaging. Neurosurgery. 2006 Feb;58(1 Suppl):ONS44-51; discussion ONS44-51.

56. Fahlbusch R, Ganslandt O, Buchfelder M, Schott W, Nimsky C. Intraoperative magnetic resonance imaging during transsphenoidal surgery. J Neurosurg. 2001 Sep;95(3):381-90.

36. Shalit MN, Israeli Y, Matz S, Cohen ML. Experience with intraoperative CT scanning in brain tumors. Surg Neurol. 1982 May;17(5):376-82.

57. McPherson CM, Bohinski RJ, Dagnew E, Warnick RE, Tew JM. Tumor resection in a shared-resource magnetic resonance operating room: experience at the University of Cincinnati. Acta Neurochir Suppl. 2003;85:39-44.

37. Thomale UW, Stover JF, Unterberg AW. The use of neuronavigation in transnasal transsphenoidal pituitary surgery. Zentralbl Neurochir. 2005 Aug;66(3):126-32; discussion 32. 38. Uhl E, Zausinger S, Morhard D, et al. Intraoperative computed tomography with integrated navigation system in a multidisciplinary operating suite. Neurosurgery. 2009 May;64(5 Suppl 2):231-9; discussion 9-40.

58. Nimsky C, Ganslandt O, Kober H, Buchfelder M, Fahlbusch R. Intraoperative magnetic resonance imaging combined with neuronavigation: a new concept. Neurosurgery. 2001 May;48(5):1082-9; discussion 9-91.

39. Walker DG, Kaye AH. Image guidance and trans-sphenoidal surgery: past, present and future. J Clin Neurosci. 2003 May;10(3):289-92.

59. Nimsky C, Ganslandt O, Von Keller B, Romstock J, Fahlbusch R. Intraoperative high-field-strength MR imaging: implementation and experience in 200 patients. Radiology. 2004 Oct;233(1):67-78.

18. Lanzino G, Laws ER, Jr. Pioneers in the development of transsphenoidal surgery: Theodor Kocher, Oskar Hirsch, and Norman Dott. J Neurosurg. 2001 Dec;95(6):1097-103.

40. Walker DG, Ohaegbulam C, Black PM. Frameless stereotaxy as an alternative to fluoroscopy for transsphenoidal surgery: use of the InstaTrak-3000 and a novel headset. J Clin Neurosci. 2002 May;9(3):294-7.

60. Schulder M, Liang D, Carmel PW. Cranial surgery navigation aided by a compact intraoperative magnetic resonance imager. J Neurosurg. 2001 Jun;94(6):936-45.

19. Laws ER, Kanter AS, Jane JA, Jr., Dumont AS. Extended transsphenoidal approach. J Neurosurg. 2005 May;102(5):825-7; discussion 7-8. 20. Liu JK, Das K, Weiss MH, Laws ER, Jr., Couldwell WT. The history and evolution of transsphenoidal surgery. J Neurosurg. 2001 Dec;95(6):1083-96.

41. Wu JS, Shou XF, Yao CJ, et al. Transsphenoidal pituitary macroadenomas resection guided by PoleStar N20 low-field intraoperative magnetic resonance imaging: comparison with early postoperative high-field magnetic resonance imaging. Neurosurgery. 2009 Jul;65(1):63-70; discussion -1.

61. Schulder M, Salas S, Brimacombe M, et al. Cranial surgery with an expanded compact intraoperative magnetic resonance imager. Technical note. J Neurosurg. 2006 Apr;104(4):611-7.

21. Martin CH, Schwartz R, Jolesz F, Black PM. Transsphenoidal resection of pituitary adenomas in an intraoperative MRI unit. Pituitary. 1999 Aug;2(2):155-62.

42. Zausinger S, Scheder B, Uhl E, Heigl T, Morhard D, Tonn JC. Intraoperative computed tomography with integrated navigation system in spinal stabilizations. Spine (Phila Pa 1976). 2009 Dec 15;34(26):2919-26.

22. Mayberg MR, LaPresto E, Cunningham EJ. Image-guided endoscopy: description of technique and potential applications. Neurosurg Focus. 2005 Jul 15;19(1):E10.

43. Zhao Y, Yu S, Wang R, Zhao J. Clinical application of a neuronavigation system in transsphenoidal surgery of pituitary macroadenoma. Neurosurg Rev. 2006 Oct;29(4):306-11; discussion 11-2.

23. McCutcheon IE, Kitagawa RS, Demasi PF, Law BK, Friend KE. Frameless stereotactic navigation in transsphenoidal surgery: comparison with fluoroscopy. Stereotact Funct Neurosurg. 2004;82(1):43-8.

44. Roberts DW, Strohbehn JW, Hatch JF, Murray W, Kettenberger H. A frameless stereotaxic integration of computerized tomographic imaging and the operating microscope. J Neurosurg. 1986 Oct;65(4):545-9.

2. Anand VK, Schwartz TH, Hiltzik DH, Kacker A. Endoscopic transphenoidal pituitary surgery with real-time intraoperative magnetic resonance imaging. Am J Rhinol. 2006 Jul-Aug;20(4):401-5.

24. Nimsky C, Ganslandt O, Fahlbusch R. Comparing 0.2 tesla with 1.5 tesla intraoperative magnetic resonance imaging analysis of setup, workflow, and efficiency. Acad Radiol. 2005 Sep;12(9):1065-79.

3. Asthagiri AR, Laws ER, Jr., Jane JA, Jr. Image guidance in pituitary surgery. Front Horm Res. 2006;34:46-63.

25. Ohhashi G, Kamio M, Abe T, Otori N, Haruna S. Endoscopic transnasal approach to the pituitary lesions using a navigation system (InstaTrak system): technical note. Minim Invasive Neurosurg. 2002 Jun;45(2):120-3.

45. Barnett GH, Kormos DW, Steiner CP, Weisenberger J. Use of a frameless, armless stereotactic wand for brain tumor localization with twodimensional and three-dimensional neuroimaging. Neurosurgery. 1993 Oct;33(4):674-8.

Disclosures The authors have no disclosures or conflicts of interest related to this manuscript.

References

1. Ahn JY, Jung JY, Kim J, Lee KS, Kim SH. How to overcome the limitations to determine the resection margin of pituitary tumours with lowfield intra-operative MRI during trans-sphenoidal surgery: usefulness of Gadolinium-soaked cotton pledgets. Acta Neurochir (Wien). 2008 Aug;150(8):763-71; discussion 71.

4. Ciric IS, Zhao J. Transsphenoidal microsurgery: past, present and future. Expert Rev Anticancer Ther. 2006 Sep;6 Suppl 9:S75-8. 5. Di Rocco F, Oi S, Samii A, et al. Neuronavigational endoscopic endonasal sellar and parasellar surgery using a 2-mm-diameter lens rigidrod endoscope: a cadaver study. Neurosurgery. 2007 Apr;60(4 Suppl 2):394-400; discussion 6. Elias WJ, Chadduck JB, Alden TD, Laws ER, Jr. Frameless stereotaxy for transsphenoidal surgery. Neurosurgery. 1999 Aug;45(2):271-5; discussion 5-7. 7. Enchev Y. Neuronavigation: geneology, reality, and prospects. Neurosurg Focus. 2009 Sep;27(3):E11.

66

Original Article | January 2011

17. Kawamata T, Iseki H, Shibasaki T, Hori T. Endoscopic augmented reality navigation system for endonasal transsphenoidal surgery to treat pituitary tumors: technical note. Neurosurgery. 2002 Jun;50(6):1393-7.

46. Barnett GH, Kormos DW, Steiner CP, Weisenberger J. Intraoperative localization using an armless, frameless stereotactic wand. Technical note. J Neurosurg. 1993 Mar;78(3):510-4.

26. Okudera H, Takemae T, Kobayashi S. Intraoperative computed tomographic scanning during transsphenoidal surgery: technical note. Neurosurgery. 1993 Jun;32(6):1041-3.

47. Reinhardt HF, Horstmann GA, Gratzl O. Sonic stereometry in microsurgical procedures for deep-seated brain tumors and vascular malformations. Neurosurgery. 1993 Jan;32(1):51-7; discussion 7.

27. Onizuka M, Tokunaga Y, Shibayama A, Miyazaki H. Computer-assisted neurosurgical navigational system for transsphenoidal surgery-technical note. Neurol Med Chir (Tokyo). 2001 Nov;41(11):565-8; discussion 9.

48. Koivukangas J, Louhisalmi Y, Alakuijala J, Oikarinen J. Ultrasoundcontrolled neuronavigator-guided brain surgery. J Neurosurg. 1993 Jul;79(1):36-42.

28. Pergolizzi RS, Jr., Nabavi A, Schwartz RB, et al. Intra-operative MR guidance during trans-sphenoidal pituitary resection: preliminary results. J Magn Reson Imaging. 2001 Jan;13(1):136-41.

49. Burkey BB, Speyer MT, Maciunas RJ, Fitzpatrick JM, Galloway RL, Jr., Allen GS. Sublabial, transseptal, transsphenoidal approach to the pituitary region guided by the ACUSTAR I system. Otolaryngol Head Neck Surg. 1998 Feb;118(2):191-4.

62. Theodosopoulos PV, Leach J, Kerr RG, et al. Maximizing the extent of tumor resection during transsphenoidal surgery for pituitary macroadenomas: can endoscopy replace intraoperative magnetic resonance imaging? J Neurosurg. 2010 Apr;112(4):736-43. 63. Schicho K, Figl M, Donat M, et al. Stability of miniature electromagnetic tracking systems. Phys Med Biol. 2005 May 7;50(9):2089-98. 64. Eboli P, Shafa B, Mayberg M. Intraoperative computed tomography registration and electromagnetic neuronavigation for transsphenoidal pituitary surgery: accuracy and time effectiveness. J Neurosurg. 2010 Jun 18. 65. Poulin F, Amiot LP. Interference during the use of an electromagnetic tracking system under OR conditions. J Biomech. 2002 Jun;35(6):7337. 66. Birkfellner W, Watzinger F, Wanschitz F, et al. Systematic distortions in magnetic position digitizers. Med Phys. 1998 Nov;25(11):2242-8. 67. Stevens F, Conditt MA, Kulkarni N, Ismaily SK, Noble PC, Lionberger DR. Minimizing electromagnetic interference from surgical instruments on electromagnetic surgical navigation. Clin Orthop Relat Res. 2010 Aug;468(8):2244-50.

Visit www.SurgRad.com for high resolution images, interactive discussions, and a PDF version of this article. www.JSurgRad.com

67


ORIGINAL ARTICLE

Neurological Deterioration Secondary to Unrecognized Instability on CT Imaging and Upright Films in a Ligamentous Chance Fracture

Original Article

Forbes et al. Unrecognized Spinal Instability

Jonathan A. Forbes, MD ∙ Michael A. Stoker ∙ Pete E. Konrad, MD  Department of Neurosurgery, Vanderbilt University Medical Center, Nashville, Tennessee  Vanderbilt School of Medicine, Vanderbilt University Medical Center, Nashville, Tennessee

Abstract

Overview This report details the clinical course of a patient who presented following a high-energy collision with a ligamentous Chance injury at T11-T12. The patient was initially evaluated with 1 mm thin-cut CT imaging – which revealed subtle bony fractures suggestive of a mechanism of distraction. He was then evaluated with upright films in thoracolumbosacral orthosis (TLSO) – read as effectively absence of malalignment or progressive kyphosis. Because the potential for ligamentous disruption on initial CT imaging went unrecognized, MRI was not initially obtained. Following a protracted recovery complicated by intraabdominal injuries, the patient developed sudden onset of neurologic deficit following mobilization on post-injury day 11. MRI of the spine was obtained and demonstrated an unstable ligamentous Chance injury at T11-T12 with new evidence of compression of the neural elements at this level. To the authors’ knowledge, this case report is the first to present definitive evidence that upright films are unreliable in the detection of pathologic instability in ligamentous Chance injuries. The radiologic presentation of Chance injuries is described and an extremely low threshold for obtaining MRI of the spine when distraction injury is suspected is advocated. Keywords flexion-distraction, plain, CT imaging, spinal, instability

J Surg Rad

Introduction

68

Neurological deterioration secondary to unrecognized spinal instability (SI) is a significant source of potentially correctible morbidity in the trauma population. Studies indicate that approximately 0.025% of all trauma patients are affected by neurologic injury relating to unrecognized SI – including 0.21% of all patients with spine injuries.1 Three root causes have been proposed as proximal etiologies behind the failure to appropriately identify SI. In a study of 24 patients who suffered neurological deterioration secondary to missed instability by Levi et al., insufficient imaging was the root cause in 14 (58%) patients, misread imaging in 8 (33%) patients, and poor quality imaging in 2 (8%) patients.1

Citation Forbes JA, Stoker MA, Konrad PE. Neurological deterioration secondary to unrecognized instability on CT imaging and upright films in a ligamentous Chance fracture. J Surg Radiol. 2011 Jan 1;2(1). Correspondence Jonathan Forbes, MD E-mail jonathan.forbes@vanderbilt.edu. Received August 31, 2010. Accepted September 23, 2010. Epub September 25, 2010.

Original Article | January 2011

Patients who arrive to medical centers following high-energy collisions receive a CT traumagram – which includes axial CT imaging of the cervical, thoracic, and lumbar spine with image spacing of 5 mm or less.2 At the majority of level I trauma centers, this imaging is comprised of thin-cut axial imaging of 2.5 mm or less and includes coronal and sagittal reconstructions. MRI is obtained on a selective basis in patients who are felt to be at increased risk of SI given the appearance of the injury on CT imaging. The practice of obtaining upright plain films has been advocated in all patients with spinal injuries who are treated non-operatively, as deformity sometimes progresses to unacceptable levels with the physiological loads associated with an upright posture.3 In the past, traumatic injuries of the spine have been subdivided using the AO Spine Classification System based on underlying mechanism into compressive, flexion-distraction, and multidirectional fraction-dislocation injuries.4 At our institution, MRI is not routinely obtained in patients who present with injuries felt to be secondary to a mechanism of compression (e.g., anterior compression, superior endplate, burst fractures, etc.). In these patients, upright films are obtained in individuals whose injuries are felt to be amenable to non-operative therapy (e.g.,

Figure 1. A. Sagittal CT image that includes subaxial cervical, thoracic, and lumbar spine. Thin arrow indicates anterior superior endplate fracture of T12. B-D. Axial cuts which include the T11, T12, and L1 vertebral bodies. In (B) the T10-T11 facet junction is visible, with no evidence of facet widening. In (C), diastasis of the right inferior articular process of T11 from the right superior articular process of T12 is illustrated by the thick arrow. E-H. Magnified consecutive sagittal images of the thoracolumbar junction are seen, proceeding from right to left. Fracture of the anterior superior endplate of T12 (indicative of avulsion of the annulus from the body secondary to distraction mechanism) is visible. In (E) diastasis of the right inferior articular process of T11 from the right superior articular process of T12 – illustrated by the arrowhead – suggests possible instability at this level. In contrast, as visualized in (H), no widening of the junction of the left inferior articular process of T11 with the left superior articular process of T12 is present. bracing) following initial CT imaging. If the patient exhibits evidence of significant progressive kyphosis when upright, the posterior ligamentous complex is assumed to be compromised and surgical reduction and stabilization is indicated. If the patient fails to exhibit evidence of progressive kyphosis when upright, the posterior ligamentous complex is assumed to be physiologically viable and conservative management is recommended. In contrast to the diagnostic work-up for compressive injuries, patients with flexion-distraction injuries are often recommended for MRI of the spine soon after arrival. This recommendation relates to an association of these injuries with unstable ligamentous disruption – occasionally in the setting of subtle findings on CT imaging.1 While the radiologic features of ligamentous Chance injuries on CT imaging can be indistinct, the diagnosis of multidirectional fracture/dislocation injuries with associated instability is usually readily apparent.5 Consequently, many patients with fracturedislocations injuries proceed to the OR for stabilization following initial CT imaging. There is a paucity of information in the literature regarding the use of upright films in lieu of MRI in the detection of spinal instability. To the authors’ knowledge, this case report is the first of its kind to present evidence that upright films are unreliable in the detection of pathologic instability in liga-

mentous Chance injuries – where compromise of the bony anterior column is often minimal. The radiologic presentation of Chance fractures is described and an extremely low threshold for obtaining MRI of the spine when distraction injury is suspected is advocated.

