Venous Thromboembolic Disease

LD2 VTD Nearly Final Pages.
indd 8 5/13/11 10:27 AM

Venous Thromboembolic Disease
LD2 VTD Nearly Final Pages.indd 1 5/13/11 10:27 AM

CME Co n T e M P o r A ry en D oVASCu l A r M A n Ag e M en T
edited by Mark g. Davies and Alan B. lumsden
The Contemporary Endovascular Management series is designed to be a focused,

relevant, and timely review of the modern aspects of imaging and intervention in
specific vascular beds. Each volume in the series addresses one vascular bed and
equips readers with the current information necessary for vascular practice using
a clear and easy-to-access format.
Volume 1 Chronic Venous Insufficiency
Volume 2 Venous Thromboembolic Disease
Volume 3 Dialysis Access: Creation and Maintenance

Forthcoming
Volume 4 Dialysis Access: Special Cases and Complications

Forthcoming
Please visit www.cardiotextpublishing.com for more information about this series.

Venous
Thromboembolic
Disease
Co n T e M P o r A ry e n D oVA S Cu l A r M A n Ag e M e n T
Volume 2
editors
Mark G. Davies MD, PhD, MBA

Vice Chairman, Finance and Administration, Department of Cardiovascular Surgery; Program
Director, Vascular Surgery Fellowship and Integrated Vascular Surgery Residency; Director of
Research and Education, Methodist DeBakey Heart & Vascular Center, The Methodist Hospi-
tal, Houston, Texas; Senior Investigator, The Methodist Hospital Research Institute; Professor of
Cardiovascular Surgery, Weill Cornell Medical College, New York, New York
Alan B. Lumsden MD
Chairman, Department of Cardiovascular Surgery; Medical Director, Methodist DeBakey
Heart & Vascular Center, The Methodist Hospital, Houston, Texas; Professor of Cardiovascular
Surgery, Weill Cornell Medical College, New York, New York
Daynene Vykoukal PhD assistant editor

Scientific Editor, The Methodist Hospital Research Institute, Methodist DeBakey Heart & Vascular Center,
The Methodist Hospital, Houston, Texas
Minneapolis, Minnesota

2011 Mark G. Davies and Alan B. Lumsden
Cardiotext Publishing, LLC
3405 W. 44th Street
Minneapolis, Minnesota 55410
USA
www.cardiotextpublishing.com
Any updates to this book may be found at:

www.cardiotextpublishing.com/titles/detail/9781935395225
Comments, inquiries, and requests for bulk sales can be directed to the publisher at:
info@cardiotextpublishing.com.
All rights reserved. No part of this book may be reproduced in any form or by any means without the prior
permission of the publisher.
All trademarks, service marks, and trade names used herein are the property of their respective owners and
are used only to identify the products or services of those owners.
This book is intended for educational purposes and to further general scientific and medical knowledge,
research, and understanding of the conditions and associated treatments discussed herein. This book is not
intended to serve as and should not be relied upon as recommending or promoting any specific diagnosis
or method of treatment for a particular condition or a particular patient. It is the readers responsibility to
determine the proper steps for diagnosis and the proper course of treatment for any condition or patient,
including suitable and appropriate tests, medications, or medical devices to be used for or in conjunction
with any diagnosis or treatment.
Due to ongoing research, discoveries, modifications to medicines, equipment and devices, and changes
in government regulations, the information contained in this book may not reflect the latest standards,
developments, guidelines, regulations, products, or devices in the field. Readers are responsible for keeping
up-to-date with the latest developments and are urged to review the latest instructions and warnings for any
medicine, equipment, or medical device. Readers should consult with a specialist or contact the vendor of
any medicine or medical device where appropriate.
Except for the publishers website associated with this work, the publisher is not affiliated with and does not
sponsor or endorse any websites, organizations, or other sources of information referred to herein.
The publisher and the author specifically disclaim any damage, liability, or loss incurred, directly or
indirectly, from the use or application of any of the contents of this book.
Unless otherwise stated, all figures and tables in this book are used courtesy of the authors.
Cover design by Brad Norr Design

Book design by Ann Delgehausen, Trio Bookworks
Library of Congress Control Number: 2011924295
ISBN: 978-1-935395-22-5
Printed in Canada
16 15 14 13 12 11 1 2 3 4 5 6 7 8 9 10

Contents
Contributors ix
Preface xiii
Anthony J. Comerota and Thomas W. Wakefield
Acknowledgments xv
Introduction 1
Mark G. Davies
Part 1: Venous Thromboembolic Disease

Overview of Part 1 3
1. Overview of Therapy for Venous Thromboembolic Disease 5

Eduardo Ramacciotti and Thomas W. Wakefield
2. Deep Venous Thrombosis and Pulmonary Embolism Pathophysiology 19
Peter K. Henke
3. Deep Venous Thrombosis: Prevention and Treatment 33
Mark G. Davies
4. Diagnostic Approach to Venous Thromboembolism 43
L. Bernardo Menajovsky, Patricia Hightower Lambden, and Ruth L. Bush

vi Contents
5. Treatment of Acute Deep Venous Thrombosis:

Whats New in the ACCP Guidelines? 57
Anthony J. Comerota
6. Anticoagulation Therapy 67
Ruth L. Bush, Brandi Huf, and CleAnn Toner
7. Pharmacological Thrombolysis: Indications, Techniques, and Outcomes 79
Mark G. Davies
8. Pharmacomechanical Thrombolysis for Acute Deep Venous Thrombosis:
Indications, Techniques, and Clinical Data 89
Elina Quiroga and Mark H. Meissner
9. Endovascular Intervention for Lower Extremity Deep Venous Thrombosis:
Techniques, Devices, and Outcomes 95
Erin H. Murphy and Frank R. Arko III
10. Surgical Thrombectomy: Indications, Techniques, and Outcomes 103
Bo Eklf
11. IVC Filters: Indications, Techniques, and Outcomes 117
Joseph P. Hart and Claudio J. Schnholz
12. Percutaneous Pulmonary Embolectomy:
Indications, Techniques, and Outcomes 135
Saher Sabri, Wael E. A. Saad, and Alan H. Matsumoto
Part 2: Thoracic Outlet Syndrome

13. Primary Axillosubclavian Venous Thrombosis: Observational Care 155

Kaj H. Johansen
14. Axillary-Subclavian Venous Effort Thrombosis: Surgical Care 163
Valerie B. Emery and Robert W. Thompson
Part 3: Central Venous Disease

15. Iliofemoral Deep Venous Thrombosis 181

Mark G. Davies

Contents vii
16. Management of Central Venous Stenosis 193

Michael J. Reardon and Mark G. Davies
17. Mesenteric Venous Thrombosis 205
Hosam F. El Sayed
Part 4: Practice Management

18. Current Venous Coding and Billing 225

Gary Burns
Index 235

Contributors
editors
Mark g. Davies MD, PhD, MBA Vice Chairman, Finance and Administration, Department
of Cardiovascular Surgery; Program Director, Vascular Surgery Fellowship and
Integrated Vascular Surgery Residency; Director of Research and Education,
Methodist DeBakey Heart & Vascular Center, The Methodist Hospital, Houston,
Texas; Senior Investigator, The Methodist Hospital Research Institute; Professor of
Cardiovascular Surgery, Weill Cornell Medical College, New York, New York
Alan B. Lumsden MD Chairman, Department of Cardiovascular Surgery; Medical Director,

Methodist DeBakey Heart & Vascular Center, The Methodist Hospital, Houston,
Texas; Professor of Cardiovascular Surgery, Weill Cornell Medical College,
New York, New York
Contributors
Frank R. Arko III MD Associate Professor, Chief of Endovascular Surgery, Division of
Vascular and Endovascular Surgery, Department of Surgery, University of Texas
Southwestern Medical Center, Dallas, Texas
Gary Burns MBA, RHIA, CIRCC Principal, Medical Asset Management, Inc., Atlanta,
Georgia
Ruth L. Bush MD, MPH Associate Professor of Surgery, Vascular Surgery, Scott & White
Healthcare; Associate Dean for Education, Temple Campus, Texas A&M Health
Science Center College of Medicine, Temple, Texas
ix

x Contributors
Anthony J. Comerota MD, FACS, FACC Director, Jobst Vascular Institute, Toledo, Ohio;
Adjunct Professor of Surgery, University of Michigan, Ann Arbor, Michigan
Bo Eklf MD, PhD Emeritus Professor, Department of Surgery, University of Hawaii; Clinical
Emeritus Professor, Lund University, Sweden
Hosam F. El Sayed MD, RVT Department of Cardiovascular Surgery, Methodist DeBakey

Heart & Vascular Center, The Methodist Hospital, Houston, Texas; Assistant Professor
of Cardiovascular Surgery, Weill Cornell Medical College, New York, New York
Valerie B. Emery RN, ANP, BC Department of Surgery, Section of Vascular Surgery, Center
for Thoracic Outlet Syndrome, Washington University School of Medicine and Barnes-
Jewish Hospital, St. Louis, Missouri
Joseph P. Hart MD, RVT, FACS Assistant Professor of Surgery and Interventional Radiology;
Chief of Endovascular Surgery, Division of Vascular Surgery, Department of Surgery,
Medical University of South Carolina, Charleston, South Carolina
Peter K. Henke MD Leland Ira Doan Professor of Surgery, Section of Vascular Surgery,
University of Michigan, Ann Arbor, Michigan
Brandi Huf PharmD Scott and White Anticoagulation Clinic, Scott and White Memorial
Hospital and Clinics, Temple, Texas
Kaj H. Johansen MD, PhD, FACS Swedish Heart & Vascular Institute, Swedish Medical
Center, Seattle, Washington
Patricia Hightower Lambden MSN, CNS-BC, APN McLennan Community College, Waco,
Texas
Alan H. Matsumoto MD, FSIR, FACR, FAHA Professor and Chair, Department of Radiology
and Medical Imaging, University of Virginia Health System, Charlottesville, Virginia
Mark H. Meissner MD Professor of Surgery, University of Washington School of

Medicine, Seattle, Washington
L. Bernardo Menajovsky MD, MS Associate Professor of Internal Medicine, Scott & White
Healthcare, Temple, Texas
Erin H. Murphy Division of Vascular and Endovascular Surgery, Department of

MD
Surgery, University of Texas Southwestern (UTSW) Medical Center, Dallas, Texas

Contributors xi
Elina Quiroga MD Division of Vascular Surgery, University of Washington, Seattle,

Washington
Eduardo Ramacciotti MD, PhD Section of Vascular Surgery, Department of Surgery,

University of Michigan, Ann Arbor, Michigan
Michael J. Reardon MD Department of Cardiovascular Surgery, Methodist DeBakey