Case Report History and Examination A 27 year-old male presented to the trauma center following an unrestrained high-speed motor vehicle collision notable for ejection. The patient was intubated on the scene and noted to be hemodynamically unstable en route to the hospital. His blood pressure eventually stabilized with fluids and vasopressors. Following arrival, he was evaluated and treated for multiple injuries visualized on CT traumagram, including complex orthopedic injuries of the left femur and right acetabulum in addition to various intraabdominal injuries. Immediately following initial evaluation and imaging, he was taken urgently to the operating room for exploratory laparotomy and hepatorrhapy, splenectomy, and right nephrectomy. A neurosurgery consultation was placed for spinal injuries noted on thin-cut www.JSurgRad.com

69


Original Article

Forbes et al. Unrecognized Spinal Instability

Original Article

Forbes et al. Unrecognized Spinal Instability (1 mm) CT imaging with reconstructions that included multiple transverse process and spinous process fractures of T10 through L4 in addition to extension of transverse process fractures to involve the pedicle at T10 and T11 (Figure 1). Fractures of the inferior-posterior endplate of T11 and the anterior superior endplate of T12 were also noted. Initial evaluation in the trauma ICU following surgery was limited secondary to the heavy sedation and paralytics required for his open abdomen. The patient returned to the OR for removal of packing and delayed primary closure of fascia approximately 6 hours following his initial procedure. On examination the following day, he was noted to be following commands weakly. Soft tissue swelling around the posterior thoracolumbar elements was noted on exam, with some apparent tenderness to palpation. Upright plain films of the thoracic and lumbar spine obtained on post-injury day (PID) 1 in a thoracolumbosacral orthosis (TLSO) failed to reveal any evidence of progressive kyphosis or malalignment (Figure 2) and the patient was cleared to mobilize as tolerated with brace when greater than 30 degrees. On PID3, the patient was taken to the OR for treatment of his orthopedic injuries. He continued to convalesce and was mobilized for the first time with physical therapy on PID9. This activity was associated with extreme pain in his back. On PID11, the patient developed new lower extremity weakness (ASIA C), again while being mobilized with the physical therapy. Neurosurgery was reconsulted emergently and MRI of the spine demonstrated a distractive injury with new instability and cord impingement at T11-12 consistent with a ligamentous chance injury (Figure 3).

Operation and Post-Operative Course The patient was taken urgently to the operating room for a T11-T12 laminectomy with T10-L1 posterior spinal fusion. Post-operative films revealed good realignment (Figure 4). The patientâ&#x20AC;&#x2122;s bilateral paresis of the lower extremities improved over the ensuing weeks. He was discharged to rehab on PID16/ POD5. At follow-up reevaluation at 6 weeks his lower extremity weakness was noted to have resolved entirely.

Discussion Flexion-distraction injuries were virtually unknown prior to routine use of the lap seat belt during motor vehicle collisions and were first formally described in 1948 by G.Q. Chance.6 Since this time, these injuries have alternatively been referred to as Chance and/or seat-belt type injuries. Flexion-distraction injuries result from a primary force vector acting along an axis of rotation anterior in location to the middle column.7 In Denisâ&#x20AC;&#x2122; sentinel paper in 1983, he described four subtypes of spinal column injury based on fracture morphology: compression fractures, burst fractures, seat-belt type fractures, and fracture-dislocations.8 Each fracture could be described in terms of associated damage to the 3 columns (anterior, middle, posterior) of the spine. Seat-belt type fractures, also known as flexion-distraction injuries, were observed to result after failure of the middle and posterior columns in distraction.8 In these injuries, the anterior column may fail in tension or compression depending on whether the instantaneous axis of rotation is anterior to the anterior spinal column.7

Figure 2. Upright AP and lateral plain films of the thoracic and lumbar spine obtained on post-injury day (PID) 1 in a thoracolumbosacral orthosis (TLSO) failed to reveal evidence of progressive kyphosis or malalignment.

Figure 3. MRI of the spine demonstrates new evidence of instability and cord impingement at T11-12 consistent with a ligamentous chance injury.

70

Original Article | January 2011

Figure 4. Post-operative upright films in brace after T11-T12 laminectomy with T10-L1 posterior spinal fusion. www.JSurgRad.com

71


Original Article

Forbes et al. Unrecognized Spinal Instability

Forbes et al. Unrecognized Spinal Instability

Original Article

Figure 6. Radiologic films of another patient who presented following a MVC with a L1 burst fracture. Sagittal (A) and axial (B) CT images of the fracture are visible. The thin arrow demonstrates the level of the L1 burst fracture in (A). Upright AP (C) and lateral (D) films demonstrated progressive kyphosis at L1 (arrow) indicative of disruption of the PLC. The patient required posterolateral fusion for stabilization. The PLC disruption was verified intra-operatively. missed diagnoses and secondary neurologic injury have been reported.1

Figure 5. The three injuries types of the AO Spine Classification are demonstrated. (A) Type A involves compression injury. (B) Type B involves distraction injury. (C) Type C involves multidirection injury with shear and/or translation. Images courtesy of M. Aebi, V. Arlet, J.K. Webb, AO-Manual of Spine Surgery, Vol. I, 2008. Thieme Publisher, Stuttgart.11 In Magerl’s (e.g., AO Spine) comprehensive injury classification system published in 1994 (Figure 5), spinal column injuries were differentiated based on mechanism.4 Type A injuries were related to compression, Type B injuries were related to distraction, and Type C injuries were related to multidirectional injury, often with translation. Based on individual characteristics, fractures could be divided into further subcategories. Regarding injuries suffered secondary to a mechanism of distraction, B1 injuries primarily involve underlying ligamentous damage while B2 injuries involve damage to the bony elements. B3 injuries involve predominant distraction anteriorly. While these respective subcategories can be divided into additional subclassifications, further analysis is beyond the scope of this report. The patient in this case would be described as having a seat-belt type fracture using the Denis classification, and an AO Spine B1 (predominantly ligamentous flexion-distraction injury) using the Magerl classification. In patients who present to trauma centers across the U.S. following high-energy collisions, the current gold standard for detection of injuries to the spinal column involves thin-cut CT axial imaging with coronal and sagittal reconstructions. The sensitivity of this imaging modality regarding the detection of bony injury approaches 100%.9 Even more important than the diagnosis of bony injury, however, is detection of pathologic instability. The potential to miss SI can be analyzed in regards to Denis’ classification of spinal column injuries and largely relates to the potential of the injury complex in question to result in occult ligamentous disruption. Compression injuries

72

Original Article | January 2011

are readily detected on CT imaging and are almost universally stable.4 Burst fractures similarly are routinely identified on CT imaging. Focal kyphosis > 30 degrees and/or loss of height >50% are suggestive of compromise of the posterior ligamentous complex (PLC) and instability in these fractures.10 When it is unclear on CT imaging as to whether true instability is present, upright films in unstable burst fractures often reveal progressive kyphosis under physiological loads (the films of another patient who presented following an MVC with a burst fracture with instability found on upright films can bee seen in Figure 6). At our institution, upright films are obtained for burst fractures felt to be amenable to non-operative therapy with bracing, based on findings on CT imaging. MRI in these patients is not routinely accomplished, as patients who demonstrate progressive kyphosis when upright require operative reduction and stabilization and patients who do not are considered to be stable and treated with bracing. It is notable that the routine use of MRI scans in all burst fractures to determine PLC integrity has been advocated by some authors.12 Other authors have questioned the clinical applicability of findings on MRI and have recommended against the routine use of MRI in patients with burst fractures because of a noted inability to influence management.13 Fracture-dislocation injuries often involve significant damage to the bony elements that is readily evident on CT imaging. These injuries are associated with instability and the highest incidence of neurologic deficit.4 While fracture-dislocation injuries are often readily detected following standard CT traumagram, cases of

Flexion-distraction injuries are a unique group of spinal column injuries associated with significant instability and, not infrequently, subtle radiologic findings on CT imaging.14 The indistinct findings on initial plain film and CT imaging occasionally lead to misdiagnosis and undetected spinal instability (refer to Table 1 for a list of all previously published case reports of delayed diagnosis of unstable, ligamentous [AO Spine

B1] Chance-type injuries.). Flexion-distraction injuries have been noted to be associated with a lower incidence of neurologic deficit on presentation than fracture-dislocation injuries4 and are relatively infrequent – representing 5 to 15% of all thoracolumbar fractures.8, 15, 16 Past descriptions have further classified flexion-distraction injuries into bony or ligamentous subtypes. Intuitively speaking, bony flexion-distraction injuries would be expected to be detected on CT more readily than those of the ligamentous variety. While this may be true,

Figure 7. Inferior displacement of T12 spinous process results in a degree of interspinous widening and “empty vertebral body sign” on AP radiograph (arrow). www.JSurgRad.com

73


Original Article

Forbes et al. Unrecognized Spinal Instability

Forbes et al. Unrecognized Spinal Instability

References

Table 1. Delayed diagnosis of unstable predominantly ligamentous (AO Spine B1) Chance-type injuries. Author and Year

Reason Injury Was Missed

Level of Injury

Retrospective Associated Radiologic Findings

Neurologic Sequelae

Months Between Injury and Diagnosis

Burdi et al. 2008

X-rays misread

T12-L1

Perched facets, widening of interspinous distance

Good result

<1

Ceroni et al. 2004

Delayed MRI

L1-L2

Widening of interspinous distance

Good result

<1

Keene et al. 1988

Insufficient imaging studies

L1-L2

Increased distance between L1-L2

Good result

7

Keene et al. 1988

X-rays misread

L2

Delayed widening of interspinous distance

Radiculopathy, hypotonic bladder

36

While ligamentous Chance injuries are often diagnosed on CT imaging, the high prevalence of subtle findings leads to the potential for misdiagnosis. In contrast, when MRI is obtained in these patients, the radiologic findings are unambiguous and allow for definitive assessment of discoligamentous instability. It is also worth noting that dynamic flexion and extension films have been reported to be accurate in delineating instability in this setting.18 Cases such as this one that involve neurological deterioration secondary to unrecognized spinal instability are useful to review so that future morbidity

74

Original Article | January 2011

14. Bernstein MP, Mirvis SE, Shanmuganathan K: Chance-type fractures of the thoracolumbar spine: imaging analysis in 53 patients. AJR Am J Roentgenol 187(4):859-68, 2006.

2. Mancini DJ, Burchard KW, Pekala JS: Optimal thoracic and lumbar spine imaging for trauma: are thoracic and lumbar spine reformats always indicated? J Trauma 69(1):119-21, 2010.

15. Gertzbein SD, Court-Brown CM: Flexion-distraction injuries of the lumbar spine: mechanisms of injury and classification. Clin Orthop Relat Res 227:52–60, 1988.

3. Benzel EC: Spine surgery: techniques, complication, avoidance, and management, ed 2. Philadelphia, PA, Elsevier Churchill Livingstone, 2005.

16. Gumley G, Taylor TKF, Ryan MD: Distraction fractures of the lumbar spine. J Bone Joint Surg Br 64B:520–525, 1982.

4. Magerl F, Aebi M, Gertzbein SD, Harms J, Nazarian S: A comprehensive classification of thoracic and lumbar injuries. Eur Spine J 3(4):184-201,1994.

can be avoided. Regarding the radiologic features previously defined, the patient described in this case report did not exhibit the traditional definition of the dissolving pedicle sign. However, the small degree of distraction that was present did result in a slightly asymmetric presentation of the pedicles on axial imaging at the T11-T12 level (refer to Figure 1c). Furthermore, while a formal “naked facet sign” was not present, axial imaging and sagittal reconstructions did demonstrate some diastasis of the right T11-T12 facet joint (refer to Figures 1c, 1e, 1f). Despite formal absence of the two previously described radiologic features, the “empty vertebral body sign” was present in the patient described (see Figure 7). This finding was felt to be secondary to the mild degree of splaying of the spinous process at the T11-T12 level in conjunction with the fractured T12 spinous process. One additional noteworthy radiologic finding in the patient described was the fracture of the anterior-superior endplate of T12 – likely consistent with distraction of the annulus at this level with avulsion of the aforementioned region of endplate. The bony findings in this patient, although subtle, were strongly indicative of a distractive injury. Knowledge that this patient had suffered a severe intraabdominal injury should have raised the index of suspicion for a flexion-distraction injury. In patients who present with radiologic findings suggestive of distractive injury in the thoracolumbar spine, MRI is recommended to fully ascertain the degree of discoligamentous disruption and definitively evaluate for spinal instability.

13. Dai LY, Ding WG, Wang XY, Jiang LS, Jiang SD, Xu HZ: Assessment of ligamentous injury in patients with thoracolumbar burst fractures using MRI. J Trauma 66(6):1610-5, 2009.

5. Liljenqvist U, Halm H, Castro WH, Mommsen U: Thoracic fracturedislocations without spinal cord injury: a case report and literature review. Eur Spine J 4(4):252-6, 1995. 6. Chance GQ: Note on a type of flexion fracture of the spine. Br J Radiol 21:452–453,1948. 7. Patel AA, Vaccaro AR: Thoracolumbar spine trauma classification. J Am Acad Orthop Surg 18(2):63-71, 2010. 8. Denis F: The three column spine and its significance in the classification of acute thoracolumbar spinal injuries. Spine 8(8):817-31, 1983. 9. Roos JE, Hilfiker P, Platz A, et al: MDCT in emergency radiology: is standardized chest or abdominal protocol sufficient for evaluation of thoracic and lumbar spine trauma. AJR Am J Roentgenol. 183: 959– 968, 2004.

17. Groves CJ, Cassar-Pullicino VN, Tins BJ, Tyrrell PN, McCall IW: Chance-type flexion-distraction injuries in the thoracolumbar spine: MR imaging characteristics. Radiology 236(2):601-8, 2005. 18. Ceroni D, Mousny M, Lironi A, Kaelin A: Pediatric seatbelt injuries: unusual Chance's fracture associated with intra-abdominal lesions in a child. Eur Spine J 13(2):167-71, 2004. 19. Burdi M, Bono CM, Kauffman CP, Hoyt D, Garfin SR: Delayed diagnosis of a flexion-distraction (seat belt) injury in a patient with multiple abdominal injuries: a case report. Am J Orthop 37(1):44-6, 2008. 20. Keene JS, Lash EG, Kling TF Jr: Undetected posttraumatic instability of "stable" thoracolumbar fractures. J Orthop Trauma 2(3):202-11, 1988. 21. Smith MW, Reed JD, Facco R, Hlaing T, McGee A, Hicks BM, Aaland M: The reliability of nonreconstructed computerized tomographic scans of the abdomen and pelvis in detecting thoracolumbar spine injuries in blunt trauma patients with altered mental status. J Bone Joint Surg Am 91(10):2342-9, 2009.

10. Heary RF, Kumar S: Decision-making in burst fractures of the thoracolumbar and lumbar spine. Indian J Orthop 41:268-76, 2007. 11. Aebi M, Arlet V, WebbJK: AO-Manual of Spine Surgery, Vol. I, 2008. Thieme Publisher, Stuttgart. 12. Petersilge CA, Pathria MN, Emery SE, Masaryk TJ: Thoracolumbar burst fractures: evaluation with MR imaging. Radiology 194:49-54, 1995.

Visit www.SurgRad.com for high resolution images, interactive discussions, and a PDF version of this article.

Conclusion Upright films are unreliable in the detection of spinal instability in Chance fractures. When CT imaging reveals evidence suggestive of a mechanism of distraction in the thoracolumbar spine, a low threshold for obtaining MRI of the thoracic and lumbar spine is advocated. If MRI is contraindicated, the use of dynamic (flexion and extension) plain films to diagnose instability in this setting has been described.

Disclosure The authors report no conflicts of interest concerning the materials or methods used in this study or the findings specified in this paper.

Photos.com

the vast majority of ligamentous flexion-distraction injuries appear to be associated with some evidence of bony injury. This is typified by a case series of 53 patients with flexiondistraction injuries by Bernstein et al., where no patient was found to have a purely ligamentous injury without evidence of bony involvement.14 Another retrospective review by Groves et al. found that only 1 of 24 patients had a purely ligamentous injury.17 The pattern of bony involvement with predominantly ligamentous (B1) flexion-distraction injuries is important, as these features may represent the only discernible evidence of underlying injury and/or instability on initial CT imaging. Location of these fractures is the first such characteristic that deserves mention. In the retrospective review by Bernstein and colleagues, 78% of flexion-distraction injuries occurred between T12 and L2.14 Associated injuries are also relevant – 40% of patients who presented with this injury complex in the aforementioned study had associated intraabdominal injury. A handful of notable radiologic findings have been described in patients who present with this injury complex. The “dissolving pedicle sign” has been described in an estimated 76% of flexion-distraction injuries.14 This feature has been defined as a gradual loss of definition of the pedicles at the level of the fracture and is particularly suggestive of a bony (AO Spine B2) Chance-type fracture. Another feature known as the “naked facet sign” – representing uncovering of the articular facets secondary to distraction injury – was seen in 40% of patients in the case series by Bernstein et al.14 One additional radiologic finding – known as the “empty vertebral body sign” – was found 100%, or 53 of 53 patients of the aforementioned series. The specificity of this latter finding is unknown. In the review of 24 MRI scans by Groves et al., a loss of anterior vertebral body height secondary to compression or corner fracture was described in 83% of patients with flexion-distraction injuries and facet disruption was seen in 71% of patients.17

1. Levi AD, Hurlbert RJ, Anderson P, Fehlings M, Rampersaud R, Massicotte EM, France JC, Le Huec JC, Hedlund R, Arnold P: Neurologic deterioration secondary to unrecognized spinal instability following trauma--a multicenter study. Spine 31(4):451-8, 2006.

Original Article

www.JSurgRad.com

75


ORIGINAL ARTICLE

Screening for Occipitoatlantal Dislocation in the Acute Trauma Setting Erin Moody, MD, MBA ∙ Alicia Mangram, MD, FACS ∙ David Ostrow, MD ∙ Robert Nisbet, MD ∙ Manual Lorenzo, MD, MBA ∙ Ernest Dunn, MD Departments of General Surgery, Neurosurgery, and Radiology, Methodist Health Systems, Dallas, Texas

Abstract

Introduction Traumatic occipitoatlantal dislocation (OAD) is infrequent, but, with improved acute care, more commonly seen. Earlier diagnosis equals better outcome, so trauma surgeons, as first-line caregivers, must be adept at early identification. Screening for OAD should be done with cervical spine computed tomography (CT-C spine), the current standard imaging modality for acute traumatic injury. This paper delineates an easy technique classically applied to radiographs, the basion-dental-interval (BDI), for evaluating CT-C spines for OAD. Methods Retrospective chart review performed from 1/1998-5/2009. 4 initial survivors identified and charts reviewed for time-to-diagnosis, method of diagnosis, outcome, and mechanism of injury. Initial CT-C spine images reviewed for diagnostic utility using the BDI screening method. Results 75% (3/4) motor vehicle accidents and 25% (1/4) motorcycle accidents. Injury Severity Score (ISS) was 30-45. 100% (4/4) received CT-C spines and could be diagnosed using the BDI method. Time to diagnosis was within 24 hours. 50% (2/4) had operative intervention and were eventually discharged. Discussion As trauma centers see more viable patients with OAD, trauma surgeons need to quickly recognize it to assure appropriate care. The BDI technique is quick and reliable using CT-C spine as the initial screening film in blunt trauma patients with significant neurological injuries. Keywords basion-dental-interval, spine trauma, cervical spine

J Surg Rad

Introduction

76

Occipitoatlantal dislocation (OAD) is an injury that is typically incompatible with life. However, in the literature within recent years, there have been several case reports describing increasing numbers of survivors with the injury.1,2,3 This is largely attributed to advanced training in first responders, prompt immobilization and airway management, increased access to tertiary care centers, and implementation of standardized protocols.3,4,5,6 At our institution, we have seen an increase in initial survivors from one in 2005 to three in 2009, with two of these patients surviving to surgical intervention and discharge. We defined initial survivor as any patient that lives beyond the first 24 hours after hospital admission.

Citation Moody E, Mangram A, Ostrow D, Nisbet R, Lorenzo M, Dunn E. Screening for occipitoatlantal dislocation in the acute trauma setting. J Surg Radiol. 2011 Jan 1;2(1). Correspondence Erin Moody, MD E-mail emoody5@gmail.com. Received August 15, 2010. Accepted September 10, 2010. Epub September 16, 2010.

Original Article | January 2011

OAD can be further categorized into three general types. Type I dislocations are anteriorly displaced, and type II dislocations are vertical distraction injuries. Type II dislocations are sometimes subdivided into type IIa with vertical displacement between the occiput and atlas and type IIb with vertical displacement between the atlas and axis, but, to simplify the screening process, we will not make this distinction here. Type III dislocations describe a posterior displacement between the occipital condyles and atlas.7,8,9

Materials and Methods Retrospective review from Jan 1998-May 2009 was performed for patients diagnosed with OAD. Medical data was analyzed for time to diagnosis, method of diagnosis, outcome, and mechanism of injury, ISS, and associated injuries. We examined all pertinent images to ascertain reproducibility of OAD diagnosis using the BDI screening method to confirm that it is a useful diagnostic tool. The BDI was measured on initial CT’s of the cervical spine for patients with known OAD with 8.5 mm as the upper limit of nor-

Moody et al. Occipitoatlantal Dislocation in Trauma

Original Article

soft tissue swelling, which is an imaging finding associated with OAD.1,14,15 The patient was transferred to the surgical intensive care unit (SICU), where he expired on hospital day two secondary to his multiple injuries. The three remaining initial survivors all presented in 2009, were status post motor vehicle accident, and had associated neurological injuries. One patient received an initial screening CT of the cervical spine, on which OAD was not diagnosed until the neurosurgeon on call reviewed the film. This patient had a BDI of 15.6 mm. Follow-up magnetic resonance imaging (MRI) of the cervical spine confirmed OAD with injury to the cruciform ligament and ligamentum flavum, as well as swelling and hematoma in the prevertebral soft tissues. This patient was admitted to the SICU and expired Figure 1. A. Normal basion-dens alignment (top left). B. Anterior displacement (top right). C. on hospital day 7 secondary to complications. No Distraction/ vertical displacement (bottom left). D. Posterior displacement (bottom right). operative intervention was attempted due to perceived mal. Measurements were made from the most inferior portion poor prognosis. of the basion to the closest point of the superior aspect of the The two remaining initial survivors underwent operative dens in the midsagittal plane.12,13 All films were reviewed with stabilization and were discharged. In one patient, an initial a neuroradiologist and/or neurosurgeon. screening lateral cervical spine X-ray was done which failed to show the OAD or related suggestive findings, such as prevertebral soft tissue swelling. Again, the neurosurgeon on call made the diagnosis of OAD on follow-up CT of the cervical spine with a BDI of 10.5 mm. Associated findings included a Four initial survivors with OAD were identified. All patients fractured right occipital condyle. This patient underwent opwere subject to blunt force trauma, 75% (3/4) motor vehicle erative stabilization on hospital day 16 and was eventually disaccidents and 25% (1/4) motorcycle accidents. ISS for all pacharged to a rehabilitation facility on day 67. The final patient tients was 30-45, and all patients received screening CT of the received an initial screening CT of the cervical spine. Diagnocervical spine. sis was, again, made by the neurosurgeon on call with a BDI of All four initial survivors were diagnosed within twenty-four 14.8 mm. Follow-up MRI of the cervical spine confirmed OAD hours on CT of the cervical spine by either the neurosurgeon and suggested transverse ligamentous injury with edema in or radiologist on call. On review, this diagnosis was possible the bilateral occipitoatlantal joints. This patient underwent using the BDI screening method. Diagnosis was confirmed via operative stabilization on hospital day 15 and was discharged direct visualization during operative intervention or autopsy. home on day 59. Morbidity was high for both patients secondOAD was suspected in the first initial survivor based on plain ary to related neurological injuries. film of the cervical spine, which was confirmed on CT of the cervical spine by the radiologist on call with a BDI of 23.8 mm. Significantly, this CT-C spine also showed prevertebral

Results

www.JSurgRad.com

77


Original Article

Moody et al. Occipitoatlantal Dislocation in Trauma

at reading screening CT’s of the chest, abdomen, and pelvis, and head, and cervical spine for gross traumatic abnormalities and fractures. OAD, though, is sufficiently rare and has historically been associated with such a poor prognosis that most trauma surgeons have not routinely screened for it. However, as patient outcomes continue to improve, the utility of screening for this condition will increase. Trauma surgeons need to incorporate the BDI screening method into their repertoire to stay current with the accepted standard of care. Patients subjected to blunt force trauma with ISS > 30, and especially those with other neurologic injuries, should be screened for OAD using the BDI method. In addition, surgeons should routinely evaluate the occipitoatlantal joint on sagittal CT-C spine images for an obvious dislocation/dissociation. Other findings associated with OAD, which can help make the diagnosis, include paravertebral hematomas, prevertebral soft tissue swelling, joint incongruity, vertebral artery injury, capsular swelling, subarachnoid hemorrhage at the craniovertebral junction, and, rarely, fractures through cranial nerve canals.1,12,17,19 In our experience, screening lateral cervical spine radiographs provided no benefit over CT of the cervical spine as CT is fast, readily available, and a far more sensitive study (52% versus 98%, respectively for all cervical spine injuries).18,19 In addition, CT of the cervical spine more readily depicts concurrent paravertebral hematomas, prevertebral soft tissue swelling, joint incongruity, vertebral artery injury, capsular swelling, and subarachnoid hemorrhage at the craniovertebral junction, if present, which can be helpful in making the diagnosis. As mentioned above, surgeons should routinely evaluate the occipitoatlantal joint in all patients with suspected injuries of the cervical spine on sagittal CT-C spine images for an obvious dislocation/dissociation, as well as utilizing the BDI method. Recently, it was shown that 8.5 mm is the upper limits of a normal BDI on CT of the cervical spine, whereas it was commonly accepted to be 12 mm on lateral cervical spine X rays.12,16 The BDI method is one of the most sensitive described in the literature, with a distance of 8.5 mm found in >95% of the population.9,10,15 MRI of the cervical spine as of yet has little utility in the immediate screening process as it has a far slower acquisition time relative to CT and is not as widely available. However, MRI can be useful for a more specific diagnosis and description of OAD.19 Figure 1. A. Normal CT of the cervical spine with BDI of 7.7 mm, designated by the white lines, measured the most inferior portion of the basion to the closest point of the superior aspect of the dens in the midsagittal plane (top left). B. An example of a grossly abnormal BDI (23.8 mm). OAD was later confirmed on autopsy; type I injury (top right). C. Type II injury (BDI 15.6 mm); OAD was confirmed during surgery (bottom left). D. A subtly abnormal BDI at 10.5 mm (type II injury) demonstrates the importance of this screening method as diagnosis could be missed on gross initial exam. Diagnosis was confirmed during surgery (bottom right).

Discussion Recent advances in first responder care, better equipment, increased access to level 1 trauma centers, implementation of standardized protocols, and advances in acute care in the hospital setting have resulted in an increasing number of initial survivors with OAD, a condition that was previously associ-

78

Original Article | January 2011

Original Article

Moody et al. Occipitoatlantal Dislocation in Trauma

Disclosures The authors have no disclosures or conflicts of interest related to this manuscript.

References

1. Bloom BM, Powell BP. Surviving atlanto-occipital dislocation.EMA 2007: 19: 379-382. 2. Hamai S, et al. Traumatic Atlanto-Occipital Dislocation With Atlantoaxial Subluxation. Spine 2006: 31; 13: E421-E424. 3. Schmal H, et al. Traumatic atlanto-occipital dislocation as part of a complex cervical spine injury. Case report in a 12 year old girl. Unfallchirurg 2007: 110; 8: 720-725. 4. Babbitz JD, Kim KD. Imaging Corner: Unknown Case. Spine 2001: 26: 12: pp 1401-1403. 5. Bundschuh CV, Alley JB, Ross M, et al. Magnetic Resonance Imaging of suspected atlanto-occipital dislocation: two case reports. Spine 1992: 17: 245-8. 6. Henry MB, Angelastro DB, Gillen JP. Unrecognized traumatic atlantooccipital dislocation. Am J Emeg Med 1998: 16: 406-8. 7. Bono CM, Carreras E. “Occipitoatlantal Injuries.” Decision Making in Spinal Care. Thieme 2007: 3-9. 8. Arakal RG, Vives M. “Cervical Trauma.” Spine. Lippincott Williams & Wilkins 2004: 35-44. 9. Hadley NM, Walters BC, Grabb PA, et al. Diagnosis and Management of Traumatic Atlanto-occipital Dislocation Injuries. Neurosurgery 2002; 50;3: S105-S113. 10. Saeheng S, Phuenpathom N. Traumatic Occipitoatlantal Dislocation. Surg Neurol 2001; 55: 35-40. 11. Bellabarba C, et al. Diagnosis and Treatment of craniocervical dislocation in a series of 17 consecutive survivors during an 8-year period. J Neurosurg Spine 2006: 4: 429-440. 12. Rojas CA, et al. Reassessment of the Craniocervical Junction: Normal Values on CT. Spine 2007:28: 1819-1823. 13. Gonzalez LF, et al. Vertical atlantoaxial distraction injuries: radiological criteria and clinical implications. J Neurosurg Spine 2004; 1: 273280. 14. Wholly MH, et al. The lateral roentgenogram of the neck; with comments on the atlanto-odontoid-basion relationship. Radiology 1958; 71: 350-356. 15. Jackson, RS, et al. Upper Cervical Spine Injuries. J Am Acad Orthop Surg 2002: 10: 271-280. 16. Harris JH Jr, Carson GC, Wagner LK, Kerr N: Radiologic diagnosis of traumatic occipitovertebral dissociation: Part 2- Comparison of three methods of detecting occipitovertebral relationships on lateral radiographs of supine subjects. Am J Radiol 1994; 162: 887-892. 17. Deliganis, AV, et al. Radiologic Spectrum of Craniocervical Distraction Injuries. Radiographics 2000; 20: S237-S250. 18. Berlin L. Cervical Spine Traum: CT vs. Radiography, A Risk Management Perspective. Healthcare Risk Manager 2008; 14; 3: 1-3. 19. Horn EM, et al. Survivors of occipitoatlantal dislocation injuries: imaging and clinical correlates. J Neurosurg Spine 2007: 6: 113-120.