Heart & Vascular Center, The Methodist Hospital, Houston, Texas; Professor of
Cardiothoracic Surgery, Weill Cornell Medical College, New York, New York
Wael E. A. Saad MBBCh, FSIR Associate Professor of Radiology, Division of Vascular and
Interventional Radiology, University of Virginia Health System, Charlottesville, Virginia
Saher Sabri MD Assistant Professor of Radiology, Division of Vascular and Interventional

Radiology, University of Virginia Health System, Charlottesville, Virginia
Claudio J. Schnholz MD Heart & Vascular Center, Interventional Radiology, Medical

University of South Carolina, Charleston, South Carolina
Robert W. Thompson MD Professor of Surgery, Vascular Surgery, Radiology and Cell

Biology and Physiology, Center for Thoracic Outlet Syndrome, Section of Vascular
Surgery, Department of Surgery, Washington University School of Medicine and
Barnes-Jewish Hospital, St. Louis, Missouri
CleAnn Toner PharmD Scott and White Anticoagulation Clinic, Scott and White Memorial
Hospital and Clinics, Temple, Texas
Thomas W. Wakefield MD S. Martin Lindenauer Professor of Surgery, Section Head of

Vascular Surgery, Department of Surgery, University of Michigan Health System,
Cardiovascular Center, Ann Arbor, Michigan
LD2 VTD Final Pages.indd 11 5/19/11 12:17 PM

Preface
Anthony J. Comerota and Thomas W. Wakefield
Venous thromboembolism (VTE) includes deep venous thrombosis (DVT) and pulmonary
embolism (PE). VTE is a national health concern, with mortality exceeding that of acute
myocardial infarction and acute stroke. VTE affects over 1,000,000 patients per year, with
over 300,000 deaths per year in the United States. VTE fatalities have remained constant over
the past 30 years, exceeding breast cancer and AIDS combined. The incidence of DVT has
been increasing with the aging of the population. In those 85-89 years old, the incidence is
reported to be as high as 310 people per 100,000 in the population. Additionally, treatment
costs are in billions of dollars per year. The local consequence of DVT, termed postthrombotic
syndrome (PTS), affects between 400,000 and 500,000 patients annually with pain, edema,
pigmentation, and skin ulcerations. It has been reported that after having an iliofemoral DVT
treated with anticoagulation alone, 95% of patients have chronic venous insufficiency, 40%
have venous claudication and nearly all have reduced quality-of-life. Even asymptomatic DVT
has been associated with PTS.
Although Virchows triad of stasis, vessel wall injury, and hypercoagulability has defined
the events that predispose individuals to DVT formation for the past 150 years, today the un-
derstanding of events that occur at the level of the vein wall, including the influence of the
inflammatory response on thrombogenesis, is increasingly becoming recognized, although we
still have much to learn.
The critical nature of venous thromboembolism to our nation and the world has been
recognized, and a Surgeon Generals conference held in Bethesda, Maryland, in May 2006
in conjunction with the National Institutes of Health led to a call to action against VTE in
2008. The fact that death due to pulmonary embolism remains the most common cause of
in-hospital death today places this problem in stark perspective.
xiii

xiv Preface
With the current emphasis that has been placed on VTE, the production of this excellent
book which addresses all aspects of venous thromboembolic disease is very timely. In part 1,
many different aspects of venous thromboembolic disease are covered, ranging from patho-
physiology of DVT and PE to diagnosis and treatment of those conditions. Treatment involves
not only pharmacologic agents, but also mechanical and pharmacomechanical approaches.
Operative thrombectomy remains a good option for treating patients with iliofemoral DVT,
and the well-designed Scandanavian trial illustrates the value of a strategy of thrombus remov-
al in patients with iliofemoral DVT. However, VTE is not limited to the lower extremities. In
part 2, upper extremity venous thrombosis related to thoracic outlet (thoracic inlet) syndrome
is discussed. Part 3 involves issues of central venous disease with a major emphasis on dialysis
access. The lifeline of patients with end-stage renal disease is a functional dialysis access. It is
evident that the majority of successful accesses are dependent on good-quality venous outflow.
Part 4 discusses practice management of venous disease.
Venous Thromboembolic Disease, volume 2 of Contemporary Endovascular Manage-
ment, has contributions by leaders in the field, with evidence-based recommendations. We
highly recommend this volume to those who are serious students of VTE and who want to
update themselves on the current status of VTE pathophysiology, diagnosis, and treatment.

Acknowledgments
The editors would like to thank the following individuals for their kind contributions to the
completion of this book: Yvette Whittier and Daynene Vykoukal, PhD, at The Methodist Hos-
pital for organizational assistance, author relations, primary editing, and manuscript assembly;
Steve Korn, Mike Crouchet, Caitlin Crouchet, and Carol Syverson at Cardiotext Publishing
for their roles in series development, content management, and technical support; and Ann
Delgehausen, Beth Wright, and Zan Ceeley at Trio Bookworks for their diligent execution of
the publication and their contribution to the series designs and concepts. Without the time
and effort of the contributors, no book can address its goals and we gratefully acknowledge the
time and effort that each of the authors devoted to this project.
Mark G. Davies
Alan B. Lumsden
xv

Introduction
Mark g. Davies
A cute venous disorders comprise the

spectrum of deep venous thrombosis
(DVT), superficial venous thrombophle-
zation occurring over the first 6 weeks after
thrombosis. Pulmonary embolism (PE) and
the postthrombotic syndrome (PTS) are the
bitis, and venous trauma.1 This volume in most important acute and chronic complica-
the CEM series will provide a focused and tions of DVT. Due to the known risk profiles
timely review of the current state of the field and ease of pharmacological intervention,
in acute venous diseases. thromboembolism prophylaxis is standard
Deep venous thrombosis has a variable of care, but appropriate measures are uti-
estimated incidence of 56 to 160 cases per lized in as few as one-third of at-risk patients.
100,000 population per year and appears to DVT may present as fever or tachycardia, as
be linked with the convergence of multiple leg swelling or pain, or as cardiopulmonary
genetic and acquired risk factors. It is a signif- events secondary to acute pulmonary embo-
icant problem in hospitalized patients. Acute lism. Duplex imaging and D-dimer testing
venous thrombosis is followed by an inflam- remain integral to DVT imaging and diag-
matory response in the thrombus and vein nosis. Ventilation-perfusion (V/Q) scans and
wall, leading to thrombus amplification, or- computed tomography angiography (CTA)
ganization, and recanalization, which results of the chest are the most common modali-
in structural defects within the wall and in ties employed to diagnose pulmonary em-
turn leads to luminal compromise and val- bolism. Once diagnosed, anticoagulation
vular dysfunction. Clinically, there is an ex- constitutes the mainstay of management for
ponential clearance of the thrombus by the a DVT, with the goal of preventing recurrent
bodys inflammatory and fibrinolytic systems venous thromboembolism (VTE). However,
over the first 6 months, with most recanali- anticoagulation is not a protection against

2 Introduction
PTS because the valves and vessel wall are system vein, conservative measures are ade-
damaged as the thrombus is cleared by the quate. For those lesions that encroach on a
body. Most recent guidelines recommend deep vein, recent investigations suggest that
catheter-directed thrombolysis or combined anticoagulation may be more effective than
pharmacomechanical thrombectomy for ligation in preventing DVT and PE. Venous
proximal DVT in the mobile and functional injuries are similarly underreported, and
patient with no absolute contraindications. current recommendations include repair of
Effective lysis of these DVTs can reduce or bypass of injuries to the major proximal
postthrombotic symptoms and improve veins. If repair is not safe or possible, ligation
quality of life after acute iliofemoral DVT. should be performed.
Inferior vena cava filters continue to have
a role among patients with contraindica-
tions to, complications of, or failure of an- reference
ticoagulation. The incidence of superficial 1. Meissner MH, Wakefield TW, Ascher E,
venous thrombophlebitis is underreported Caprini JA, Comerota AJ, Eklof B, et al. Acute
and is considered to occur in approximate- venous disease: venous thrombosis and venous
ly 125,000 patients per year in the United trauma. J Vasc Surg. 2007;46 (suppl) S:
States. For lesions not encroaching on a deep 25S-53S.

1
PA R T
Venous
Thromboembolic
Disease
T he occurrence of deep ve-
nous thrombosis (DVT) and pulmonary embolism continues be a
significant problem in clinical practice. guidelines for prevention of
DVT have been updated by the American College of Chest Physi-
cians (ACCP) and DVT prophylaxis remains a high-priority qual-
ity measure for hospitals and practitioners. Prevention of DVT and
pulmonary embolism can save lives and decrease morbidity. As
we refine our knowledge of the biology and progression of DVT,
we will be able to develop better anticoagulants and alternative
strategies to control DVT propagation. While the efforts to prevent
DVT continue, the ability to clear existing DVTs has become more
refined. While use of pharmacological lysis is a mainstay, introduc-
tion of mechanical declotting and ultrasound generating devices
and combined pharmacomechanical techniques have accelerated
treatment strategies and decreased lysis-related complications. In
certain patients with limb-threatening DVT, open surgical embolec-
tomy remains a viable and necessary option in the carefully selected
patient. There has been a surge in the prophylactic placement of

inferior vena cava filters, without any significant rise in the removal
of retrievable filters, and the basic designs have not changed much
since their introduction. Finally, a more aggressive stance has been
taken when pulmonary embolism is detected and both lytic and
mechanical devices are now available to facilitate a reduction in
the cardiac and pulmonary sequelae of major pulmonary emboli.
The evolution of the management of acute DVT and pulmonary
embolism is well illustrated in this section.