Visit www.SurgRad.com for high resolution images, interactive discussions, and a PDF version of this article.

ated with almost certain death. Therefore, it is reasonable to assume we may see more of these critically injured patients surviving beyond initial presentation. As such, it is imperative that trauma surgeons recognize OAD as early as possible, as failure to diagnose this injury can lead to complications and increased morbidity and mortality.10,11 Out of necessity, trauma surgeons are generally quite skilled www.JSurgRad.com

79


ORIGINAL ARTICLE

Successful Endovascular Management of a Transected Horseshoe Kidney

Paxton et al. Transected Horseshoe Kidney

Original Article

Ben E. Paxton, MD ∙ Charles Y. Kim, MD ∙ Michael J. Miller, MD ∙ Rajan T. Gupta, MD Department of Radiology, Duke University Medical Center, Durham, North Carolina

Abstract

Overview The horseshoe kidney is more prone to blunt abdominal trauma because of its low position and the presence of the isthmus across the midline. This is a rare case of complete transection of a horseshoe kidney at the isthmus due to blunt abdominal trauma with two sites of active extravasation on initial CT imaging. This extravasation was successfully treated by embolization with coils. Superselective embolization may be used for effective, minimally invasive control of active extravasation due to blunt renal trauma, even in kidneys with congenital malformations such as the horseshoe kidney.

Figure 1. Single axial contrast-enhanced CT image at the Figure 4. One of two primary sites of active extravasation level of L3 demonstrates patient's known horseshoe kidney emanating from the right moiety at its fracture site (arrowon prior unrelated CT imaging. head). Extravascular contrast is also visualized dependently within the retroperitoneal hematoma (arrow).

Keywords embolization, blunt trauma, renal, malformation, transection, superselective

Case Report

J Surg Rad

A 35-year-old male with no significant past medical history presented to Duke University Medical Center status post motor vehicle collision. The patient was an unrestrained moped driver when he collided with a stationary car. Physical examination at the time of patient presentation to the emergency department was remarkable for diffuse abdominal pain, without rebound or guarding. The patient's vital signs at the time of the primary trauma survey were as follow: blood pressure 98/52, heart rate 78 beats per minute, 18 respirations per minute, and oxygen saturation 98% on room air. He was hemodynamically stable at this time. Laboratory work-up revealed a hematocrit of 0.33. His urine was noted to be grossly bloody upon Foley catheter insertion.

80

The patient had unrelated contrast-enhanced CT imaging prior to the accident which demonstrated the presence of a horseshoe kidney (Figure 1). In the emergency department, the patient underwent urgent contrast-enhanced CT of the abdomen and pelvis, which demonstrated a complete transection at the isthmus of the horseshoe kidney with an associated large hematoma measuring 8.9 x 8.7 cm that displaced the right side of the kidney anteriorly (Figure 2). In terms of arterial supply, the right aspect Citation Paxton BE, Kim CY, Miller MJ, Gupta RT. Successful endovascular management of a transected horseshoe kidney. J Surg Radiol. 2011 Jan 1;2(1). Correspondence Rajan T. Gupta, MD E-mail rajan.gupta@duke.edu. Received August 31, 2010. Accepted September 30, 2010. Epub September 30, 2010.

Original Article | January 2011

of the horseshoe kidney was supplied by a single right renal artery arising from the abdominal aorta in a normal expected location. The left aspect of the horseshoe kidney was supplied by a main left renal artery and an accessory left renal artery, which supplied the lower moiety and originated from the left common iliac artery (Figure 3). Two distinct sites of active extravasation were identified emanating from the right moiety at its fracture site on CT (Figures 4 and 5). No definite active extravasation was visualized in association with the left aspect of the kidney. Delayed images taken approximately 10 minutes after injection of IV contrast demonstrated the bilateral collecting systems to be grossly intact without definite extravasation of excreted contrast. The patient was subsequently transferred to Vascular and Interventional Radiology for further management based on the findings of active arterial extravasation. Abdominal arteriogram was performed which again demonstrated a transected horseshoe kidney with active extravasation from branches of the right renal artery (Figure 6). The renal parenchyma enhanced fairly symmetrically on both sides. The arterial supply to the horseshoe kidney was again confirmed, with a right and left renal arteries arising from their normal expected locations from the aorta and an accessory left renal artery from the left common iliac artery (Figure 7). Selective left renal arteriography demonstrated no evidence of active extravasation. At this point in the procedure, the patient developed hemodynamic instability, requiring aggressive volume resuscitation. Subsequently, a  Mikaelsson catheter was then used to select the right renal artery and arteriography was performed (Figure 8). A microcatheter and 90 degree angled 0.018 inch hydrophilic guidewire were advanced into the

Figure 2. Complete transection of horseshoe kidney at the Figure 5. Another site of active extravasation (arrowhead) isthmus with a large hematoma displacing transected horse- located inferiorly to the extravasation in Figure 4, also emshoe kidney anteriorly (arrows). anating from the right moiety at its fracture site. Extravascular contrast is again visualized dependently within the retroperitoneal hematoma (arrow). branch artery supplying the right lower moiety of the fractured horseshoe kidney, and a superselective right renal arteriogram was performed. This demonstrated two foci of active extravasation originating from two branches of the right renal artery (Figures 9 and 10). These two arterial branches were each successfully embolized with detachable microcoils until stasis of flow was achieved (Figure 11). Final renal arteriogram demonstrated no evidence of continued extravasation. Furthermore, the patient’s hemodynamic parameters stabilized at this point in time. Post procedure, the patient did well with stabilization of serial hematocrit levels. A  follow-up IV contrast-enhanced CT scan of the abdomen and pelvis 2 days later demonstrated extravasation of contrast Figure 3. Accessory left renal artery arising from the left com- into the retroperitoneum on delayed images consistent with mon iliac artery (arrow). injury to the renal collecting systems. No active arterial exwww.JSurgRad.com

81


Original Article

Figure 6. Delayed phase abdominal arteriogram in the nephrographic phase again depicts complete transection of the horseshoe kidney with separation of the left and right moieties, with relatively well-preserved parenchymal perfusion. A tiny site of active extravasation can be visualized from a branch of the right renal artery (arrow).

Paxton et al. Transected Horseshoe Kidney

Paxton et al. Transected Horseshoe Kidney

Original Article

Figure 7. Abdominal arteriogram in the arterial phase demonstrates the arterial supply to the transected horseshoe kidney, with a single right renal artery (black arrow), dominant left renal artery (white arrow), and accessory left renal artery from the left common iliac artery (white arrowhead).

Figure 10. The injured branch artery illustrated in Figure 9 was embolized with coils; superselective injection distal to this branch demonstrates active extravasation of another small branch. travasation was identified. The patient underwent urological intervention with placement of bilateral ureteral stents. The patient's subsequent hospital course was unremarkable and he was discharged in good condition. A follow up angiogram performed 4 weeks after initial stent placement showed no active extravasation, arteriovenous fistula, or pseudoaneurysm.

Discussion

Figure 8. Selective right renal arteriogram was performed, again demonstrating active extravasation from a branch of the right renal artery (arrow).

82

Original Article | January 2011

Figure 9. Superselective right renal arteriogram demonstrates prominent active extravasation from a branch of the right renal artery (arrows).

Congenital fusion anomalies of the kidneys can generally be placed into 2 main categories: (1) horseshoe kidney and its variants and (2) crossed fused ectopia. Horseshoe kidney is the most common fusion anomaly with an overall prevalence of 0.2%, and is more common in males.1,2 It is caused by congenital fusion of the lower kidney poles with an isthmus that is either fibrous or parenchymal.3 Vascularization of the isthmus and adjacent parenchyma is markedly variable, and may arise from the aorta, the iliac arteries (common, internal, or

external), the inferior mesenteric artery, or the sacral arteries.4,5 Bilateral single arteries occur in one-third of cases.4 However, various combinations of single and multiple renal hilar and isthmus vessels are seen in two-thirds of cases.3 The isthmus of the horseshoe kidney is supplied by a single vessel from the aorta in 65% of cases.6 The current case highlights this variable arterial supply with multiple bilateral renal arteries seen at angiography, with a single dominant superior renal artery bilaterally. Furthermore, the inferior left moiety had arterial supply from the left common iliac artery (Figures 3 and 7).  The horseshoe kidney is more prone to blunt abdominal trauma because of its low position and the presence of the isthmus across the midline.3 The fused kidney is not well protected by the ribs due to its low position in the abdomen, and it is prone to compression or fracture against the lumbar vertebral bodies in response to blunt abdominal force due to its location anterior to these structures.  www.JSurgRad.com

83


Original Article

Paxton et al. Transected Horseshoe Kidney

Paxton et al. Transected Horseshoe Kidney

Original Article

Disclosures

8. Armenakas NA, Duckett CP, McAninch JW. Indications for nonoperative management of renal stab wounds. J Urol1999; 161:768–771.

The authors have no disclosures or conflicts of interest related to this manuscript.

9. Altman AL, Haas C, Dinchman KH, Spirnak JP. Selective nonoperative management of blunt grade 5 renal injury. J Urol 2000; 164:27–30.

References

1. Glodny B, Petersen J, Hofmann KJ, et al. Kidney fusion anomalies revisited: clinical and radiological analysis of 209 cases of crossed fused ectopia and horseshoe kidney. BJU Int 2008;103:224-35. 2. O'Brien J,  Buckley O,  Doody O,  Ward E,  Persaud T,  Torreggiani W. Imaging of horseshoe kidneys and their complications. J Med Imaging Radiat Oncol. 2008 Jun;52(3):216-26. 3. Murphy JT, Borman KR, Dawidson I. Renal autotransplantation after horseshoe kidney injury: a case report and literature review. J Trauma 1996; 40: 840–44. 4. Davidovic LB, Kostic DM, Jakovljevic NS, et al. Abdominal aortic surgery and horseshoe kidney. Ann Vasc Surg 2004;18:725–8. 5. Gay SB, Armistead JP, Weber ME, Williamson BR. Left infrarenal region: anatomic variants, pathologic conditions, and diagnostic pitfalls. Radiographics 1991; 11: 549–70. 6. Ferko A, Krajina A, Jon B, et al. Juxtarenal aortic aneurysm associated with a horseshoe kidney. Transfemoral endoluminal repair. Arch Surg 1997; 132: 316–17. 7. Kawashima A, Sandler CM, Corl FM, West OC, Tamm EP, Fishman EK, Goldman SM. Imaging of renal trauma: a comprehensive review. Radiographics. 2001 May-Jun;21(3):557-74.

10. Danuser H, Wille S, Zoscher G, Studer U. How to treat blunt kidney ruptures: primary open surgery or conservative treatment with deferred surgery when necessary? Eur Urol 2001; 39:9–14. 11. Dinkel HP, Danuser H, Triller J. Blunt renal trauma: minimally invasive management with microcatheter embolization experience in nine patients. Radiology. 2002 Jun;223(3):723-30. 12. Esquena Fernández S, Trilla Herrera E, Abascal Junquera JM,  Pérez M,  Morote Robles J. Arterial embolization in the treatment of renal trauma of a horseshoe kidney. Arch Esp Urol. 2005 Dec;58(10):1075-7. [Article in Spanish] 13. Trottier V,  Lortie MA,  Gouin E,  Trottier F. Renal artery avulsion from blunt abdominal trauma in a horseshoe kidney: endovascular management and an unexpected complication. Can J Surg. 2009 Dec;52(6):E291-2.  14. Pascual Samaniego M, Bravo Fernandez I, Ruiz Serrano M, et al. Traumatic rupture of a horseshoe kidney. Actas Urol Esp 2006;30: 424-8.

Visit www.SurgRad.com for high resolution images, interactive discussions, and a PDF version of this article.

Figure 11. The second small branch illustrated in Figure 10 was embolized with coils, and additional coils (arrows) were continued to be packed proximally to the origin of the preceding embolized branch. Follow up selective angiogram confirms successful embolization with stasis of flow.

84

Original Article | January 2011

In the current case, we present a rare complete transection of a horseshoe kidney at the isthmus due to blunt abdominal trauma with two sites of active extravasation on initial CT imaging. This extravasation was successfully treated by embolization with coils. Few similar cases of transected horseshoe kidneys have been reported in the literature with endovascular treatment by embolization,12 endovascular treatment by stenting,13 and surgical management.14 In conclusion, superselective embolization may be used for effective, minimally invasive control of active extravasation due to blunt renal trauma, even in kidneys with congenital malformations such as the horseshoe kidney.

MedicalRF

The sequelae of blunt renal trauma range from simple contusion to complete shattering of the kidney or avulsion of the vascular pedicle, with the vast majority of renal injuries requiring only conservative treatment.7 The advent of minimally invasive therapies for the treatment of renal injury has led to a decrease in frequency of surgical intervention for the management of renal injury.8-10 Vascular injury from penetrating or blunt trauma in the hemodynamically stable or unstable patient, including rupture of renal parenchyma, shattered kidneys, and pedicle avulsions, can frequently be effectively treated with superselective catheter embolization.11 In situations where catheter-based therapy is unsuccessful or not clinically indicated, the range of surgical options can vary and often depend on the type and severity of the injury with considerations of additional factors such as collecting system involvement.

www.JSurgRad.com

85


ORIGINAL ARTICLE

Original Article

Dy et al. Nipple Adenoma

Nipple Adenoma Benzon Dy, MD ∙ Cindy Tortorelli, MD ∙ Sejal Shah, MD ∙ Judy C. Boughey, MD Department of Surgery, Radiology, and Pathology, Mayo Clinic, Rochester, Minnesota.

Abstract

Introduction Nipple adenomas are rare benign neoplasms that closely mimic malignant disease. They arise from lactiferous ducts and milk sinuses. A retrospective review of nipple adenomas at our institution from 1992 to 2010 was performed with review of pathology, radiographic studies and clinical histories. Methods and Results Twelve patients with a nipple adenoma were identified. Mean age at diagnosis was 58.5 years (range 43-76). Ten women (83%) were symptomatic. Symptoms included nipple discharge (58.3%), palpable lesions (50%), and ulcerative skin changes (25%). Workup included mammography (12), ultrasound (8) and magnetic resonance imaging (MRI) (1). The lesion size ranged from 0.4-1.5 cm (mean 0.7 cm). Four patients had complete nipple excision, six had partial nipple excision, one had removal via percutaneous biopsy, and one had a mastectomy for history of breast cancer. With 38.3 months follow-up, 3 women (25%) had a diagnosis of breast cancer prior to or after diagnosis of nipple adenoma, and there has been 1 recurrence of nipple adenoma. Discussion Nipple adenoma is a rare condition with non-specific presentation. Diagnosis is usually made by complete surgical excision. It is important to differentiate these lesions from malignancy. Although no clear correlation has been shown between nipple adenoma and breast cancer, further investigation is warranted.

J Surg Rad

Keywords benign breast disease, papillomatosis, sclerosing adenomatosis

86

Introduction

Materials and Methods

Adenomas of the nipple (erosive adenomatosis, papillary adenoma, florid papillomatosis) are rare neoplasms that arise from the lactiferous ducts and milk sinuses. Clinically, they can present as ulcerations, erosions, nipple discharge, palpable masses, or as erythematous growths. Nipple adenomas are considered benign; however, they often prove to be a diagnostic dilemma because their clinical and radiologic appearances may mimic Paget’s disease, low grade well differentiated invasive ductal adenocarcinoma, or lymphoma of the breast.1 Mammography and ultrasound provide important information in the workup. The definitive diagnosis and treatment is complete surgical excision with histologic confirmation.2 We present the clinical presentation, radiographic findings, pathology, treatment, and outcomes of 12 patients with nipple adenoma.