CHAPTER
overview of Therapy
for Venous Thromboembolic Disease
eduardo ramacciotti and Thomas W. Wakefield
V enous thomboembolism (VTE), a

disease that includes deep venous
thrombosis (DVT) and pulmonary embo-
The ideal treatment should relieve
the edema and pain, reduce or remove the
thrombus, and prevent death by pulmonary
lism (PE), is serious and frequent. Nine hun- embolism in the early stages of treatment.
dred thousand cases of VTE are estimated In the long term, treatment should prevent
to occur per year in the United States, with recurrence of DVT as well as avoid post-
300,000 deaths every year from PE.1 Addi- thrombotic syndrome and chronic pulmo-
tionally, the development of long-term com- nary hypertension secondary to PE. In the
plications is common, despite appropriate acute phase, treatment is targeted at:
acute phase treatment. Following the acute
thrombotic process, damage to the vein Arresting thrombus growth, thereby
wall and valves leads to a chronic condition preventing recurrent thrombotic
called postthrombotic syndrome (character- disease in the chronic phase
ized by pain and leg swelling, with eventual Preventing embolization, thereby
formation of distal ulcers). The incidence of preventing PE and secondary
postthrombotic syndrome is as high as 30% hypertension in the chronic phase
over 8 years.2 Dissolving or removing the clot,
thereby preventing venous dysfunction
and chronic venous insufficiency
Venous Thromboembolic Disease. Contemporary
Endovascular Management series. 2011 Mark G.
Limiting progressive swelling of the
Davies md and Alan B. Lumsden md, eds. leg, thereby preventing an increase in
Cardiotext Publishing, ISBN 978-1-935395-22-5. compartmental pressure

6 Venous Thromboembolic Disease
Once the diagnosis is confirmed, son with UFH, the small molecular weight
immediate systemic anticoagulation should of LMWH means that it has a more predict-
be started, unless contraindicated. A thera- able anticoagulant effect, can be given sub-
peutic dose is required, due to a 4- to 6-fold cutaneously, causes less platelet activation
risk of recurrence with insufficient antico- and bleeding, and is associated with a lower
agulation in the first 24 hours.3 Adequate rate of heparin-induced thrombocytopenia.7
anticoagulation has been shown to prevent In addition, LMWH may be administered
the development of fatal PE both during the subcutaneously in a weight-based manner
initial treatment and after treatment is com- (q.d. or b.i.d.), and in many instances, may
plete.4 However, the recurrence rates are still be administered in the outpatient setting.
high. It is expected that one-third of properly These compounds do not require monitor-
treated patients will present recurrent DVT ing except in certain circumstances such
after an 8-year follow-up period.5 as renal failure and morbid obesity, and
Historically, intravenous unfraction- during pregnancy.8 However, its use in the
ated heparin (UFH) was the initial standard outpatient settings usually requires a team
therapy. Due to large individual variation in approach and a coordinated effort of mul-
heparin effect response, frequent adjustment tiple healthcare providers. There is also lim-
of the dose is required and the treatment is ited evidence that LMWH may decrease the
maintained for an average of 5 days. Patients incidence of postthrombotic syndrome.
usually have to be hospitalized to undergo There are different LMWHs available
the frequent monitoring and dose adjust- for the treatment of VTE. Since they are bio-
ments necessary to maintain therapeutic logic preparations, they are not interchange-
plasma concentrations of UFH. Monitor- able. Moreover, the regimen and doses for
ing of activated partial thromboplastin time the compounds differ, as listed in Table 1.1.
(aPTT) is crucial during the initial phase of
treatment. Patients who fail to achieve thera-
peutic levels of heparin within 24 hours of
starting treatment have a 15 times higher Prevention of
risk of recurrence than patients who achieve
heparin concentrations within the target Recurrent Venous
therapeutic range, aPPT 1.5 to 2.5 x control.6
In this period, treatment with oral vitamin
Thromboembolism
K antagonist (VKA) is instituted (usually Warfarin (and other VKAs) remains the only
warfarin, 5 mg/day) and an international available oral anticoagulant proven to reduce
normalized ratio (INR) therapeutic (usually the risk of recurrent VTE. Several studies
above 2) is required before stopping hepa- have established that for the vast majority
rin infusion. The usual minimum time for of patients, anticoagulation results in an
heparin initial therapy is 5 days. annual risk of recurrence of less than 2%.9,10
Recently, the treatment of VTE has Because warfarin has a narrow therapeutic
been markedly simplified by the introduc- index, monitoring the INR is necessary. A
tion of low-molecular-weight heparins target INR value of 2 to 3 is appropriate for
(LMWHs). LMWHs are derived from stan- patients with VTE.
dard heparin after exposure to different Warfarin should be started only after
chemical processes. However, in compari- heparinization is therapeutic to prevent war-

Overview of Therapy for Venous Thromboembolic Disease 7
Table 1.1. FDA-ApprovedAnticoagulantAgentsforTreatmentofVTE

Routeof
Agent Administration Dose NeedforMonitoring
UFH IV 80 U/kg initial bolus, followed Adjusted to achieve aPTT 1.5

by 18 U/kg/h maintenance 2.5 x control
UFH SQ 333 IU/kg SQ initially, No. However, reserved

followed by as second-line therapy
250 IU/kg SQ
LMWH
Enoxaparin SQ 1.0 mg/kg b.i.d. or No
1.5 mg/g q.d.
Dalteparin SQ 100 mg/kg b.i.d. or No

200 mg/kg q.d.
Tinzaparin SQ 175 IU/kg q.d. No

PENTASACCHARIDE
Fondaparinux SQ 7.5 mg q.d. No

(510 mg q.d. based on
weight)
UFH = unfractionated heparin IV, LMWH= low-molecular-weight heparin. IV= intravenous,SQ= subcutaneous,
aPTT= activated partial thromboplastin time. Taking all of the evidence together, LMWH is now preferred over
standard UFH for the initial treatment of VTE with a level of evidence 1A.
Source: van Dongen CJ, van den Belt AG, Prins MH, Lensing AW. Fixed dose subcutaneous low molecular
weight heparins vs. adjusted dose unfractionated heparin for venous thromboembolism. Cochrane Db Syst Rev.
2004(4):CD001100.
farin-induced skin necrosis, usually on the with shorter administration (3 months) of

first day of therapy. This condition occurs VKA.12 Calf-level thrombi may be treated for
due to transient hypercoagulability, which shorter periods, ranging from 6 to 12 weeks
occurs for the first few days after warfarin is of warfarin. Patients with a second episode
administered. Warfarin causes inhibition of of VTE require prolonged warfarin therapy,
protein C and protein S before most coagu- unless contraindicated, due to high rates of
lation factors are inhibited by warfarin, lead- recurrence.
ing to a transient prothrombotic state. The warfarin therapy length in other
After a first episode of VTE, the rec- clinical situations is controversial. Inherited
ommended duration of anticoagulation is or acquired hypercoagulable states, such as
3 to 6 months.11 For patients with VTE that the presence of homozygous factor V Leiden
occurred in association with a well-defined, (FVL) and prothrombin G20210A muta-
time-limited risk factor (eg, major surgery, tions (FII), protein C/S or antithrombin
pregnancy, trauma), the future risk of recur- III deficiency, antiphospholipid syndrome,
rent thrombosis appears to be lower than 5%, and active cancer, significantly increase the

risk of VTE recurrence, and prolonged oral at an increased risk for VTE recurrence.17
anticoagulation is warranted. However, first Moreover, another recent study has demon-
episodes of VTE in patients with heterozy- strated a statistically significant advantage to
gous mutations of FVL/ FII do not carry the resuming Coumadin if the D-dimer assay
same high risk as those in their homozygous is positive compared to remaining off Cou-
counterparts, shortening the required length madin during an average 1.4-year follow-up
of oral anticoagulation. period (odds ratio [OR] 4.26, P = 0.02).18
Another particularly controversial treat- One other factor that impacts the recurrence
ment situation involves patients experienc- rates is the quality of anticoagulation. One
ing unprovoked VTE. These subjects are trial has demonstrated that patients exposed
at substantial risk for recurrent disease after to an INR <1.5 during the first 3 months of
warfarin discontinuation. During 3 years treatment presented higher risk for recurrent
following warfarin discontinuation, approxi- VTE.19
mately 20% of such patients will suffer recur- Decisions about the duration of antico-
rent VTE, particularly on the same vascular agulant therapy are further complicated by
site as the index event.13 The actual length the need to individualize the hazards asso-
of therapy and dose of warfarin best suited ciated with anticoagulation by weighing the
for this particular group are not known. The risks vs. the benefits of anticoagulation. After
PREVENT multicenter trial suggested that, estimating both the risk of thrombosis with-
for idiopathic DVT, low-dose warfarin (INR out warfarin as well as the risk of bleeding
1.52.0) is superior to placebo over a 4-year with treatment, the clinician should con-
follow-up period with a 64% risk reduction sider factors such as adherence history, risk
for recurrent DVT after the completion of of falling, and underlying diseases before
an initial 6 months of standard warfarin making a recommendation about when or
therapy.14 However, a second study has sug- if to discontinue anticoagulation.20 In addi-
gested that full-dose warfarin (INR 23) is tion, if the anticoagulation is not discon-
superior to low-dose warfarin in these same tinued, decreasing risk of bleeding dose
patients, without a difference in bleeding.15 adjustments and using anticoagulation clin-
Taken together, this data suggest that long- ics are emphasized.
term therapy is required for unprovoked Thus, the decisions for discontinua-
VTE patients, with INR targeting 2 to 3. tion of oral anticoagulation should include
When estimating the risk of recurrent thrombosis risk assessment, evaluation of
disease for an individual patient, certain residual thrombus burden, and coagulation
factors have been linked to a higher likeli- system activation (as suggested by D-dimer
hood of future events: an unprovoked ini- measurements). These criteria are given a
tial clot, and if the index event is PE, rather level of evidence of 1A.17-19
than DVT.16 Recently, one additional crite-
rion has been used to determine the length
of anticoagulation. This validated criterion Complications of
involves D-dimer testing obtained 1 month
after warfarin is completed. If the D-dimer
Anticoagulant Therapy
level is elevated above normal, warfarin The most common complication of antico-
should be continued, as this result suggests agulation is bleeding. Increasing the anti-
that the patient is still prothrombotic and coagulant effect of a pharmacological agent

increases the risk of bleeding. By decreasing mediated reaction, and it seems to be safe as
the anticoagulant effect of the compound, a substitute anticoagulant for HIT patients.22
the risk of bleeding decreases, but the risk The use of these alternative agents is given a
of recurrence or even treatment failure 2C and 1C level of evidence.
increases. For UFH, the risk of bleeding dur-
ing the initial 5 days is around 10%. With
the addition of warfarin at an INR between 2
and 3, the annual bleeding risk is 6%. Prolonged
Another complication associated with
heparin use is heparin-induced thrombo-
Anticoagulation
cytopenia (HIT). This immune-mediated in Cancer Patients
disorder occurs in 0.6% to 30.0% of patients
exposed to heparin. Heparin-dependent The safety of LMWH compared with that
antibodies mediated by platelet factor 4 of warfarin has led to a consideration of
(PF4) bind to platelets, activating them. This the long-term use of LMWH as a replace-
activation leads initially to platelet aggrega- ment for oral vitamin K antagonists. Rates
tion and release of thrombogenic micropar- of recanalization have been reported to be
ticles, which can produce thrombosis, both higher in certain venous segments using
arterial and venous. Thrombocytopenia due LMWH vs. traditional oral agents.
to consumption occurs, increasing the risk of Cancer patients are reported to have
bleeding. Both UFH and LMWH have been longer periods out of the therapeutic INR
associated with HIT, although the incidence range compared to patients free of cancer.
is lower with LMWH. Clinically, the patient Additionally, the use of LMWH has been
presents a drop on platelet count greater found to result in improved outcomes in can-
than 50% of baseline, or below 100,000/L cer patients compared with standard heparin
during heparin therapy. Eventually, throm- or LMWH/warfarin therapy when used for 6
bosis occurs.21 The test of choice for diag- months, without differences in major bleed-
nosis is an enzyme-linked immunosorbent ing.23 The use of LMWH in selected can-
assay (ELISA) that detects the antiheparin cer patients for prolonged anticoagulation is
antibody in the patients plasma. Cessation given a 1A level of evidence.
of heparin is mandatory. LMWH should LMWH has also been found to pro-
not be used as substitute, due to a strong vide better DVT prophylaxis compared with
immunogenic cross-reactivity. In addition, placebo for extended 4-week prophylaxis in
warfarin alone is also contraindicated. The patients undergoing abdominal/pelvic can-
initial drop in protein C/S associated with cer surgery, decreasing by half the incidence
beginning VKA treatment in conjunction of postoperative VTE, with similar bleeding
with HIT may produce severe paradoxal rates.24
thrombosis. Direct thrombin inhibitors hiru-
din (lepirudin/Refludan) and argatroban
are the treatments approved by the FDA, Standard Therapy
although other agents such as fondaparinux
have been found to treat this syndrome as
for VTE Algorithm
well.22 Fondaparinux, a synthetic penthas- The algorithm in Figure 1.1 summarizes the
sacharide, does not produce this immune- standard pharmacological therapy for DVT.

DVT
Confirmed
Indication for Fibrinolysis = CDT

Formal indication for standard therapy or Venous Thrombectomy
2 consecutive INR > 2
Discontinue Heparin
UFH or LMWH UFH or LMWH UFH or LMWH UFH or LMWH UFH or LMWH
or Fondaparinux or Fondaparinux or Fondaparinux or Fondaparinux or Fondaparinux Day 6
Warfarin 5 mg Warfarin 5 mg Warfarin ? mg Warfarin ? mg Warfarin ? mg Warfarin (?) mg
Warfarin (?) mg
Day 1 Day 2 Day 3 Day 4 Day 5
Day 7 Warfarin (?) mg
Day 8
Warfarin (?) mg
Day 9
Day 10
Warfarin (?) mg
Day 11
Decision to stop or continue VKA
1. weigh risk versus benefit

2. clinical evaluation
3. concomitant diseases
4. cancer Month 3 - 6
5. warfarin compliance Warfarin (?) mg
6. risk of falling INR = 2 - 3
7. D-dimer
8. CUS Warfarin (?) mg
9. New Biomarkers (?)
Stop VKA or Continue VKA: Target INR = 2.0 3.0
Figure 1.1. Algorithm for the standard pharmacological therapy for DVT.

Nonpharmacological patients with iliofemoral DVT and offers the

best long-term outcome.
Treatments
Pain and swelling after an above-the-knee Rationale for
DVT can be decreased by approximately
50% by the use of strong compression
Thrombolytic Therapy
stockings. Additionally, walking with good In gathering the reported data of the patho-
compression does not increase the risk of physiology of postthrombotic syndrome, as
PE, while significantly decreasing the inci- well as the reported series of venous throm-
dence and severity of pain and swelling bectomy and randomized trials of venous
after DVT. It is recommended that once thrombectomy vs. anticoagulation for acute
patients are therapeutic on anticoagulants, iliofemoral DVT, a persistent observation
they ambulate while wearing compression surfaces. Patients in whom thrombosed
stockings. The use of strong compression venous segments are restored to patency
and early ambulation after initiation of have the lowest ambulatory venous pres-
DVT treatment can significantly reduce sure and the fewest postthrombotic symp-
the long-term morbidity of pain and swell- toms.26 Those patients with persistent venous
ing resulting from the DVT and carries a obstruction have the most severe postthrom-
1A level of evidence.25 botic symptoms, and this is associated with
the highest venous pressures. Therefore, it is
apparent that the long-term benefits of treat-
Invasive Therapies ment are directly related to maintenance of
for Acute DVT and PE a patent deep venous system.
There are 4 major pathophysiology con-
Proximal DVT, particularly of the iliofemo- cepts that support aggressive fibrinolytic or
ral system, is a severe form of venous occlu- thrombus removal for treatment of proximal
sion that in the long term leads to severe DVT of the lower extremities:
postthrombotic morbidity in a large number
of patients. Ninety percent of iliofemoral 1. Residual thrombus (thrombus burden)
DVT patients who are treated with antico- increases the odds of rethrombosis
agulation therapy alone will have ambula- 2. Thrombus burden is related to increased
tory venous hypertension, resulting in severe thrombus activity
chronic venous insufficiency. Up to 40% 3. Persistent thrombus activity favors
will have symptoms of venous claudication, rethrombosis
and within 5 years, up to 15% may develop 4. Rethrombosis increases postthrombotic
venous ulceration. In addition, the severity morbidity
of the acute venous thrombotic event is pre-
dictive of the degree of postthrombotic mor- Various authors have led strong advo-
bidity, with consequent impact on quality of cacy of early thrombus removal worldwide.
life. This is especially true in patients with Thrombus removal strategies promote early
iliofemoral DVT. Recent studies are provid- restoration of patency and improved venous
ing solid evidence that a strategy of throm- return. Recently, new evidence-based data
bus removal is the preferred management for have encouraged a more aggressive approach

to remove proximal thrombus, rather than a Catheter-Directed Thrombolysis

conservative treatment based upon antico-
agulation alone. The initial attempts to dissolve DVT were
The 2004 Seventh ACCP consensus based on systemic infusion of plasminogen
conference section stated that there is no activators. Higher rates of bleeding compli-
evidence that supports the use of thrombo- cations associated with poor reduction in
lytic agents for the initial treatment of DVT postthrombotic morbidity were frustrating.
in the large majority of patients.27 Recom- However, some of these patients achieved
mendations were against the routine use good results, preserving their valvular func-
of catheter-directed thrombolysis (CDT) tion, with reduced postthrombotic morbidity.
(grade 1C). In addition, it was emphasized The concept of directly delivering
that thrombolytic therapy should be con- smaller amounts of thrombolytic agents
fined to patients requiring limb salvage into the thrombus results in higher rates of
(grade 2C). clot dissolution, shorter treatment times,
More recently, the 2008 Eighth ACCP and reduced bleeding complications. Many
consensus conference changed its statements. reports have documented good outcomes
The recommendations regarding CDT for of catheter-directed thrombolysis for acute
acute DVTs are: In selected patients with DVT. Generally, success rates in the 75%
extensive acute proximal DVT (eg, ileo- to 90% range can be anticipated. Bleeding
femoral DVT, symptoms for <14 days, good complications have been reported in up to
functional status, life expectancy of 1 year) 11% of cases; however, in the majority of the
who have a low risk of bleeding, we suggest reports published within the past 6 years,
that catheter-directed thrombolysis may be bleeding complications are 5% or less, with
used to reduce acute symptoms and post- very few intracranial bleeds. Symptomatic
thrombotic morbidity if appropriate exper- PE during thrombolytic infusion is uncom-
tise and resources are available (grade 2B). mon and fatal PE is a rarity.
After successful CDT, we suggest correction
of underlying venous lesions using balloon
angioplasty and stents (grade 2C).28 Addi- Pharmacomechanical Thrombolysis
tionally, pharmacomechanical thrombolysis Good results have been reported with CDT.
(eg, with inclusion of thrombus fragmenta- However, reports indicate an average treat-
tion or aspiration) in preference to CDT ment time for CDT of 71 hours. This dura-
alone to shorten treatment time if appropriate tion of acute care is logistically difficult for
expertise and resources are available, is given many practitioners and medical centers.
a grade 2C recommendation. Percutaneous mechanical techniques
alone or in combination with thromboly-
sis have been developed to more rapidly
Treatment Options clear the venous system. These techniques
include catheters with 2 occluding balloons,
Several methods for removal of thrombus drug infusion holes between the balloons,
are available, including catheter-directed and mechanical drug dispersion capabili-
thrombolysis, percutaneous pharmacome- ties. This pharmacomechanical combina-
chanical thrombolysis, and operative venous tion enables focused treatment of thrombus
thrombectomy. within a targeted vessel. Other modalities

include catheters with microsonic (ultra- for <7 days, good functional status, and life
sound emission) capabilities to facilitate expectancy of >1 year) can be treated with
fibrin dissolution, and catheters that utilize operative venous thrombectomy to reduce
the Venturi effect. acute symptoms and postthrombotic mor-
The effectiveness of mechanical throm- bidity, if appropriate expertise and resources
bectomy alone or in combination with are available (grade 2B). If such patients do
pharmacologic thrombolysis was recently not have a high risk of bleeding, CDT is usu-
compared. Mechanical thrombectomy alone ally preferable to operative venous thrombec-
was successful for removing a thrombus that tomy, with a grade 2C recommendation.28
developed intraprocedurally (which is gen-
erally gelatinous and not cross-linked with
fibrin). However, in patients with thrombosis Future Medical
already installed, only 26% of the thrombus
was removed by mechanical thrombectomy.
Treatments for DVT/PE
The addition of a plasminogen activator solu- The evolution of anticoagulant drugs for
tion to the mechanical technique removed VTE from the 1930s to the present day is a
82% of the thrombus.29 series of agents with increasing specificity.
Traditional therapy has been standard intra-
venous unfractionated heparin, which is a
Operative Venous Thrombectomy good drug for preventing fatal PE. However,
Contemporary venous thrombectomy for heparin has a number of potential prob-
iliofemoral venous thrombosis has been lems, including bleeding, the development
shown to be effective in both short- and long- of osteoporosis and alopecia with high doses
term follow-ups. The long-term benefits of over long periods of time, the development of
venous thrombectomy relate to its ability HIT (heparin-induced thrombocytopenia),
to achieve proximal patency and maintain and importantly, the need for IV administra-
distal valve competency. Both outcomes are tion and frequent monitoring. Additionally,
influenced by the initial technical success contaminated heparins were a problem ear-
and the avoidance of recurrent thrombosis. lier in 2008. Low-molecular-weight heparins
Therefore, attention to operative detail in (LMWHs) are an alternative to standard
terms of complete thrombus removal, cor- unfractionated heparin, which has become
recting underlying venous stenoses, and the gold standard for the prophylaxis and
maintaining therapeutic anticoagulation treatment of VTE. LMWHs work more on
postoperatively is crucial. Pooled data from a inhibiting factor Xa than on inhibiting fac-
number of contemporary reports on iliofem- tor IIa (thrombin). Advantages of LMWHs
oral venous thrombectomy demonstrate that include an improved pharmacokinetic pro-
early and long-term patency of the iliofemo- file, a half-life that is not dose-dependent,
ral venous segment is 70% to 80% compared less physiologic antiplatelet activity, more
with 30% of patients treated with anticoagu- constant anti-factor Xa activity, less protein
lation alone.1 C antigen decrease, less complement activa-
The indication for venous thrombec- tion, less inhibition of physiologic platelet
tomy in proximal DVT patients was reviewed aggregation, and importantly the ability for
in the latest (2008) ACCP consensus: subcutaneous administration with no moni-
patients with proximal DVT (eg, symptoms toring necessary (except in renal failure,