Institutional Review Board approval was obtained for a retrospective review of the clinical database to identify all cases of nipple adenoma treated at our institution between 1992 and 2010. Histology slides were reviewed by a breast pathologist to confirm diagnosis of nipple adenoma. Radiological studies and clinical histories were also reviewed.

Citation Dy B, Tortorelli C, Shah S, Boughey JC. Nipple adenoma. J Surg Radiol. 2011 Jan 1;2(1). Correspondence Judy C. Boughey, MD E-mail boughey.judy@mayo.edu. Received September 5, 2010. Accepted September 11, 2010. Epub September 12, 2010.

Original Article | January 2011

Results We identified 12 patients with a confirmed histologic diagnosis of nipple adenoma. Mean age at diagnosis was 58.5 years (range 43-76). Ten women (83%) were symptomatic at the time of diagnosis. Seven patients (58.3%) presented with nipple discharge, 6 patients (50%) had palpable lesions, and 3 patients (25%) had ulcerative skin changes. Other findings included pain, discoloration, crusting, erythema and pruritus. The duration of symptoms ranged from 1 week to 24 months (mean 6.2 months). Preoperative differential diagnosis among breast surgeons and dermatologists included nipple adenoma, Paget’s disease, ductal carcinoma in situ (DCIS), and infiltrating ductal carcinoma. Two patients were asymptomatic at the time of diagnosis. In one patient, the nipple adenoma was found on screen-

Figures 1 and 2. Magnification mammograms demonstrating pleomorphic calcifications within the left nipple. ing mammogram and was completely excised by percutaneous biopsy. The second asymptomatic patient had the nipple adenoma incidentally identified on pathologic review after a simple mastectomy. The patient had undergone breast conservation and radiation for an invasive ductal carcinoma two years previously with a poor cosmetic outcome and decided on mastectomy and reconstruction.

Radiologic Imaging All 12 women had mammography, 8 women had an ultrasound performed in addition to mammography, and 1 had an MRI in addition to mammography and ultrasound. A distinct mass or suspicious calcifications in the nipple were preoperatively identified in 6 patients.

within the nipple on T1 weighted images (Figures 4 and 5).

Treatment One patient had percutaneous biopsy for diagnostic purposes in which the adenoma was completely removed and did not undergo further intervention. Eleven patients underwent surgical excision for diagnosis as well as therapeutic management. Four patients had complete excision of their nipple, six had partial nipple excision with a focus of the involved duct, and one woman had mastectomy. On mastectomy, incidental finding of nipple adenoma in permanent sectioning of the nipple was found. One woman with partial nipple excision

On mammography, the most common finding was calcifications in four patients in the subareolar breast tissue and within the nipple (Figures 1 and 2). Calcifications when they were present were most often pleomorphic (Figure 3). Adenomas can present as masses on mammogram and ultrasound. The nipple masses can have circumscribed, lobulated, or spiculated margins. Lesion size on imaging correlated well with size on histologic examination. The one case which underwent MRI had both negative sonographic and mammographic studies. Her imaging was performed with contrast and showed focal, intense enhancement

Figure 3. Left breast ultrasound demonstrating hyperechoic mass with internal calcifications within the left nipple. www.JSurgRad.com

87


Original Article

Dy et al. Nipple Adenoma

Original Article

Dy et al. Nipple Adenoma

Figures 4 and 5. Sagittal and axial breast MRI post contrast T1 weighted images demonstrating marked focal uptake within the right nipple.

also underwent wide local excision of a separate biopsy site due to a synchronous DCIS lesion.

Pathology Pathologic size of the nipple adenoma ranged from 0.4-1.5 cm (mean 0.7 cm). In general, these lesions are grossly firm and illdefined nodules and are characterized by proliferating ductal structures lined by columnar epithelium. Apical intraluminal projections are commonly found.

Figures 6 and 7. Nipple adenoma with proliferation of benign ducts and glands within a fibrotic stroma (Figure 6 - H&E x40) with intact myoepithelial cells (Figure 7, facing page, H&E x100).

88

Original Article | January 2011

In our series, microscopic examination revealed relatively well-demarcated proliferation of benign glands and ducts with an intact myoepithelial cell layer within a fibrotic stroma (Figure 6). Most cases showed expansion of glands by ductal hyperplasia of usual type. One case showed nipple adenoma with focal atypical ductal hyperplasia. None of the cases showed in situ or invasive malignancy arising within a nipple

adenoma. In one of the cases, however, a concurrent DCIS was identified in the breast parenchyma adjacent to the nipple adenoma. Because there are a myriad of clinical presentations for nipple adenoma, these key histologic features are essential to diagnosis.

Outcome Three women (25%) have had a diagnosis of DCIS or invasive ductal carcinoma either before or after discovery of their nipple adenoma. One woman had a wide local excision of invasive ductal carcinoma and two years later had incidental discovery of her nipple adenoma during mastectomy with breast reconstructive surgery. Another woman subsequently developed ipsilateral DCIS and contralateral invasive ductal carcinoma eight years after excision of the nipple adenoma. The third woman had concurrent contralateral DCIS biopsied www.JSurgRad.com

89


90

Original Article | January 2011

Mammo, US Bloody No No

Mammo, US None No Yes

Mammo, US None No Yes

Mammo, US Serous No Yes

Mammo, US Bloody Yes No

Mammo, US None No Yes

Recurrent Adenoma 0.4 x 0.3 cm Mammo, US, MRI Yes

Yes

Bloody

Concurrent DCIS - IDC 8 years after nipple adenoma excision Mammo, US None

Concurrent contralateral DCIS Mammo Bloody No

Previous ipsilateral breast IDC

Mammo

Mammo None

Bloody

Serous No

No No

8. Matsubayashi RN, Adachi A, Yasumori K, et al. Adenoma of the nipple: correlation of magnetic resonance imaging findings with histologic features. J Comput Assist Tomogr 2006; 30(1):148-50.

Yes

7. Taylor HB, Robertson AG. Adenomas of the nipple. Cancer 1965; 18:995-1002.

No

6. Clune JE, Kozakewich HP, VanBeek CA, et al. Nipple adenoma in infancy. J Pediatr Surg 2009; 44(11):2219-22.

Mammo

5. Ishii N, Kusuhara M, Yasumoto S, Hashimoto T. Adenoma of the nipple in a Japanese man. Clin Exp Dermatol 2007; 32(4):448-9.

Yes

4. Albers SE, Barnard M, Thorner P, Krafchik BR. Erosive adenomatosis of the nipple in an eight-year-old girl. J Am Acad Dermatol 1999; 40(5 Pt 2):834-7.

No

3. Perzin KH, Lattes R. Papillary adenoma of the nipple (florid papillomatosis, adenoma, adenomatosis). A clinicopathologic study. Cancer 1972; 29(4):996-1009.

No

Malignancy Imaging Discharge

2. Healy CE, Dijkstra B, Walsh M, et al. Nipple adenoma: a differential diagnosis for Paget's disease. Breast J 2003; 9(4):325-6.

0.6 x 0.5 cm

55

47

11

12

Partial Nipple Excision

0.5 x 0.5 cm

65 10

Partial Nipple Excision

1.5 x 1.0 cm

51 9

Complete Nipple Excision

1.3 x 1.2 cm

43 8

Complete Nipple Excision

0.5 x 0.4 cm

63 7

Complete Nipple Excision

0.4 x 0.3 cm

46 6

Partial Nipple Excision

0.6 x 0.4 cm

60 5

Partial Nipple Excision

0.5 x 0.3 cm

70 4

Core Biopsy

0.7 x 0.5 cm

53 3

Partial Nipple Excision

0.6 x 0.3 cm

76 2

Mastectomy

0.5 x 0.3 cm

73 1

Visit www.SurgRad.com for high resolution images, interactive discussions, and a PDF version of this article.

Partial Nipple Excision

11. Gudjonsdottir A, Hagerstrand I, Ostberg G. Adenoma of the nipple with carcinomatous development. Acta Pathol Microbiol Scand A 1971; 79(6):676-80.

Complete Nipple Excision

10. Kuflik EG. Erosive adenomatosis of the nipple treated with cryosurgery. J Am Acad Dermatol 1998; 38(2 Pt 1):270-1.

0.6 x 0.5 cm

9. Handley RS, Thackray AC. Adenoma of nipple. Br J Cancer 1962; 16:187-94.

Table 1.

Nipple adenomas are regarded as benign lesions. However, there have been isolated reports of malignant synchronous foci in these lesions.11 Some have questioned as to whether these adenomas serve as a precursor for malignant lesions

1. Miller L, Tyler W, Maroon M, Miller OF, 3rd. Erosive adenomatosis of the nipple: a benign imitator of malignant breast disease. Cutis 1997; 59(2):91-2.

Ulceration

It is important to differentiate these lesions from malignant lesions, such as Paget’s disease and invasive adenocarcinoma, since the treatments are quite different. Imaging can assist in differentiating between benign and malignant disease processes, but there are few, if any, pathognomonic radiologic findings for nipple adenoma. MRI findings have been shown to correlate with certain histologic features but do not provide diagnostic certainty in the absence of a tissue sample.8 Definitive treatment of nipple adenoma has traditionally involved complete excision of the nipple areolar complex.9 However, recent reports, including our own, have shown acceptable cure rates with partial nipple resection when the adenoma is excised with negative margins.2 Novel treatments such as cryosurgery and Mohs surgery have been reported with good outcomes as well when negative margins are achieved.10

References

Palpable

In addition to its rarity, these neoplasms are difficult to diagnose because of their non-specific presentation. Duration of symptoms prior to a clinical diagnosis ranges from 1 month to 14 years.7 In our series, all cases were diagnosed within two years of symptomatology. The most common presentation of these lesions in our series is nipple discharge. Other frequent symptoms include palpable mass, crusting, ulceration, and erythema of the nipple. Preoperative biopsy is challenging due to the lesions location within the nipple, therefore, surgical excision is both diagnostic and therapeutic.

The authors have no disclosures or conflicts of interest related to this manuscript.

Size

Nipple adenoma is a rare condition most commonly found in women with peak rates in the 5th decade.3 They have also been reported in men as well as in children as young as 5 months old.4-6

Disclosures

Type of Surgery

Discussion

or increase risk for future carcinoma. Epithelial hyperplasia has been suggested as a mechanism for increased risk of carcinoma development. In this series, three patients developed ductal carcinoma in situ or invasive ductal carcinoma either before or after their diagnosis of nipple adenoma. No clear correlation has ever been shown with nipple adenoma and development of breast cancer. Further investigation may be warranted to delineate a relationship between diagnoses.

Age

With a mean follow up of 38.3 months (range = 1-155 months, median 29.0 months), there has been 1 recurrence of nipple adenoma. This recurrence occurred in a patient who had a partial excision of the nipple for resection of a 0.6 cm adenoma and 42 months later presented with bloody nipple discharge. This was treated with a re-excision of part of the nipple that revealed a 0.4 cm nipple adenoma. With 18 months follow up after re-excision, there has been no sign of additional disease.

Original Article

Dy et al. Nipple Adenoma

Patient

one month after removal of her nipple adenoma.

Dy et al. Nipple Adenoma

No

Original Article

www.JSurgRad.com

91


ORIGINAL ARTICLE

Molecular Breast Imaging for Women Presenting With a History of Non-Reproducible Bloody Nipple Discharge and Negative Findings on Routine Imaging Studies: A Pilot Study

Wahner-Roedler et al. Molecular Breast Imaging Evaluation

developed in the 1990s and has been the subject of considerable investigation. However, conventional scintimammography does not reliably detect tumors less than 10 mm,5 and thus this technique has not been widely adopted in the United States. In an attempt to overcome the limitations of conventional scintimammography, several gamma cameras with a small field of view have been developed that permit imaging of the breast in a similar manner and orientation to conventional mammography. The close proximity of small detectors to the breast provides a considerable improvement in resolution compared with conventional scintimammography.

Dietlind L. Wahner-Roedler, MD ∙ Carrie B. Hruska, PhD ∙ Michael K. O’Connor, PhD ∙ Stephen W. Phillips, MD ∙ Dana H. Whaley, MD ∙ Ruth E. Johnson, MD ∙ Amy C. Degnim, MD ∙ Judy C. Boughey, MD ∙ Deborah J. Rhodes, MD  Division of General Internal Medicine, Mayo Clinic, Rochester, Minnesota.  The Department of Radiology, Mayo Clinic, Rochester, Minnesota.  The Division of Preventive, Occupational and Aerospace Medicine, Mayo Clinic, Rochester, Minnesota.  The Division of Gastroenterologic and General Surgery; Mayo Clinic, Rochester, Minnesota.

Abstract

Mayo Clinic in Rochester, Minnesota, developed an imaging system that uses two small, opposing semiconductor gamma camera detectors. This technique has been labeled molecular breast imaging (MBI).6 Previous studies have showed a high sensitivity for the detection of small breast tumors, including mammographically occult tumors.7 The purpose of the present pilot study was to evaluate MBI in the work-up of patients with a history of BND that was not reproducible in the clinic and had negative results on routine imaging studies.

Introduction The purpose of this study was to assess the role of molecular breast imaging (MBI) for women with a history of non-reproducible bloody nipple discharge (BND) and negative findings on clinical breast examination, mammography, and ultrasonography. Methods and Results MBI was performed on 14 women meeting these criteria and interpreted as negative (n=7), benign (n=6), and probably benign (n=1). Two patients with a negative result underwent duct exploration, revealing papilloma and radial scar. Median follow-up was 20 months (range, 0-35 months). One patient had another episode of BND at 37 months; excisional biopsy result was benign. Invasive carcinoma and phyllodes tumor diagnosed in two patients at 17 and 25 months were assumed to be unrelated to the reported BND episode. Discussion The concordance between MBI, mammography, and ultrasonography regarding the absence of suspicious findings in women presenting with a history of non-reproducible BND is reassuring. Larger studies are needed to confirm these preliminary findings. Keywords breast, drainage, molecular diagnostic techniques, nipples

Introduction

J Surg Rad

The appearance of secretion or discharge from nipple ducts is a common occurrence. As many as 80% of the women are able to express one or more drops of fluid from a nipple at some time during their reproductive years.1 Most isolated nipple discharge (ND) is benign in origin. In one series, for example, the overall incidence of malignancy was 4.8% in 1,145 women with ND evalu-

92

Citation Wahner-Roedler DL, Hruska CB, O’Connor MK, Phillips SW, Whaley DH, Johnson RE, Degnim AC, Boughey JC, Rhodes DJ. Molecular breast imaging for women presenting with a history of nonreproducible bloody nipple discharge and negative findings on routine imaging studies: a pilot study. J Surg Radiol. 2011 Jan 1;2 (1). Correspondence Dietlind L. Wahner-Roedler, MD E-mail wahnerroedler.dietlind@mayo.edu. Received October 1, 2010. Accepted November 17, 2010. Epub November 23, 2010.