pregnancy, or morbid obesity).30 LMWH is an oral direct factor Xa inhibitor that inhib-
a little better than standard unfractionated its free and fibrin-bound factor Xa activity
heparin regarding recurrent venous throm- and thrombotic activity. It has potent anti-
boembolic events, major hemorrhage, and coagulant effects, and it does not directly
even mortality compared to standard hepa- inhibit thrombin but instead inhibits throm-
rin. Even proximal above-knee thrombi are bin generation via inhibition of factor Xa
associated with improvements with LMWH activity. It has a rapid onset of action (within
therapy. In fact, the latest 2008 ACCP guide- 24 hours), high bioavailability (greater
lines give LMWH a 1A recommendation for than 80%), and does not affect agonist-
the initial treatment of VTE.28 induced platelet aggregation and therefore
Newer drugs include fondaparinux, has no direct effects on primary hemostasis.
among others. This indirect thrombin It does not require a cofactor such as anti-
inhibitor is a synthetic pentasaccharide that thrombin, it may be safely administrated
is identical to the antithrombin III binding with concomitant agents, and there are no
sequence of heparin. Fondaparinux is useful dosage requirement adjustments needed
for the prophylaxis and treatment of VTE, for gender, age, or extreme body weight.
and it has been FDA-approved for the pro- It has been evaluated in treatment studies
phylaxis of DVT, which may lead to PE in and prophylaxis studies, where it has been
patients undergoing abdominal surgery, found to be quite effective in reducing the
hip fracture surgery, extended prophylaxis risk for VTE, between approximately 30%
for hip fracture surgery, hip replacement and 80%. Apixaban, another factor Xa
surgery, and knee replacement surgery. inhibitor, has also successfully undergone
Additionally, it is indicated for the treat- both treatment and prophylaxis protocols.
ment of acute DVT when administered in Finally, dabigatran etexilate is a new oral
conjunction with warfarin sodium and for direct thrombin inhibitor. This drug has
the treatment of PE when administered in predictable anticoagulant effects, no need
conjunction with warfarin sodium and ini- for monitoring, and binds directly to throm-
tial therapy is given in the hospital. A cousin bin with high affinity and specificity. This
of fondaparinux, Idraparinux, has a 10-day drug was approved for prophylaxis in total
half-life. However, it did not meet the non- hip replacement and total knee replace-
inferiority target against PE in a recent study ment in Western Europe and Canada, but
and was also associated with significant approval will require additional trials in the
intracranial bleeding. Thus, its development United States as testing against LMWH 30
was halted. The drug has been biotinylated mg b.i.d. failed to meet the noninferiority
so that its effects can be reversed by avidin. target. This drug shows a low rate of bleed-
This drug is called SSR126517 and is under- ing comparable to LMWH, demonstrates
going evaluation. elevated liver enzymes in only a small por-
Other new drugs target specifically tion of patients (comparable with LMWH
either factor Xa or IIa and are oral agents. treatment), and offers oral dosing without
These include rivaroxaban, apixaban, and coagulation monitoring.
dabigatran.31 A comparison of these 3 drugs HIT requires special components for
reveals that they all are orally administered, its treatment. Agents available for patients
they have varying bioavailability, and their with HIT include direct thrombin inhibi-
half-lives are not the same. Rivaroxaban is tors lepirudin (Refludan), argatroban, and

fondaparinux.32 Refludan is a 65-amino- 2. Prandoni P, Lensing AW, Prins MR. Long-

acid protein that is the most potent throm- term outcomes after deep venous thrombosis
bin inhibitor, can be difficult to monitor, of the lower extremities. Vasc Med.
and undergoes renal excretion. Bivalirudin 1998;3(1):57-60.
(Angiomax) is an alternative, a shorter-act- 3. Hull RD, Raskob GE, Brant RF, Pineo GF,
ing agent requiring a drip. Argatroban is a Valentine KA. The importance of initial
synthetic thrombin inhibitor with little or heparin treatment on long-term clinical
no effect on factor Xa and that is metabo- outcomes of antithrombotic therapy. The
lized by the liver. There is a fairly predict- emerging theme of delayed recurrence. Arch
able aPPT nomogram starting with infusion Intern Med. 1997;157(20):2317-21.
of 2 to 4 g/kg/h, with a maximum of 10 4. Douketis JD, Kearon C, Bates S, Duku
g/kg/h. With Argatroban and Coumadin EK, Ginsberg JS. Risk of fatal pulmonary
administration, the INR must be greater embolism in patients with treated venous
than 4 before stopping therapy due to a false thromboembolism. JAMA. 1998;279(6):
elevation of INR with Argatroban. Moreover, 458-62.
fondaparinux has also been used success- 5. Prandoni P, Lensing AW, Cogo A, Cuppini
fully in HIT, but it has not been approved by S, Villalta S, Carta M, et al. The long-
the FDA for this indication. term clinical course of acute deep venous
Other antithrombotic agents are being thrombosis. Ann Intern Med. 1996;125(1):1-7.
evaluated, including plasminogen activation 6. Hull RD, Raskob GE, Hirsh J, Jay RM,
inhibitors (PAI-1), P-selectin inhibitors, and Leclerc JR, Geerts WH, et al. Continuous
PSGL-1 inhibitors (the ligand for P-selectin). intravenous heparin compared with
The use of P-selectin and PSGL-1 inhibitors intermittent subcutaneous heparin in the
is an area of intense ongoing research in initial treatment of proximal-vein thrombosis.
our laboratory. Such an anti-inflammatory N Engl J Med. 1986;315(18):1109-14.
approach uses an antithrombotic agent that 7. Weitz JI. Low-molecular-weight heparins.
does not cause direct anticoagulant activities N Engl J Med. 1997;337(10):688-98.
and thus raises the possibility of a therapeu- 8. Ageno W, Turpie AG. Low-molecular-weight
tic compound without bleeding potential. heparin in the treatment of pulmonary
Preclinical studies are promising. Aptamers embolism. Semin Vasc Surg. 2000;13(3):
that can bind P-selectin, thereby reducing 189-93.
inflammation and thrombus progression, are 9. Agnelli G, Prandoni P, Santamaria MG,
under development. Without directly target- Bagatella P, Iorio A, Bazzan M, et al. Three
ing the coagulation cascade, and therefore months versus one year of oral anticoagulant
dramatically reducing the risk of bleeding, therapy for idiopathic deep venous thrombosis.
these new compounds are focused on safety: Warfarin Optimal Duration Italian Trial
with an aptamer, a corresponding strand of Investigators. N Engl J Med. 2001;345(3):
genetic material can be made to immedi- 165-9.
ately reverse the effects of the drug. 10. Kearon C, Gent M, Hirsh J, Weitz J, Kovacs
MJ, Anderson DR, et al. A comparison of
references three months of anticoagulation with extended
1. Wakefield TW, Caprini J, Comerota AJ. anticoagulation for a first episode of idiopathic
Thromboembolic diseases. Curr Prob Surg. venous thromboembolism. N Engl J Med.
2008;45(12):844-99. 1999;340(12):901-7.

11. Hyers TM, Agnelli G, Hull RD, Morris TA, 19. Palareti G, Legnani C, Cosmi B, Guazzaloca
Samama M, Tapson V, et al. Antithrombotic G, Cini M, Mattarozzi S. Poor anticoagulation
therapy for venous thromboembolic disease. quality in the first 3 months after unprovoked
Chest. 2001;119(1 suppl):176S-193S. venous thromboembolism is a risk factor for
12. Schulman S, Rhedin AS, Lindmarker P, long-term recurrence. J Thromb Haemost.
Carlsson A, Larfars G, Nicol P, et al. A 2005;3(5):955-61.
comparison of six weeks with six months of 20. Garcia DA, Spyropoulos AC. Update in the
oral anticoagulant therapy after a first episode treatment of venous thromboembolism.
of venous thromboembolism. Duration of Sem Resp Crit Care M. 2008;29(1):40-6.
Anticoagulation Trial Study Group. N Engl J 21. Alving BM. How I treat heparin-induced
Med. 1995;332(25):1661-5. thrombocytopenia and thrombosis. Blood.
13. Eichinger S, Weltermann A, Minar E, 2003;101(1):31-7.
Stain M, Schonauer V, Schneider B, et al. 22. Kovacs MJ. Successful treatment of
Symptomatic pulmonary embolism and the heparin induced thrombocytopenia (HIT)
risk of recurrent venous thromboembolism. with fondaparinux. Thromb Haemost.
Arch Intern Med. 2004;164(1):92-6. 2005;93(5):999-1000.
14. Ridker PM, Goldhaber SZ, Danielson 23. Lee AY, Levine MN, Baker RI, Bowden
E, Rosenberg Y, Eby CS, Deitcher SR, C, Kakkar AK, Prins M, et al. Low-
et al. Long-term, low-intensity warfarin molecular-weight heparin versus a coumarin
therapy for the prevention of recurrent for the prevention of recurrent venous
venous thromboembolism. N Engl J Med. thromboembolism in patients with cancer.
2003;348(15):1425-34. N Engl J Med. 2003;349(2):146-53.
15. Kearon C, Ginsberg JS, Kovacs MJ, Anderson 24. Bergqvist D, Agnelli G, Cohen AT, Eldor
DR, Wells P, Julian JA, et al. Comparison A, Nilsson PE, Le Moigne-Amrani A, et
of low-intensity warfarin therapy with al. Duration of prophylaxis against venous
conventional-intensity warfarin therapy thromboembolism with enoxaparin
for long-term prevention of recurrent after surgery for cancer. N Engl J Med.
venous thromboembolism. N Engl J Med. 2002;346(13):975-80.
2003;349(7):631-9. 25. Prandoni P, Lensing AW, Prins MH, Frulla
16. Agnelli G, Prandoni P, Becattini C, Silingardi M, Marchiori A, Bernardi E, et al. Below-
M, Taliani MR, Miccio M, et al. Extended knee elastic compression stockings to
oral anticoagulant therapy after a first episode prevent the post-thrombotic syndrome: a
of pulmonary embolism. Ann Intern Med. randomized, controlled trial. Ann Intern Med.
2003;139(1):19-25. 2004;141(4):249-56.
17. Prandoni P, Lensing AW, Prins MH, 26. Eklof B, Kistner RL. Is there a role for
Bernardi E, Marchiori A, Bagatella P, et al. thrombectomy in iliofemoral venous
Residual venous thrombosis as a predictive thrombosis? Semin Vasc Surg. 1996;9(1):
factor of recurrent venous thrombo- 34-45.
embolism. Ann Intern Med. 2002;137(12): 27. Bller HR, Agnelli G, Hull RD, Hyers TM,
955-60. Prins MH, Raskob GE. Antithrombotic
18. Palareti G, Cosmi B, Legnani C, Tosetto A, therapy for venous thromboembolic
Brusi C, Iorio A, et al. D-dimer testing to disease: the Seventh ACCP Conference on
determine the duration of anticoagulation Antithrombotic and Thrombolytic Therapy.
therapy. N Engl J Med. 2006;355(17):1780-9. Chest. 2004;126(3 suppl):401S-28S.