Original Article | January 2011

ated between January 2, 1960, and December 1974.2 The risk of cancer is greater when the discharge is bloody or positive for guaiac, is unilateral, is associated with a breast mass, or occurs in a woman older than 40 years. Bloody nipple discharge (BND) occurs when a lesion in the duct is bleeding. There are three main causes of BND. Papillomata account for 35% to 48% of the cases; ductal ectasia associated with an inflammatory infiltrate accounts for 17% to 36%; and carcinoma accounts for 5% to 21% of cases.3 Evaluation of BND is focused on excluding a malignant cause. A thorough physical examination is imperative because associated palpable masses increase the risk of malignancy.4 Although a mammogram and an ultrasonographic scan (US) are indicated in the evaluation of unilateral, spontaneous ND, they do not exclude the possibility of a malignant cause. When BND is clinically reproducible, management traditionally involves surgical excision, regardless of imaging findings. When BND is not reproducible and mammogram and US results are negative, management usually

Original Article

Methods Study Participants

Figure 1. Dual-Detector Molecular Breast Imaging System. relies on watchful waiting – serial breast imaging and continued observation for recurrent BND. However, watchful waiting provokes anxiety among patients who do not have a clear explanation for their BND. Breast magnetic resonance imaging (MRI) has not been studied systematically in this context and is costly. The availability of a less costly method to either identify a cause or confirm the validity of a negative mammogram and US would be of great value to these patients and their physicians. Scintimammography with technetium Tc 99m sestamibi was

Eligible participants were women who presented to a multidisciplinary breast clinic for evaluation of BND that was not reproducible in the clinic and who subsequently had negative results on a clinical breast examination and on diagnostic mammogram and US studies. Women who were pregnant or breastfeeding were excluded. Informed consent was obtained from all participants, and the protocol was approved by the Mayo Clinic Institutional Review Board and registered as NCT00566280 in ClinicalTrials.gov.

Equipment A dual-detector cadmium-zinc-telluride gamma camera syswww.JSurgRad.com

93


Original Article

Wahner-Roedler et al. Molecular Breast Imaging Evaluation

Wahner-Roedler et al. Molecular Breast Imaging Evaluation tem mounted on a modified mammography gantry (Figure 1) was used to perform MBI. This system has been described in detail.6-8 Briefly, each 20 × 15 cm detector (Gamma MedicaIdeas, Inc, Northridge, California) consisted of pixilated arrays of 128 × 96 cadmium-zinc-telluride elements with pixel sizes of 1.6 × 1.6 mm. This small pixel size offers improved spatial resolution compared with continuous sodium iodide crystals used in conventional scintimammography. In addition, the cadmium-zinc-telluride material has superior energy resolution compared with conventional gamma cameras – about 3.8% at 140 keV – and offers improved ability to reject scattered gamma rays in the image.

RMLO

LMLO

RCC

LCC

Figure 2. MBI - Interpretation negative (uniform uptake of Tc-99m sestamibi in the breast with no focal areas of uptake – except Figure 2. uptake). Molecular Imaging LMLO of Patient RCC 6. Interpretation was a negative result RMLO LCC(unifor nipple (Patient Breast # 1)

form uptake of technetium Tc 99m sestamibi in the breast, with no focal areas of uptake Figure 2. MBI - Interpretation negative (uniform uptake of Tc-99m sestamibi in the breast with no focal areas of uptake – except except left craniocaudal; LMLO, left mediolateral oblique; RMLO nipple uptake). LCC indicates LMLO RCC LCC for nipple uptake). (Patient # 1) RCC, right craniocaudal; RMLO, right mediolateral oblique. Figure 2. MBI - Interpretation negative (uniform uptake of Tc-99m sestamibi in the breast with no focal areas of uptake – except for nipple uptake). (Patient # 1)

Imaging Procedure Patients received an intravenous injection of 20 mCi of Tc 99m sestamibi. Craniocaudal and mediolateral oblique MBI views of each breast were obtained at 10 minutes per view. The breast was positioned between the two detectors, and light, pain-free compression (15-lb force) was applied to reduce breast thickness and limit movement artifact.

MBI Scoring System

RMLO

LMLO

RCC

LCC

Figure 3. MBI - Interpretation benign (patchy [non-mass like] uptake or uptake in areas of known inflammation, mild uptake in RMLO LMLO RCC LCC the nipple). (Patient #5) Figure 3. MBI - Interpretation benign (patchy [non-mass like] uptake or uptake in areas of known inflammation, mild uptake in

RMLO 3. Molecular Breast Imaging of LMLO RCC LCC Figure Patient 14. Interpretation was a benign result (patchy the nipple). (Patient #5) [nonmasslike] uptake or uptake in areas of known inflammation, with mild uptake in the Figure 3. MBI - Interpretation benign (patchy [non-mass like] uptake or uptake in areas of known inflammation, mild uptake in nipple). LCC indicates left craniocaudal; LMLO, left mediolateral oblique; RCC, right craniothe nipple). (Patient #5) caudal; RMLO, right mediolateral oblique.

RMLO

LMLO

RCC

LCC

Figure 4. MBI - Interpretation probably benign (one or more areas of mild focal uptake, focal uptake only seen on one view). RMLO LMLO RCC LCC (Patient # 12) Figure 4. MBI - Interpretation probably benign (oneLMLO or more areasRCC of mild focal uptake, focal uptake only seen on one LCC view). RMLO (Patient # 12) Figure 4. MBI - Interpretation probably benign (one or more areas of mild focal uptake, focal uptake only seen on one view). Figure 4. Molecular Breast Imaging of Patient 12. Interpretation was of a probably benign (Patient # 12) result (≥1 area of mild focal uptake, focal uptake seen on 1 view only). LCC indicates left craniocaudal; LMLO, left mediolateral oblique; RCC, right craniocaudal; RMLO, right mediolateral oblique.

94

Original Article | January 2011

A radiologist reviewed the MBI views for presence of abnormal radiotracer uptake. Clinical history details, mammographic images, and ultrasounds were available of each patient. The following scoring system was used to evaluate MBI images. • Negative: uniform uptake of Tc 99m sestamibi in the breast, with no focal areas of uptake except nipple uptake. • Benign: patchy (non-mass-like) uptake or uptake in areas of known inflammation, with mild nipple uptake. • Possibly benign: One or more areas of mild focal uptake, and focal uptake seen on one view only. • Suspicious for malignancy: One or more areas of moderate focal or diffuse uptake. • Highly suggestive of malignancy: One or more focal areas of intense uptake. In cases where the MBI was interpreted as negative or benign, the usual standard of care was followed with a recommendation of a 12-month follow-up screening mammogram or follow-up at any time if BND recurred. In cases with a positive finding on MBI (classified as probably benign, suspicious, or highly suggestive of malignancy), the radiologist determined appropriate additional work-up, which may have included special mammographic views, focused breast US, contrast-enhanced breast MRI, image-guided biopsy, or a 6-month follow-up mammogram, US, or MBI, or a combination of these techniques. Patients were advised to return at any time if BND recurred or for any breast changes of concern.

Original Article

Results Between October 1, 2006, and December 31, 2007, 14 women were recruited into the study. Their demographic characteristics and clinical and imaging findings are summarized in the table. Examples of a negative (patient 6), benign (patient 14), and probably benign (patient 12) MBI interpretation are depicted in figures 2-4. The median number of days between the episode of BND and presentation to the clinic were 22 (range, 2-76 d); the median number of days between presentation to the clinic and the performance of MBI was 22 (range, 0-163 d). The median length of follow-up after performance of MBI was 20 months (range, 0-35 mo). For nine of the 14 patients with spontaneous non-reproducible BND, a surgical consultation was recommended by the breast clinic staff physician. Seven patients had this appointment with a surgeon. For five of these seven patients, observation was advised and the patients were asked to monitor for future episodes of BND, recording when possible the location of the involved duct or the subareolar quadrant to which the application of pressure elicited drainage. In the other two patients, subareolar duct excision was performed because of the patient’s concern about a family history of breast cancer. Pathologic evaluation showed a small radial scar in one patient (patient 1) and a small papilloma (0.3 × 0.2 × 0.2 cm) in the other (patient 4). These benign lesions were undetected by mammogram, US, and MBI. All patients were advised to return if they noticed recurrent BND. One patient (patient 5) returned eight months after the initial visit with an episode of non-reproducible clear ND, but not BND. In consideration of her significant family history, an MRI of the breast was performed and showed negative results. Surgical consultation was obtained and observation advised. Another patient (patient 7) returned one year after the initial episode, with a history of non-reproducible greenish ND, but not BND. Clinical breast examination and standard imaging studies showed negative results. Observation was advised. Patient 12 with both an MBI interpretation of probably benign and a spontaneous BND underwent post-MBI special mammographic views and targeted repeat US examinations, which showed negative results. At the radiologist’s recommendation, the patient returned for a 6-month follow-up MBI, and this study was interpreted as benign. Two patients required subsequent breast surgery, one (patient 4) with a benign phyllodes tumor at 25 months and one (patient 8) with cancer at 17 months. Patient 4 had undergone subareolar duct excision after MBI, with a finding of benign intraductal papilloma. Follow-up for this patient included screening mammogram at 13 months and at 25 months after MBI. Screening mammography at 25 months after MBI showed a region of asymmetrical density in the ipsilateral breast. Directed US showed a 23 ×  20  ×  10 mm

www.JSurgRad.com

95


96

Original Article | January 2011

42

45

55

11

12

13

Arrow indicates progression. b

74 10

44

36 9

14

75 8

26 5

65

53 4

46

37 3

6

36 2

7

66

In all the patients, the bloody nipple discharge could not be reproduced in the clinic. Clinical breast examination showed normal results, and all 14 patients had negative findings on mammography of the symptomatic breast. Prominent ducts were noted on mammography of 1 patient (patient 10).

13 None Left

Spontaneous

Negative

Benign

Surgical consultation → observation

2

21 None Left

Expressed

Negative

Negative

Surgical consultation → observation

2

19 Right

Spontaneous

Negative

Probably benign

None

2

19

Surgical consultation → observation

1 None Observation Negative Negative Spontaneous Left

19 Right

Spontaneous

Prominent ducts

Negative

None

2

8 1 None Observation

Surgical consultation advised → patient did not have

Negative Negative Unknown Left

30 5 17 mo after MBI: infiltrating ductal carcinoma and ductal carcinoma in situ Surgical consultation advised → patient did not have Benign Dilated ducts Spontaneous Right

12

25 2

1 None

None

Observation

Observation Benign

Negative Prominent ducts

Dilated ducts Expressed

Spontaneous Left

Left

4 Left

Expressed

Negative

Benign

1 None Surgical consultation → observation

25 2 Negative Bilateral

Left: spontaneous; right: expressed

Benign

Surgical consultation → left breast subareolar duct excision

0 0 None

2 d after MBI: left breast intraductal papilloma with associated duct ectasia 25 mo after MBI: left breast benign phyllodes tumor

Observation Negative Prominent ducts Spontaneous Right

0 None Left

Expressed

Negative

Benign

Surgical consultation → observation

0

24 2 Left

Spontaneous

Dilated ducts

Negative

5 d after MBI: dilated ducts with periductal fibrosis and inflammation; focal florid ductal hyperplasia and small radial scar Surgical consultation → left breast subareolar duct excision

Interpretation

Abbreviations: BND, bloody nipple discharge; MBI, molecular breast imaging; MG, mammogram; US, ultrasonographic scan.

a

14

21

19

21

19

8

30

12

28

12

31

0

35

31

visit, mo

and

Last mo

Last MG/US,

Time Between MBI MG Since

MBI, No.

Mayo Clinic at

Follow-up

Pathologic Findings

1

In a retrospective chart review of 306 patients with suspicious ND and negative results on standard evaluation (clinical breast examination, mammogram, US), 52 patients underwent an

In patients presenting with a history of BND that is not reproducible, caution must be exercised in assuming that a patient report of “bloody” discharge is indeed of blood. Many breast secretions, especially those that are green, black, or brown, may look like blood to a patient. In general, breast surgeons do not consider a patient report of a bloody discharge alone

y

A preliminary study reported retrospectively on 23 patients with ND who underwent MRI.12 Mammographic findings were negative in 22 patients. In 11 of the 15 patients who underwent an excisional biopsy, MRI findings correlated with histopathologic findings. The investigators concluded that MRI can help identify both benign and malignant causes of ND.

If BND is clinically reproducible, management traditionally involves surgical excision, even when results of imaging studies are negative.4,17 A terminal duct excision is both diagnostic and, for discharges that turn out to have a benign cause, therapeutic. Usually, the duct with discharge is cannulated before the incision, for use as a guide to dissection and to ensure removal of the involved duct.

Age,

Breast MRI is an effective diagnostic technique to identify clinically and mammographically occult cancers.9 The role of MRI has not been defined yet in the evaluation of patients with pathologic ND in whom other nonsurgical methods have failed to identify an underlying etiologic cause. There are isolated case reports of patients with pathologic ND in whom routine imaging studies had negative results and MRI was helpful in finding an underlying malignancy.10,11

Preliminary patient studies with the MBI system of our institution indicate that this system is capable of reliably detecting 5-10 mm lesions in the breast with Tc 99m sestamibi. In our ongoing MBI studies of 150 patients with suspicious mammographic abnormalities, we have obtained an overall sensitivity of 91%. The sensitivity of MBI for cancer of 5 mm or less, 6 to 10 mm, and greater than 10 mm was 69%, 91%, and 97%, respectively.7 In addition, MBI has showed approximately 10% more carcinomas in these patients that were otherwise undetectable with conventional methods (mammogram and clinical examination).6,15

Treatment Decisionb

Women presenting with a history of BND that cannot be reproduced in the clinic who have negative results on work-up with conventional imaging procedures (i.e., mammography and US) present a diagnostic problem in clinical practice. Usually, observation is advised and the patients are asked to return if BND recurs. This watchful waiting can provoke anxiety in the patient because of the inability to explain the cause of the discharge at presentation and since an underlying malignancy cannot be excluded. Patients or their clinicians, or both, may request an additional imaging procedure (such as MRI) to exclude a diagnosis of cancer.

These incidental findings can have an obvious impact on both the patient and the health care delivery system and as such should be considered when integrating breast MRI into the evaluation of patients with suspicious ND. In addition to these limitations, the high cost of MRI underscores the need for alternative imaging approaches in the evaluation of nonreproducible BND.

MBI

Discussion

Ultrasonographic Findings

During a mammographic examination 12 months after the diagnosis of breast cancer, the patient was noted to have an episode of BND. The mammogram showed negative results, US showed a hypoechoic mass, and excision biopsy found benign breast parenchyma with duct ectasia associated with intraductal necrotic debris and dense stromal fibrosis.

Although breast MRI is more sensitive than mammography for the detection of breast cancer, its low specificity limits its clinical utility.14 Diagnostic MRI may detect additional incidental lesions in either the ipsilateral or contralateral breast that require follow-up imaging or biopsy. In the series reported by Morrogh et al,13 24 of the 52 patients (46%) who underwent MRI for further evaluation of suspicious ND were reported to have additional incidental findings in the index breast. Repeat imaging was recommended and undertaken for all 24 patients, three of whom subsequently required biopsy, yielding benign results in all three cases. The median number of follow-up scans for these 24 patients was one (range, 1-7), and at last follow-up, no further cancers were identified. Of 45 women who had bilateral MRI, one patient required biopsy for a suspicious lesion in the contralateral breast, which ultimately proved to be benign.

Nature of BND

Only one patient had breast cancer in the course of follow-up. Patient 8, a 75-year-old woman with a history of spontaneous right ND, received a diagnosis of invasive breast cancer of the right breast 17 months after undergoing evaluation in the breast clinic and an MBI. Mammogram at that time showed worrisome calcifications with some associated soft-tissue density in the upper outer quadrant of the right breast 12 cm from the nipple. US showed the area of calcification but did not show a classic malignant-appearing mass. Stereotactic biopsy showed invasive ductal carcinoma of Nottingham grade 3 of 3 with microcalcifications and associated ductal carcinoma in situ. Wide local excision found a 0.8 × 0.7 × 0.7 cm mass; three sentinel lymph nodes were negative.