28. Kearon C, Kahn SR, Agnelli G, Goldhaber of the efficacy and safety of once-daily
S, Raskob GE, Comerota AJ. Antithrombotic dose of enoxaparin versus unfractionated
therapy for venous thromboembolic disease: heparin in the treatment of proximal lower
American College of Chest Physicians limb deep-vein thrombosis. Thromb Res.
Evidence-Based Clinical Practice Guidelines 2004;114(3):149-53.
(8th ed.). Chest. 2008;133(6 suppl):454S-545S. 31. Gross PL, Weitz JI. New anticoagulants
29. Vedantham S, Vesely TM, Parti N, Darcy M, for treatment of venous thromboembolism.
Hovsepian DM, Picus D. Lower extremity Arterioscl Throm Vas. 2008;28(3):380-6.
venous thrombolysis with adjunctive 32. Baldwin ZK, Spitzer AL, Ng VL, Harken
mechanical thrombectomy. J Vasc Interv AH. Contemporary standards for the
Radiol. 2002;13(10):1001-8. diagnosis and treatment of heparin-
30. Ramacciotti E, Araujo GR, Lastoria S, Maffei induced thrombocytopenia (HIT). Surgery.
FH, Karaoglan de Moura L, Michaelis W, 2008;143(3):305-12.
et al. An open-label, comparative study

CHAPTER
2
Deep Venous Thrombosis
and Pulmonary embolism
Pathophysiology
Peter K. Henke
Acute Venous pathophysiology of thrombosis is important

for furthering medical and surgical therapies.
Thrombosis
Venous thromboembolism (VTE) is a sig-
nificant healthcare problem in the United Venous Thrombosis Pathways
States, with an estimated 900,000 cases of Hemostasis is often initiated by damage to
deep venous thrombosis (DVT) and pulmo- the vessel wall and disruption of the endo-
nary embolism (PE) reported yearly, with thelium, although it may be initiated in the
approximately 300,000 deaths.1 For the past absence of vessel wall damage, particularly
150 years, the pathogenesis of VTE cen- in venous thrombosis (VT),4 and new para-
tered on Virkows triad of stasis, changes in digms are emerging.5 Vessel wall damage
the vessel wall, and thrombogenic changes simultaneously results in release of tissue
in the blood. Stasis is probably not a direct factor (TF), a cell membrane protein, from
cause, while infection and systemic inflam- injured cells and the circulation, with subse-
mation may be more causal than previously quent activation of the extrinsic pathway of
thought.2,3 Beyond prevention, which is the the coagulation cascade (Figure 2.1). These
best-case scenario, understanding the basic 2 events are critical to the activation and
acceleration of thrombosis. Tissues also vary
with regard to their susceptibility to throm-
bosis, and the local mechanisms may be
somewhat different.6
Davies md and Alan B. Lumsden md, eds. The adhesion of platelets to exposed
Cardiotext Publishing, ISBN 978-1-935395-22-5. subendothelial collagen is one mechanism
19

Figure 2.1 Pathways of thrombosis. Two primary pathways that initiate thrombosis
are shown with the primary factors of platelets, tissue factor, and thrombin. These
converge on the coagulation pathway, and fibrin is formed, with thrombus produced.
PDI = protein disulfide isomerase; TF = tissue factor; gp = glycoprotein; vWF = von
Willebrand factor. Based on Furie and Furie, New Engl J Med. 2008;359:938(5).
for an effective hemostatic platelet plug to in the presence of calcium. Feedback ampli-
form, resulting in platelet activation. This fication occurs as VIIa, IXa, and Xa are all
interaction is mediated by von Willebrand capable of activating VII to VIIa, especially
factor (vWF), whose platelet receptor is gly- when bound to TF.7 Factor Xa is also capable
coprotein (Gp) Ib, and Gp VI and IX are of activating factor V to Va. Factors Xa, Va,
also involved. Direct TF de-encryption can and II (prothrombin) form on the platelet
occur by endothelial protein disulfide isom- phospholipid surface in the presence of Ca2+
erase, leading to platelet activation. Activa- to initiate the prothrombinase complex,
tion of platelets also leads to the release of the which catalyzes the formation of thrombin
prothrombotic contents of platelet granules, from prothrombin. Thrombin feedback
containing receptors for coagulation factors amplifies the system not only by activating
Va and VIIIa, as well as fibrinogen, vWF, factor V to Va, but also factors VIII (normally
and ADP. Fibrinogen forms bridges between circulating bound to vWF) to VIIIa and XI
platelets by binding to the GpIIb/IIIa recep- to XIa. After activation, factor VIIIa dissoci-
tor, resulting in further platelet aggregation. ates from vWF and assembles with factors
Platelet activation also leads to the elabora- IXa and X on the platelet surface in the pres-
tion of arachidonic acid metabolites such as ence of Ca2+ to form a complex called the
thromboxane A2, further promoting platelet Xase complex, which catalyzes the activa-
aggregation. tion of factor X to Xa. It is important to rec-
The extrinsic clotting pathway begins ognize that these events occur at the vessel
with TF complexing with factor VII, caus- wallplatelet interface.
ing activation (VIIa). The TF-VIIa complex Thrombin (factor II) is central to coagu-
then activates factors IX and X to IXa and Xa lation by its action of cleavage and release of

Deep Venous Thrombosis and Pulmonary Embolism Pathophysiology 21
fibrinopeptide A (FPA) from the -chain of E-selectin has been found up-regulated at
fibrinogen and fibrinopeptide B (FPB) from later time points.11
the -chain of fibrinogen. This causes fibrin Microparticles (MPs) are involved in
monomer polymerization and cross-linking, the initiation and amplification of throm-
stabilizing the thrombus and the initial plate- bosis. MPs are small (< 1m) phospholipid
let plug. Thrombin also activates factor XIII vesicles shed from platelets, leukocytes, and
to XIIIa, which catalyzes this cross-linking endothelial cells in a calcium-dependent
of fibrin as well as that of other plasma pro- fashion.12-14 MPs lack DNA and RNA but are
teins, such as fibronectin and 2-antitrypsin. protein rich, and subpopulations of MPs rich
In addition, factor XIIIa activates platelets as in TF and phosphatidylserine have been
well as factors V and VIII, further amplify- identified.15 Fusion of MPs with activated
ing thrombin production. platelets is another mechanism that results
The physiologic importance of the in decryption of TF and the initiation of
intrinsic pathway is not completely clear thrombosis.16 Several circulating markers of
and is probably not as physiologically impor- inflammation once thought to be soluble are
tant in the venous system as the extrinsic actually carried by MPs.17
system. Intrinsic pathway activation occurs In veins, CAMs and MPs interface in
with activation of factor XI to XIa, which promoting thrombosis and inflammation.
subsequently converts factor IX to IXa, pro- Both venous stasis and hypoxia result in the
moting formation of the Xase complex and up-regulation of P-selectin, which localizes
ultimately thrombin. Another mechanism prothrombotic MPs to the area of stasis and
by which this occurs in vitro is through the promotes DVT formation.18-20 The P-selec-
contact activation system, in which factor tin receptor, P-selectin glycoprotein ligand
XII (Hageman factor) is activated to XIIa 1(PSGL-1), is expressed on leukocytes and
when complexed to prekallikrein and high- platelets, as well as on their derived MPs.
molecular-weight kininogen (HMWK) on a Indeed, MPs coexpressing TF and leukocyte
negatively charged surface; factor XIIa then markers have been shown to accumulate in
activates factor XI to XIa. growing thrombi in a P-selectin:PSGL-1
dependent fashion.10 Further, P-selectin:
PSGL-1 interactions stimulate the produc-
Inflammation and Thrombosis tion of thrombogenic MPs from leukocytes,
In the early 1970s, the relationship between along with platelets and endothelial cells.21
thrombosis and inflammation was first sug- The importance of P-selectin:PSGL-1
gested.8 Inflammation increases TF, mem- to thrombosis also depends on the nature of
brane phospholipids, platelet reactivity, and the stimulus and the role of TF, which is nor-
fibrinogen, while decreasing thrombomodu- mally abundant in the outer portion of the
lin (TM) and inhibiting fibrinolysis.9 Coin- vessel wall. With significant vascular injury
cidently, cell adhesion molecules (CAM) and the exposure of vein wall TF, this TF is
are up-regulated on inflamed endothelium likely more important in the thrombogenic
and allow leukocyte transmigration. The process than the TF that is brought to the
selectins (P- and E-selectin) are integrally point of thrombogenesis by activated MPs,
involved in venous thrombosis.10 P-selectin in contrast to arterial thrombosis.22 Further-
is up-regulated in the vein wall as early as more, monocyte-derived MPs deliver TF to
6 hours after thrombus induction, while areas of injury and inflammation by binding

to P-selectin mobilized to the surface of acti- aminoglycans, including both heparan and
vated platelets and endothelial cells, result- dermatan sulfate, but its deficiency is not
ing in the generation of fibrin. associated with increased VTE risk.23
Natural Anticoagulants Fibrinolysis