MRI.13 MRI detected seven cancers and three high-risk lesions but failed to identify one cancer and one high-risk lesion. The positive and negative predictive values of MRI in this review were 56% and 87%, respectively (sensitivity, 77%; specificity, 62%). The investigators concluded that MRI should not replace major duct excision as the gold standard to exclude malignancy in patients with ND and negative results on standard evaluation.

Side of BND

lesion and US-guided biopsy showed a phyllodes tumor with usual ductal hyperplasia. Excisional biopsy found a benign phyllodes tumor (1.8 × 1.6 × 1.4 cm).

Original Article

Wahner-Roedler et al. Molecular Breast Imaging Evaluation

Patient No.a

Wahner-Roedler et al. Molecular Breast Imaging Evaluation

Table. Demographic Characteristics and Clinical and Imaging Findings of 14 Patients Presenting With a History of Bloody Nipple Discharge

Original Article

www.JSurgRad.com

97


Original Article

Wahner-Roedler et al. Molecular Breast Imaging Evaluation

Wahner-Roedler et al. Molecular Breast Imaging Evaluation

a sufficient indication for surgery.18 In a study by Dawes et al,18 4 of 91 patients with ND reported BND that could not be reproduced. The patients’ evaluation was unremarkable, and they had no further breast problem with observation alone.

regarding absence of suspicious findings documented in this study is reassuring. Larger studies are needed to confirm these preliminary findings.

16. O’Connor MK, Hruska CB, Phillips SW, Whaley DH, Blevis I, Wagenaar DJ, et al. Molecular breast imaging: a new technique for the accurate detection of smalls breast lesions [abstract]. Breast Cancer Res Treat. 2006;100(Suppl 1):S33-4.

Of our 14 patients, seven were seen in surgical consultation, and duct exploration was advised for two, mainly because of a positive family history. A subareolar duct excision was performed. In both patients, nonmalignant etiologic factors for the BND were found. These benign lesions were not detected on mammogram, US, and MBI.

Disclosures

17. Wahner-Roedler DL, Reynolds C, Morton MJ. Spontaneous unilateral nipple discharge: when screening tests are negative: a case report and review of current diagnostic management of a pathologic nipple discharge. Breast J. 2003 Jan-Feb;9(1):49-52.

We assume that the finding of an invasive ductal carcinoma in one patient (patient 8), diagnosed 17 months after her initial evaluation for non-reproducible, nonrecurrent BND, was not related to her ND since the location of her tumor was peripheral rather than central. It is unlikely that the ND in patient 4 was related to the phyllodes tumor, since this patient did undergo a subareolar duct excision for evaluation of her BND, showing a papilloma that explained her BND and resulted in resolution of the discharge. To our knowledge, the present study is the first study to examine the role of MBI – an imaging procedure that has been shown to detect mammographically occult lesions – in the diagnostic evaluation of ND. This pilot study evaluated MBI as an additional diagnostic procedure for patients presenting with a history of non-reproducible BND. The concordance between MBI, mammogram, and US in regard to the absence of suspicious findings is reassuring. There were no false-positive findings on MBI, in contrast to the large number of false-positive MRI findings reported in the study by Morrogh et al.13 Although two patients who underwent duct excision were noted to have benign lesions, none of these lesions were visible on MBI. Advantages of MBI relative to MRI include the cost and the complexity of interpretation. It is estimated that the cost of MBI will be 4 to 6 times less than MRI. A bilateral MRI generates approximately 1,000 images, whereas dual-head bilateral MBI generates 4 to 8 images. A larger study comparing MBI with MRI in this clinical context is needed to assess the sensitivity, specificity, and costeffectiveness of each of these procedures for clinical practice.

Conclusion Although the gold standard of diagnostic and therapeutic approach for patients presenting with BND regardless of imaging findings is surgical duct excision, the therapeutic management of patients presenting with a history of non-reproducible BND and negative findings on clinical examination and routine imaging studies relies on watchful waiting. The availability of an additional diagnostic method – less costly than MRI – to either identify a cause or confirm the validity of standard imaging tests would be of great value to clinical practice. In this pilot study, MBI was performed on 14 women presenting with a history of non-reproducible BND and negative findings on clinical examination and routine imaging studies. The concordance between MBI, mammography, and ultrasonography

98

Original Article | January 2011

Presented in part at the 12th World Congress on the Menopause, Madrid, Spain, May 19-23, 2008. Portions of this manuscript have been published in abstract form: Climacteric. 2008 Feb; 11 Suppl 1:290-1. PS-413.

Original Article

Visit www.SurgRad.com for high resolution images, interactive discussions, and a PDF version of this article.

18. Dawes LG, Bowen C, Venta LA, Morrow M. Ductography for nipple discharge: no replacement for ductal excision. Surgery. 1998 Oct;124(4):685-91.

The study sponsors had no role in any aspect of study design, collection, analysis and interpretation of data, or writing and submitting the manuscript. No competing financial interests exist.

References

1. Isaacs JH. Other nipple discharge. Clin Obstet Gynecol. 1994 Dec;37(4):898-902. 2. Murad TM, Contesso G, Mouriesse H. Nipple discharge from the breast. Ann Surg. 1982 Mar;195(3):259-64. 3. Vargas HI, Romero L, Chlebowski RT. Management of bloody nipple discharge. Curr Treat Options Oncol. 2002 Apr;3(2):157-61. 4. Morrogh M, Park A, Elkin EB, King TA. Lessons learned from 416 cases of nipple discharge of the breast. Am J Surg. 2010 Jan 13. Epub ahead of print. 5. Khalkhali I, Villanueva-Meyer J, Edell SL, Connolly JL, Schnitt SJ, Baum JK, et al. Diagnostic accuracy of 99mTc-sestamibi breast imaging: multicenter trial results. J Nucl Med. 2000 Dec;41(12):1973-9. 6. O’Connor M, Rhodes D, Hruska C. Molecular breast imaging. Expert Rev Anticancer Ther. 2009 Aug;9(8):1073-80. 7. Hruska CB, Boughey JC, Phillips SW, Rhodes DJ, Wahner-Roedler DL, Whaley DH, et al. Scientific Impact Recognition Award: Molecular breast imaging: a review of the Mayo Clinic experience. Am J Surg. 2008 Oct;196(4):470-6. Epub 2008 Aug 23. 8. Hruska CB, Phillips SW, Whaley DH, Rhodes DJ, O’Connor MK. Molecular breast imaging: use of a dual-head dedicated gamma camera to detect small breast tumors. AJR Am J Roentgenol. 2008 Dec;191(6):1805-15. 9. Morris EA, Libermann L. Breast MRI: diagnosis and intervention. New York (NY): Springer; c2005. p. 7. 10. Mortellaro VE, Marshall J, Harms SE, Hochwald SN, Copeland EM III, Grobmyer SR. Breast MR for the evaluation of occult nipple discharge. Am Surg. 2008 Aug;74(8):739-42. 11. Tjalma WA, Verslegers IO. Nipple discharge and the value of MR imaging. Eur J Obstet Gynecol Reprod Biol. 2004 Aug 10;115(2):234-6. 12. Orel SG, Dougherty CS, Reynolds C, Czerniecki BJ, Siegelman ES, Schnall MD. MR imaging in patients with nipple discharge: initial experience. Radiology. 2000 Jul;216(1):248-54. 13. Morrogh M, Morris EA, Liberman L, Borgen PI, King TA. The predictive value of ductography and magnetic resonance imaging in the management of nipple discharge. Ann Surg Oncol. 2007 Dec;14(12):336977. Epub 2007 Sep 26. 14. Fischer U, Kopka L, Grabbe E. Breast carcinoma: effect of preoperative contrast-enhanced MR imaging on the therapeutic approach. Radiology. 1999 Dec;213(3):881-8. 15. Rhodes DJ, O’Connor MK, Phillips SW, Smith RL, Collins DA. Molecular breast imaging: a new technique using technetium Tc 99m scintimammography to detect small tumors of the breast. Mayo Clin Proc. 2005 Jan;80(1):24-30.

It is commonly believed that men lack a rib on one side, and that men have one rib fewer than women. This is plainly absurd, even if Moses did say in the second chapter of Genesis that Eve was created by God out of Adam's rib. Granted that perhaps Adam's bones, had someone articulated them into a skeleton, might have lacked a rib on one side, it does not necessarily follow on that account that all men are lacking a rib as well. Aristotle attributed only eight ribs to humans, and was ready to allow that certain members of the race of the Turduli were born with only seven ribs on each side, provided he established this on the actual testimony of some suitable authority. But as in the latter instance Aristotle was willing to support his opinion only with the testimony of others, it is also not unlikely that in the former instance he ascribed eight ribs to man on hearsay evidence, and in this manner wrongly handed down to us something he had not seen. For if we discover that he was suppositious so many times concerning the fabric of man, what judgement shall we make about the rest of his research into animals?

Andreas Vesalius, 1514-1564

De humani corporis fabrica libri septem www.JSurgRad.com

99


CASE STUDY

J Surg Rad 100

Case Study: A 61-Year-Old Male with Hematuria and a Pancreatic Tail Mass

Case Study

Tatli and Ashley. Brigham and Women's Hospital

Servet Tatli, MD ∙ Stanley W. Ashley, MD‚  Division of Abdominal Imaging and Intervention, Department of Radiology, Brigham and Women's Hospital / Harvard Medical School, Boston, Massachusetts. ‚ Department of Surgery, Brigham and Women's Hospital / Harvard Medical School, Boston, Massachusetts.

Figure 1. Axial enhanced CT scan image shows a 1.2 cm mass (arrow) in the tail of the pancreas. Note that the mass enhances uniformly and is isodense with the spleen.

Figure 2. Axial fat suppressed, T1-weighted gradientecho MR image shows that the mass (arrow) is hypointense compared to the rest of the pancreas.

Presentation

Findings

A 61-year-old man presented with hematuria. An abdominal CT scan showed no reason for hematuria but demonstrated a 1.2 cm rounded, hypervascular mass (Figure 1) in the tail of the pancreas.

The lesion was isodense with the spleen and raised the question of ectopic splenic tissue. MRI of the pancreas confirmed a 1.2 cm, well circumscribed, pancreatic tail mass, which was seen only on pre (Figure 2) and post -contrast (Figure 3), fat suppressed T1-weighted images as homogenously hypointense compared to the normal pancreatic parenchyma with ring-like enhancement. The rest of the pancreas looked normal and there was no lymphadenopathy. Because of its well-defined border, the mass was felt unlikely to be an adenocarcinoma. Although, the patient did not have any hormonal abnormality, it was felt to most likely represent a neuroendocrine tumor. Acinar cell carcinoma and solid pseudopapillary neoplasm can also be well-circumscribed but the later most frequently is seen in young women as an encapsulated mass with hemorrhagic, necrotic areas. To more completely exclude ectopic splenic tissue in the pancreatic tail, nuclear medi-

Citation Tatli S, Ashley SW. Case Study: A 61-year-old male with hematuria and a pancreatic tail mass. J Surg Radiol. 2011 Jan 1;2(1).

Case Study | January 2011

Figure 3. Axial fat suppressed, contrast enhanced T1weighted gradient-echo MR image shows the mass (arrow) enhances uniformly but less than the normal pancreatic parenchyma with thin ring-like peripheral enhancement.

Figure 4. Axial nuclear medicine splenic imaging with Tc99m labeled, heat-denatured red blood cells shows that the mass does not accumulates the tracer; therefore, it is not splenic tissue.

cine splenic imaging with Tc-99m labeled, heat-denatured red blood cells (Figure 4) was performed and was negative. The patient underwent distal pancreatectomy and splenectomy, and pathology revealed a well-differentiated neuroendocrine tumor.

renchyma. Rarely, splenic tissue may be ectopically present in the pancreatic tail and may mimic a pancreatic neoplasm. Although this ectopic splenic tissue can be identified with MRI because of its signal intensity similar to the splenic parenchyma on all MRI sequences, a nuclear medicine study may sometimes be necessary for definite diagnosis.

Diagnosis & Management Pancreatic neuroendocrine tumors arise from the endocrine pancreas (islet cells), and may present with endocrine abnormalities because of their hormone secretion. These hormonally active tumors are named after their dominant secreted hormone, for example insulinoma or gastrinoma. Some neuroendocrine tumors may not be hormonally active (nonfunctioning) but tend to be symptomatic due to their large size. Nonfunctioning tumors are believed to all have the potential for malignancy. Malignancy of neuroendocrine tumors of the pancreas is determined not by their hormonal status but by either local invasion into adjacent organs or distant metastases, most commonly to the liver. On CT or MR imaging, neuroendocrine tumors are seen as small, well-circumscribed, hypervascular masses. They are typically hyperintense on T2-weighted images. Post-contrast images should be obtained in dynamic fashion since these tumors tend to enhance during arterial phase, which may be uniform or ring-like. Tumors can rarely be hypovascular. Unenhanced, T1-weighted, fat-suppressed MR images are also valuable to demonstrate these tumors as a hypointense mass on a background of hyperintense normal pancreatic pa-

References 1. Buetow PC, Parrino TV, Buck JL, et al. Islet cell tumors of the pancreas: pathologic-imaging correlation among size, necrosis and cysts, calcification, malignant behavior, and functional status. AJR 1995;165:1175-1179. 2. Mergo PJ, Helmberger TK, Buetow PC, Helmberger RC, Ros PR. Pancreatic neoplasms: MR imaging and pathologic correlation. RadioGraphics 1997; 17:281-301. 3. Cantisani V, Mortele KJ, Levy A, et al. MR imaging features of solid pseudopapillary tumor of the pancreas in adult and pediatric patients, AJR 2003;181:39-401. 4. Tatli S, Mortele KJ, Levy AD, Glickman JN, Ros PR, Banks PA, Silverman SG. CT and MRI features of pure acinar cell carcinoma of the pancreas in adults. AJR 2005;184(2):511-519.

www.JSurgRad.com

101


CASE STUDY

Case Study: A 61-Year-Old Male with Chest and Back Pain Following Blunt Trauma Timothy A. Plerhoples, MD, MPH ∙ John Sherck, MD ∙ John Brawley, MD ∙ Ralph S. Greco, MD Department of Surgery, Stanford University Hospital and Santa Clara Valley Medical Center, Stanford, California.

Presentation A 61-year-old man was brought into our Level I trauma center by EMS as a pedestrian struck by a vehicle. Witnesses reported the vehicle was moving at 25 mph when it impacted the pedestrian, dragging him several feet and eventually lodging him under a guardrail. After several minutes of unconsciousness, the man was extricated by EMS. He awoke, but was confused, with stable vital signs, and complained of chest and back pain.

Findings By the time he arrived to the trauma bay, he had a blood

pressure of 90/61, pulse of 85, respiratory rate of 20, and venous oxygen saturation of 97% on room air. His airway was intact, breath sounds were equal bilaterally, he had good radial and dorsalis pedis pulses, he was moving all extremities, and he had a Glasgow Coma Scale of 14 (one point deduction for confused language). A Focused Assessment with Sonography for Trauma (FAST) was negative, and physical exam was remarkable only for mid-lumbar spine tenderness with an overlying abrasion, a left elbow abrasion, and left hip tenderness. Two liters of saline were given; his blood pressure improved to 124/67. After chest and pelvis plain films were obtained and deemed negative, the decision was made to bring him to the CT scanner.