Physiologic anticoagulants balance throm- In addition to preventing thrombus genera-
bin formation and localize thrombotic tion, controlled thrombolysis serves to limit
activity to sites of vascular injury by sev- thrombus extension as well as to metabolize
eral mechanisms. Antithrombin (AT) is a the thrombus to allow scar tissue to form.
central anticoagulant protein that binds to The central fibrinolytic enzyme is plasmin,
thrombin, and interferes with coagulation a serine protease generated by the proteolytic
by 3 major mechanisms. First, inhibition of cleavage of the proenzyme plasminogen. Its
thrombin prevents the removal of FPA and main substrates include fibrin, fibrinogen,
FPB from fibrinogen, limiting fibrin forma- and other coagulation factors. The degrada-
tion. Second, thrombin becomes unavail- tion of fibrin polymers by plasmin ultimately
able for factor V and VIII activation, slowing results in the creation of fragment E and 2
the coagulation cascade. Third, thrombin- molecules of fragment D, which are released
mediated platelet activation and aggregation as a covalently linked dimer (D-dimer).24
are inhibited. In the presence of heparin, Activation of plasminogen occurs by
inhibition of thrombin by AT is accelerated, several mechanisms. In the presence of
resulting in systemic anticoagulation. Anti- thrombin, vascular endothelial cells pro-
thrombin has been shown to directly inhibit duce and release tissue plasminogen activa-
factors VIIa, IXa, Xa, XIa, and XIIa. tor (tPA) as well as 2-antiplasmin, a natural
A second natural anticoagulant mecha- inhibitor of excess fibrin-bound plasmin. In
nism is activated protein C (APC), which is contrast to free-circulating tPA, fibrin-bound
produced on the surface of intact endothe- tPA is an efficient activator of plasminogen.
lium when thrombin binds to its receptor, A second endogenous activator of plas-
thrombomodulin (TM), and endothelial minogen is the urokinase-type plasminogen
protein C receptor (EPCR). The thrombin- activator (uPA), also produced by endothe-
TM complex inhibits the actions of throm- lial cells, but with less affinity for fibrin. The
bin and also activates protein C to APC. activation of uPA in vivo is not completely
APC, in the presence of its cofactor protein understood. However, it is hypothesized
S, inactivates factors Va and VIIIa, therefore that plasmin in small amounts (produced
reducing Xase and prothrombinase activity. through tPA) activates uPA, leading to fur-
The third innate anticoagulant is tis- ther plasminogen activation and amplifica-
sue factor pathway inhibitor (TFPI). This tion of fibrinolysis.25 This protease is most
protein binds the TF-VIIa complex, thus important for experimental DVT resolution,
inhibiting the activation of factor X to Xa as compared with tPA.26
and formation of the prothrombinase com- The third mechanism of plasminogen
plex. Finally, heparin cofactor II is another activation involves factors of the contact
inhibitor of thrombin whose action is in the activation system; activated forms of fac-
extravascular compartment. The activity of tors XII, kallikrein, and XI can each inde-
heparin cofactor II is augmented by glycos- pendently convert plasminogen to plasmin.

These activated factors may also catalyze the thelium; hence, the endothelium is one of
release of bradykinin from HMWK, which the pivotal regulators of homeostasis (Figure
further augments tPA secretion. Finally, 2.2). Under normal conditions, endothelial
APC has been found to proteolytically inac- cells maintain a vasodilatory and local fibri-
tivate plasminogen activator inhibitor type nolytic state in which coagulation, platelet
1 (PAI-1), an inhibitor of plasmin activators, adhesion, and activation are suppressed.
released by endothelial cells in the presence A nonthrombogenic endothelial surface is
of thrombin.27 maintained through a number of mecha-
In plasma, plasminogen activator nisms, including: (1) endothelial production
inhibitor (PAI-1) is the primary inhibitor of of TM and subsequent activation of protein
plasminogen activators and is stored in the C; (2) endothelial expression of heparan sul-
alpha-granules of quiescent platelets.25 It fate and dermatan sulfate, which accelerate
is secreted in an active form from liver and AT and heparin cofactor II activity; (3) con-
endothelial cells, and stabilized by binding stitutive expression of TFPI; and (4) local
to vitronectin (and inhibits thrombin in this production of tPA and uPA. In addition, the
form).28 Plasminogen activator inhibitor-1 production of nitric oxide (NO) and pros-
levels are elevated by hyperlipidemia, and tacyclin by the endothelium inhibits the
PAI-1 elevation appears to synergize with fac- adhesion and activation of leukocytes and
tor V Leiden genetic abnormalities, possibly produces vasodilation.29
contributing to pathological thrombosis. During states of endothelial distur-
bances, a prothrombotic and proinflamma-
tory state of vasoconstriction is supported by
Endothelium and Hemostasis the endothelial surface. Release of platelet-
Most of the thrombosis-thrombolysis pro- activating factor and endothelin-1 promotes
cesses occur in juxtaposition to the endo- vasoconstriction, while production of vWF,
Figure 2.2 Counterbalancing pro- and antithrombotic endothelial functions. At rest,

the balance is tipped toward an anticoagulant milieu. With inflammation or with local
damage, the balance tips to a procoagulant state by several mechanisms.

TF, PAI-1, and factor V augment thrombo- is the primary fibrinolytic mechanism in the
sis.30 Indeed, vWF is expressed to a greater PA, including the arterioles.38 Although ani-
extent on the endothelium of veins as com- mal models can mimic the acute to chronic
pared with arteries, and tPA is less commonly PE pathophysiology, DVT characteristics
expressed in venous endothelium. Loss of that predispose to PE as compared to no PE
endothelium likely also contributes to the have not yet been forthcoming, but would be
vein wall fibrosis, as well as the predisposi- of significant clinical interest.
tion to recurrent venous thrombosis. Experi-
mental DVT was associated with decreased
expression of homeostatic endothelial genes Differential Risk
such as NO and TM, as compared with
controls, and correlated with loss of vWF
of DVT vs. PE
positive cell luminal staining.31 Other inves- Approximately 50% of patients with a DVT
tigators have found that prolonged venous may suffer a PE, although many of the
stasis is associated with decreased plasmin- PEs are asymptomatic and of small magni-
ogen activators, probably related to loss of tude.33 Conversely, pulmonary symptoms
endothelium.32 often point the physician to a suspected PE.
Defining those at risk for PE after DVT is
diagnosed, given similar levels of risk, and
Experimental anticoagulation, is important, and data are
Pulmonary Embolism lacking. Adequate anticoagulation certainly
reduces the risk, but some patients present
Pulmonary embolism is the fatal complica- with primary PE without a DVT source de-
tion of DVT, primarily related to acute right fined and may have symptoms related to fail-
heart failure and less so to hypoxemia.33 ure of local DVT resolution.
While the basic processes of thrombolysis in From a large epidemiological study,
all vascular beds are similar, several experi- several factors including age and corticoste-
mental studies have suggested that pulmo- roid use are independently associated with
nary artery (PA) injury is different from in the PE, but not DVT.2 From a large patient
inferior vena cava (IVC). In an experimen- registry, predictors of fatal PE in patients
tal study in rats using an in vivogenerated with incident DVT included older age (>75
thromboembolism, the early PA response is years), cancer, and immobility related to
associated with elevated monocyte chemo- neurological disease.39 Recent studies also
tactic protein-1 (MCP-1), early monocyte suggest that very high D-dimer levels may be
influx, and later intimal hyperplasia.34 Little predictive of PE over isolated DVT, with a
elevation in PA selectins is observed, in con- threshold of >4000 ng/mL.40 Similarly, those
trast to the IVC response.35 Interestingly, the patients whose initial clinical event was a PE
mechanical obstruction in the PA induces rather than simply a DVT are at higher risk
significant injury, and the resulting intimal of recurrent PE.41 In hospitalized patients,
hyperplasia is independent of the thrombus the postoperative state may also confer an
effect.36 Moreover, LMWH, but neither tPA increased risk of PE, but studies specifically
nor a gIIb/IIIa antagonist, was associated defining the DVT prior to PE are few.42 It
with attenuated intimal hyperplasia.37 Con- is likely that the differences are related to
sistent with what is observed in the IVC, uPA the balance of thrombolysis, thrombusvein

wall adherence, and thrombus composition The most common inherited thrombophil-
related to undefined genetic factors. ias are divided among the deficiencies of the
natural anticoagulants, the genetic factor
abnormalities, and a miscellaneous category
Factors That (Table 2.1). The suggested screening tests
Increase VTE Risk to evaluate for the most common thrombo-
philias are as follows: antithrombin, protein
Most clinical VTEs have a proximate cause, C/S; factor V Leiden; prothrombin 20210A;
including environmental risks and genetic homocysteine; factor VIII; and antiphospho-
predispositions, which may account for lipid antibody panel.
most of the VTEs that manifest clinically.
The most common risk factors for VTE are
prior DVT, malignancy, immobility, intra- Biomarkers for VTE
venous catheters, increased age, major sur-
gery, trauma, infections such as pneumonia While atherosclerotic disease has numerous
and urinary tract, and certain chemothera- biomarkers for clinical symptom risk, such
pies.2,41,43 Some medications such as oral as C-reactive protein (CRP), cholesterol sub-
contraceptives and hormonal replacement types, Lipoprotein a, and more, markers for
therapies also increase the risk of VTE. A VT occurrence have been less forthcoming.
particularly aggressive acquired hyperco- Part of this has been a lack of understand-
agulable state is antiphospholipid antibody ing of VT pathophysiology and the nature
syndrome. of VTE occurrence as compared with arte-
Although beyond the scope of this rial events. Venous thrombosis biomarkers
chapter, it is important to emphasize that would be most useful to determine primary
under specific circumstances, evaluating risk, risk for recurrence, and perhaps defin-
for an inherited thrombophilic state is war- ing therapy duration of VTE.
ranted.44 Clinical scenarios include early age Venous thrombosis biomarkers are
of VTE (<40 years), recurrent unprovoked several and are variably studied clinically45
VTE, thrombosis at unusual sites, and a (Table 2.2). Some of these, such as factor
strong family history of VTE. The inter- VIII, may be etiologic for VTE, whereas
section of environmental stressor (eg, post- others such as D-dimer and thrombin-
surgery) with an inherited thrombophilia antithrombin complex (TAT) may reflect
is common; a form of the 2-hit hypothesis. the thrombotic process.46 D-dimer is cur-
Table 2.1 InheritedThrombophilias
Natural
AnticoagulantDeficiencies FactorGeneticAbnormalities Miscellaneous
Antithrombin Factor V Leiden Hyperhomocysteimia
Protein C PT G20210A Factor VIII
Protein S Factor IX
Antiphospholipid syndrome

Table 2.2 SelectedBiomarkersforAcuteVTE
Etiologic VTEdx Clinically

Biomarker Association Sens/Spec(%) Available?
D-dimer47 No 100/78 Yes
P-selectin53 Yes 71/81 No
Microparticles53 Yes 59/62 No
Factor VIII45 Yes UNK Yes
Thrombin antithrombin45 No 100/18 Yes (not all)