Diagnosis While well protected in the retroperitoneum, the abdominal aorta may be injured by blunt force since it is fixed in position by the vertebral column and lumbar vessels. Today’s trauma surgeons see far fewer blunt force-caused injuries to the abdominal aorta than to the thoracic aorta due to their high onscene mortality, accounting for less than five percent of all injuries to the aorta.1 A majority are seen in patients involved in motor vehicle accidents either as passengers or as drivers. The incidence of such injuries is rising, likely due to the required use of seatbelts in most states.2 Injuries may range from simple contusion to frank rupture, and include intimal disruption with dissection, intramural hematoma, and pseudoaneurysm. Direct force tends to result in laceration from associated vertebral fractures. Indirect force stems from compression between organs and the vertebral column, or deceleration causing shearing forces between free and relatively fixed segments of the aorta (more common in the thoracic aorta).

J Surg Rad

In terms of location, over 90% occur in the infrarenal aspect, with most of the remaining occurring near the insertion of the inferior mesenteric artery.1 It remains debatable whether atherosclerosis plays a role, with pathogenesis related to weakening of the intima with a loss of elasticity and compliance.3

102

Figure 1. One image from a contrast CT scan obtained after primary and secondary trauma evaluation.

Citation Plerhoples TA, Sherck J, Brawley J, Greco RS. Case Study: A 61-year-old male with chest and back pain following blunt trauma. J Surg Radiol. 2011 Jan 1;2(1).

Case Study | January 2011

Case Study

Plerhoples et al. Stanford University Hospital

forded little attention in the literature. There is more recent interest in using “less invasive” techniques especially in stable patients with concurrent head or thoracic injuries,6 although many surgeons are wary of using anticoagulation in such patients. While a low incidence limits its general use, endovascular repair options should be considered for hemodynamically-stable blunt force trauma patients with abdominal aortic injuries.

Management Our patient was taken to the operating room for angiography, which revealed the infrarenal pseudoaneurysm at the level of the third lumbar vertebral body. An endovascular pseudoaneurysm repair was done using an endograft cuff with good result. The patient remained in the intensive care unit overnight, and was ultimately discharged from the hospital four days later.

References

1. Roth SM, Wheeler JR, Gregory RT, Gayle RG, Parent FN, Demasi R, Riblet J, Weireter LJ, Britt LD. Blunt Injury of the Abdominal Aorta: A Review. The Journal of Trauma: Injury, Infection, and Critical Care. 1997;42(4):748-755. 2. Katsoulis E, Tziopis C, Sparks I, Giannoudis PV. Compressive blunt trauma of the abdomen and pelvis associated with abdominal aortic rupture. Acta Orthop. Belg. 2006;72:492-501.

Figure 2. Intra-operative angiogram obtained prior to placement of 3. Beless DJ, Muller DS, Early clinical signs and endograft cuff. Perez H: Aortoiliac occlusion symptoms of blunt injusecondary to atherosclerotic plaque rupture as the result of ry to the abdominal aorta blunt trauma. Ann Emerg Med. 1990;19:922-924. include abdominal pain, acute arterial insufficiency, acute ab4. Lock JS, Huffman AD, Johnson RC. Blunt Trauma to the Abdominal domen, weakness/paralysis/paresthesia, abdominal wall conAorta. The Journal of Trauma, 1987;27(6):674-677. tusion/defect, and back pain. Delayed signs that may suggest 5. Solovei G, Alame A, Bardoux J et al. Paraplegia and dissectionof the an injury include abnormal peripheral pulses or claudication, abdominal aorta after closed trauma. A propos of a case. Current rean abdominal mass, and an abdominal bruit.4 Approximately view of the literature (1982-1993). J Chir (Paris) 1994;131:236-244. one third of cases reported in the literature resulted from de6. Starnes BW, Arthurs ZM. Endovascular Management of Vascular Traulayed diagnosis, due either to veiled vascular symptoms or to ma. Perspect Vasc Surg Endovasc Ther. 2006;18:114. concurrent visceral lesions.5 Over the past few decades, catheter-based techniques have been increasingly used in the management of the trauma patient, including embolization of pelvic vessels or solid organs. Despite this history, the use of endovascular techniques in the management of acute traumatic vascular injuries has been afwww.JSurgRad.com

103


IMAGES

J Surg Rad 104

Images: Portal Venous Gas and Pneumatosis Intestinalis Diana L. Diesen, MD â&#x2C6;&#x2122; Dan G. Blazer III, MD Department of Surgery, Duke University Medical Center, Durham, North Carolina.

Diesen and Blazer. Duke University Medical Center An 82-year-old man with an EUS T3 N1 obstructing distal gastric cancer who was undergoing neoadjuvant chemoradiation therapy presented at a routine clinic visit complaining of dehydration, fatigue, and crampy lower abdominal pain. He reported one day of progressive malaise and abdominal pain. In clinic, he became short of breath and orthostatic. The patient was taken to the emergency room where he was noted to be more somnolent and hypotensive with diffuse abdominal pain and tenderness. IV fluids, antibiotics, and pressors were initiated. Abdominal

Images x-ray revealed gas in the portal venous system and airfluid levels in the small bowel. An abdominal pelvic CT was performed which showed diffuse portal venous gas and pneumatosis intestinalis involving the majority of the small and large bowel with nonenhancement of the bowel wall concerning for ischemia. Considering the extensive pneumatosis and portal venous gas on the CT scan, the patientâ&#x20AC;&#x2122;s advanced cancer diagnosis, and his poor performance status, the patient's family felt that comfort measures would best. The patient expired within a few hours.

Citation Diesen DL, Blazer DG. Images: Portal venous gas and pneumatosis intestinalis. J Surg Radiol. 2011 Jan 1;2(1).

Images | January 2011

www.JSurgRad.com

105


CLINICAL REVIEW

Rutherford. Transient Ischemic Attacks

Transient Ischemic Attack

ease and the risk of subsequent stroke, the great majorities of strokes occur as the initial event, emphasizing the importance of an asymptomatic carotid lesion as a potential marker for subsequent stroke risk:

Robert B. Rutherford, MD

1. They do not “most often” occur within 24 hours before the stroke.

Emeritus Professor of Surgery, University of Colorado School of Medicine, Denver, Colorado All of the following can be accepted as true examples of transient ischemic attacks except: A. Transient loss of strength in the left arm and left leg B. Transient loss of vision in the upper field of one eye C. Transient numbness in one arm D. Transient episode of dizziness and light-headedness clearing completely in one hour E. Transient episode of garbled speech Explanation: Transient ischemic attacks refer to episodes of transient loss of function in the distribution of either the carotid or vertebral artery as it affects the hemisphere, brain stem, or eye, as in a, b, d and e. Symptoms often related to transient global ischemia such as (c) dizziness and light-headedness, are most often related to transient drops in cardiac output or blood pressure and do not relate directly to arterial occlusive disease of the cerebral circulation. Territorial transient ischemic events are best explained by A. Reduced bloodflow secondary to a hemodynamically significant carotid stenosis

2. They do not occur a year before the stroke “in the majority”.

secondary embolization from irregular or ulcerated surfaces. The evidence for the repetitive nature or stereotypic presentation of transient ischemic events is based upon the concept of laminar bloodflow. Thus, when a particle is introduced from the surface of an atherosclerotic plaque in the flowing blood, particles within that particular portion of the flow stream will inevitably end up in the same distribution and terminal vessel of the intracranial circulation.

3. They characteristically occur in the same distribution as the stroke.

C. Sub plaque hemorrhage may be associated with increased carotid narrowing or disturb the overlying atheromatous debris or platelet aggregates, but associated vasospasm does not play a major role in TIAs, similar to that theorized for coronary artery disease.

Explanation: 1. The landmark publication by Eastcott, Pickering, and Robb in The Lancet in 1954 first called attention the possibility that direct repair of a carotid artery would be helpful in alleviating neurologic symptoms. However, their operation consisted of a resection and a re-anastomosis rather than endarterectomy. 2. Debakey claims to have done the first carotid endarterectomy in 1956. However, it was not published until he reported a 19-year follow-up of the “first case” in 1975.

4. A retrospective review of patients who have suffered a stroke has indicated that only 30-50% of those individuals had experienced any form of transient cerebral ischemia prior to the stroke event. This is the correct statement.

3. The Argentinean surgeons, Carrea, Molins, and Murphy, published their experience with three carotid endarterectomies in 1955. In their article they described the operations as being performed in 1953, but they withheld publication until they had at least a two-year follow-up.

The first published report of carotid endarterectomy is attributed to

4. Javid, and Julian, reported sizable early experiences with carotid endarterectomy, and Javid was responsible for the firstly specially designed shunt for maintaining flow during carotid endarterectomy, but he did not publish the first reported case.

A. Eastcott, Pickering, and Robb B. DeBakey C. Carrea, Molins, and Murphy D. Javid

D. Platelet deposition clearly occurs and platelet aggregates may embolize, causing a TIA, but hypercoagulable states have rarely been implicated in this process.

Clinical Review

E. Wiley

5. Edwin J. Wiley was a pioneer in developing endarterectomy as a technique in the USA after observing it performed in Portugal, but he did not perform or publish the first report of carotid endarterectomy.

E. Emboli of cardiac origin must be considered in someone presenting with TIA’s but they are not “territorial” in that they characteristically have the same cerebral arterial distribution.

B. Emboli of arterial origin

D. A hypercoagulable state causing platelet deposition

J Surg Rad

E. Emboli of cardiac origin

106

Explanation: A. While high-grade stenoses are normally associated with neurologic events, TIA’s are not a hemodynamic event, i.e. related to transiently flow decreases caused by a carotid stenosis. B. TIAs are secondary to the plaque burden associated with a lesion that results in fragmentation of atheromatous debris or platelet aggregation and

Citation Rutherford RB. Transient ischemic attacks. J Surg Radiol. 2011 Jan 1;2(1).

Clinical Review | January 2011

Which of the following statements regarding the timing and frequency of hemispheric transient ischemic attacks preceding a hemispheric stroke is correct? A. Antecedent TIA’s most often occur within 24 hours of the stroke event B. TIA’s occur at least one year before the stroke event in the majority C. Antecedent transient ischemic attacks occur in multiple distributions before the final stroke D. Antecedent TIAs occur in only 30-50% of those patients ultimately suffering from a stroke E. Antecedent TIAs closely mimic the subsequent stroke, except that they resolve within 48 hours Explanation: While transient ischemic attacks are an important warning of extracranial arterial occlusive dis-

Photos.com

C. Transient vasospasm following hemorrhage under a plaque

www.JSurgRad.com

107


CLINICAL REVIEW

May-Thurner Syndrome

A. Compression of the left common iliac vein by the right common iliac artery

Assistant Professor of Radiology, Duke University Medical Center, Durham, North Carolina

B. Development of focal stenosis related to focal intimal hyperplasia

Which of the following describes a classic clinical presentation for a patient with May-Thurner syndrome? B. 79 year old female with first-time presentation of acute right leg swelling. C. 91 year old female with bilateral ultrasound-proven lower extremity DVT D. 35 year old female with painful and marked left leg swelling, with a history of mild episodic but selfresolving left leg swelling E. 17 year old male with left knee pain and knee swelling Explanation: A. Although the structural venous compression is assumed to be present long before presentation with acute thrombosis, the development of intimal hyperplasia-related stenosis takes years to develop. The most common age range at presentation is the third to fifth decades of life. B. May-Thurner syndrome is compression of the left common iliac vein by the right common iliac artery; therefore, a patient with right leg swelling is unlikely to have this syndrome. C. A patient with bilateral lower extremity DVT is also unlikely to have May-Thurner syndrome. Such a patient is much more likely to have DVT due to an alternate etiology, such as hypercoaguable state. D. This patientâ&#x20AC;&#x2122;s age falls into the most common age range, and accurately describes a common presentation.

J Surg Rad

May-Thurner syndrome is characterized by all of the following except:

Charles Y. Kim, MD

A. Newborn with marked enlargement of the left leg

108

Kim. May-Thurner Syndrome

E. May-Thurner syndrome is at least twice more common in females than males. However, more importantly, the clinical symptoms in this patient are typically that of proximal left leg DVT, and this presentation suggests a musculoskeletal etiology.

Citation Kim CY. May-Thurner Syndrome. J Surg Radiol. 2011 Jan 1;2(1).

Clinical Review | January 2011

Optimal treatment of May-Thurner syndrome is: A. Long-term anticoagulation with warfarin B. Thrombolysis and angioplasty C. Thrombolysis, angioplasty, and stenting followed by long-term anticoagulation D. Palmaâ&#x20AC;&#x2122;s crossover procedure E. Implantation of a PTFE bypass graft from the left common femoral vein to IVC Explanation: A. Anticoagulation without any additional intervention results in persistent and recurrent symptoms, because the underlying structural etiology is not addressed. B. Thrombolysis followed by angioplasty results in a dismal one year patency rate. This is due to early recurrence of the stenosis, at least partially secondary to the extrinsic compression by the right common iliac artery. C. The ideal management in the majority of cases involves thrombolysis followed by angioplasty and stenting of the left common iliac vein origin, with subsequent anticoagulation. Stenting likely prolongs patency by minimizing the effect of the extrinsic compression. The long term patency rate is excellent, with a 10-year patency rate reported at 93%. D. The Palma crossover procedure involves transecting the right greater saphenous vein and creating an anastomosis with the left common femoral vein to bypass the left iliac venous system, sometimes with creation of a temporary AV fistula. This alternative surgical option provides a good long term patency, reported as 83% at four years. E. Implantation of a PTFE bypass graft is associated with poor patency rates, reported as <50% at two years.

C. Collateral vein formation D. Atherosclerotic disease E. Deep vein thrombosis

Clinical Review initial imaging modality followed by CT, MR, or conventional venography. In all of these modalities, acute and/or sequelae of prior DVT are typically present. In the setting of acute DVT, the left lower extremity veins are typically thrombosed to the origin of the left common iliac vein. In patients who have just undergone thrombolysis, or those imaged prior to acute thrombosis, a stenosis of the origin of the left common iliac vein is classically visualized on CTV, MRV, and conventional venography, as are multiple collateral veins in the pelvis and to the IVC.

Explanation: May-Thurner syndrome is characterized primarily by compression of the left common iliac vein by the overlying right common iliac artery. Over time, the chronic vibratory pulsation incites an intimal hyperplasia response in the compressed segment of vein, with resulting progressive stenosis. Because of the slow and chronic nature of this process, as the stenosis becomes hemodynamically significant, collateral veins to the contralateral iliac veins and to the IVC enlarge over time. When the flow is slow enough, spontaneous thrombosis can occur, particularly when the patient is subject to other risk factors for DVT, such as a hypercoaguable state. Patients may have had prior episodes of clinically apparent or subclinical DVT. MayThurner syndrome is a pathologic state of the venous system, and so atherosclerosis is not associated. Which of the following imaging modalities can be used for evaluation of potential MayThurner syndrome? A. Ultrasound B. CT venography C. MR venography D. Conventional venography E. All of the above Explanation: Imaging for May-Thurner syndrome can be performed with ultrasound, MR venography(MRV), CT venography(CTV), and conventional venography. Frequently, ultrasound is the www.JSurgRad.com

109


FUTURE DIRECTIONS

J Surg Rad 110

As 3D printing

and materials technology improve, customized real-time construction of endovascular grafts tailored to a patient's particular anatomy will provide superior results and virtually eliminate most types of endoleak. Sapan S. Desai, MD, PhD Executive Editor

Future Directions | January 2011


Journal of Surgical Radiology January 2011  

Volume 2 Issue 1

Advertisement