(>1.6 ng/mL)
Brain natriuretic No > 80%/40% for major complications Yes

peptide*55
OR=5.2, 95% CI = 3.38.4 for
Troponin*55 No cardiac death Yes
*These are specific for PE patients.
rently used to define the likelihood of VTE P-selectin levels correlate highly in patients
probability in the clinical setting.47 Studies with malignancy.41 Inhibition of P-selectin
are well established, confirming its use to can confer protection against VTE as sug-
effectively rule out, but not rule in, a VTE. gested by nonhuman primate studies.54
Defining VTE recurrence risk has been Specific anti-P-selectin agents are being
improved with D-dimer measurement and developed for human trials and may offer
now is included in the AACP guidelines.48 prophylaxis benefit with anticoagulation
For example, longer duration of VTE oral risks. Although E-selectin plays a role in
anticoagulation therapy is suggested by sig- DVT, it has not been examined as a bio-
nificantly elevated D-dimer levels after 3 marker in humans.
months of oral coagulant therapy.49 Throm- Both probrain natriuretic peptide
bin-antithrombin complex, although not (BNP) and troponin are useful biomarkers
widely available, is also useful for acute VTE reflecting the physiological stress of a PE.
probability46 and prediction of VTE recur- For example, a low BNP is associated with
rence.50 Other measures, such as assessment good clinical outcome, whereas BNP 1000
of fragment 1 + 2, may further improve these ng/mL is associated with a 12-fold increased
predictions.51 risk of cardiac complications in combination
Soluble P-selectin is an etiological fac- with an abnormal ECHO.55 Similarly, those
tor in VTE,52 and studies suggest its utility patients with significantly elevated troponin-
as additive for VTE prediction.53 Elevated I have a 5.9-fold increased risk of in-hospital

mortality. These biomarkers may be helpful The risk of VTE was increased 2.0-fold for
in selecting those who would benefit from obesity, 1.5-fold for hypertension, 1.4 for
aggressive thrombus removal and close clini- diabetes, 1.2 for smoking, and 1.2 for hyper-
cal support.56 cholesterolemia.62 Interestingly, mean HDL
On the horizon is potentially more levels were lower in patients with VTE as
specific VTE biomarker risk assessment by compared with controls, although this find-
proteonomic evaluation.57,58 Limiting this ing is not universal.63 Reports are forthcom-
will be ease of use and rapidity of test results. ing that HMG-CoA reductase inhibitors
However, refinements in technology may (statins) may be associated with decreased
make this very useful in the future for assess- DVT incidence. Registry cohort patient data
ment of VTE risk, particularly if specific suggest a 22% risk reduction of symptomatic
enough to allow risk prediction in asymp- VTE in patients on a statin, as compared
tomatic patients. to matched patients not on a statin.64 More
impressive are results of a randomized clini-
cal trial of patients with normal cholesterol
but elevated hsCRP who were assigned to
Commonalities rosuvastatin or placebo. A 45% highly sig-
nificant reduction of incident VTE was
of Arterial and observed.65 In addition to the endothelial
Venous Thrombosis protective effects of statins, these agents may

increase TM expression and enhance APC
Several epidemiologic and basic science activity, tipping the balance of endothelium
studies have suggested a common interface toward a more anti-inflammatory and anti-
between venous and arterial vascular pathol- coagulant state. Future research will better
ogy. In a prospective study of nearly 300 define the vascular biology of venous throm-
patients with DVT and a control group of bosis, which may reveal further similarities
150 subjects, evaluation of carotid plaque in arterial and venous risk factors and pos-
progression was done. Patients with a spon- sibly response to therapy.
taneous DVT had a 2.3-fold increased
incidence of atherosclerosis progression, as
compared with control subjects, which was references
independent of other atherosclerotic risk 1. Heit JA, Cohen AT, Anderson FJ. Estimated
factors.59 Similarly, those patients with a annual number of incident and recurrent,
spontaneous VTE had a 1.6-fold increased non-fatal venous thromboembolism (VTE)
incidence of symptomatic atherosclerosis,60 events in the US. Blood. 2005;106(11):
suggesting overlap in vascular inflammation 910.
and risk. Interestingly, the proven effective- 2. Gangireddy C, Rectenwald JR, Upchurch
ness for antiplatelet therapy in preventing GR, Wakefield TW, Khuri S, Henderson
arterial thrombotic events does not translate WG, et al. Risk factors and clinical
to prevention of VTE,48,61 despite the fact impact of postoperative symptomatic
that platelets play a role. venous thromboembolism. J Vasc Surg.
The concept that VTE and atheroscle- 2007;45(2):335-41.
rosis may share common risk factors has 3. Meissner MH, Wakefield TW, Ascher E,
also been assessed by a recent meta-analysis. Caprini JA, Comerota AJ, Eklof B, et al. Acute

venous disease: venous thrombosis and venous vivo is dependent upon microparticle
trauma. J Vasc Surg. 2007;46 suppl S:25S-53S. P-selectin glycoprotein ligand 1 and platelet
4. Mackman N, Tilley RE, Key NS. Role of the P-selectin. J Exp Med. 2003;197(11):1585-98.
extrinsic pathway of blood coagulation in 16. Osterud B. The role of platelets in decrypting
hemostasis and thrombosis. Arterioscl Throm monocyte tissue factor. Semin Hematol.
Vas. 2007;27(8):1687-93. 2001;38(4 suppl 12):2-5.
5. Furie B, Furie BC. Mechanisms of thrombus 17. Ahn ER, Lander G, Jy W, Bidot CJ,
formation. N Engl J Med. 2008;359(9):938-49. Jimenez JJ, Horstman LL, Ahn YS.
6. Mackman N. Tissue-specific hemostasis in Differences of soluble CD40L in sera and
mice. Arterioscl Throm Vas. 2005;25(11): plasma: implications on CD40L assay as
2273-81. a marker of thrombotic risk. Thromb Res.
7. Dahlback B. Blood coagulation. Lancet. 2004;114(2):143-8.
2000;355(9215):1627-32. 18. Myers DD, Hawley AE, Farris DM, Wrobleski
8. Stewart GJ. Neutrophils and deep venous SK, Thanaporn P, Schaub RG, et al. P-selectin
thrombosis. Haemostasis. 1993;23 suppl 1:1 and leukocyte microparticles are associated
27-40. with venous thrombogenesis. J Vasc Surg.
9. Esmon CT. Inflammation and thrombosis. 2003;38(5):1075-89.
J Thromb Haemost. 2003;1(7):1343-8. 19. Myers DD, Wakefield TW. Inflammation-
10. Celi A, Pellegrini G, Lorenzet R, De Blasi A, dependent thrombosis. Front Biosci.
Ready N, Furie BC, et al. P-selectin induces 2005;10:2750-7.
the expression of tissue factor on monocytes. 20. Polgar J, Matuskova J, Wagner DD. The
P Natl Acad Sci USA. 1994;91(19):8767-71. P-selectin, tissue factor, coagulation triad.
11. Myers D Jr, Farris D, Hawley A, Wrobleski J Thromb Haemost. 2005;3(8):1590-6.
S, Chapman A, Stoolman L, et al. Selectins 21. Andre P, Hartwell D, Hrachovinova I,
influence thrombosis in a mouse model of Saffaripour S, Wagner DD. Pro-coagulant
experimental deep venous thrombosis. J Surg state resulting from high levels of soluble
Res. 2002;108(2):212-21. P-selectin in blood. P Natl Acad Sci USA.
12. Gilbert GE, Sims PJ, Wiedmer T, Furie 2000;97(25):13835-40.
B, Furie BC, Shattil SJ. Platelet-derived 22. Day SM, Reeve JL, Pedersen B, Farris DM,
microparticles express high affinity Myers DD, Im M, et al. Macrovascular
receptors for factor VIII. J Biol Chem. thrombosis is driven by tissue factor derived
1991;266(26):17261-68. primarily from the blood vessel wall. Blood.
13. Mesri M, Altieri DC. Endothelial cell 2005;105(1):192-8.
activation by leukocyte microparticles. 23. Corral J, Aznar J, Gonzalez-Conejero R, Villa
J Immunol. 1998;161(8):4382-7. P, Minano A, Vaya A, et al. Homozygous
14. Sabatier F, Roux V, Anfosso F, Camoin L, deficiency of heparin cofactor II: relevance of
Sampol J, Dignat-George F. Interaction of P17 glutamate residue in serpins, relationship
endothelial microparticles with monocytic with conformational diseases, and role in
cells in vitro induces tissue factor- thrombosis. Circulation. 2004;110(10):1303-7.
dependent procoagulant activity. Blood. 24. Hassouna HI. Laboratory evaluation of
2002;99(11):3962-70. hemostatic disorders. Hematol Oncol Clin N.
15. Falati S, Liu Q, Gross P, Merrill-Skoloff G, 1993;7(6):1161-249.
Chou J, Vandendries E, et al. Accumulation 25. Sidelmann JJ, Gram J, Jespersen J, Kluft
of tissue factor into developing thrombi in C. Fibrin clot formation and lysis: basic

mechanisms. Semin Thromb Hemost. al. Viral IL-10 gene transfer decreases
2000;26(6):605-18. inflammation and cell adhesion molecule
26. Singh I, Burnand KG, Collins M, Luttun expression in a rat model of venous
A, Collen D, Boelhouwer B, et al. Failure thrombosis. J Immunol. 2000;164(4):2131-41.
of thrombus to resolve in urokinase-type 36. Zagorski J, Debelak J, Gellar M, Watts JA,
plasminogen activator gene-knockout mice: Kline JA. Chemokines accumulate in the
rescue by normal bone marrow-derived cells. lungs of rats with severe pulmonary embolism
Circulation. 2003;107(6):869-75. induced by polystyrene microspheres.
27. Esmon CT. The regulation of natural J Immunol. 2003;171(10):5529-36.
anticoagulant pathways. Science. 37. Rectenwald JE, Deatrick KB, Sukheepod P,
1987;235(4794):1348-52. Lynch EM, Moore AJ, Moaveni DM, et al.
28. Stoop AA, Lupu F, Pannekoek H. Experimental pulmonary embolism: effects
Colocalization of thrombin, PAI-1, and of the thrombus and attenuation of pulm-
vitronectin in the atherosclerotic vessel wall: onary artery injury by low-molecular-
a potential regulatory mechanism of thrombin weight heparin. J Vasc Surg. 2006;43(4):
activity by PAI-1/vitronectin complexes. 800-8.
Arterioscl Throm Vas. 2000;20(4):1143-49. 38. Bdeir K, Murciano JC, Tomaszewski J,
29. Becker MD, ORourke LM, Blackman WS, Koniaris L, Martinez J, Cines DB, et al.
Planck SR, Rosenbaum JT. Reduced leukocyte Urokinase mediates fibrinolysis in the
migration, but normal rolling and arrest, in pulmonary microvasculature. Blood.
interleukin-8 receptor homologue knockout 2000;96(5):1820-6.
mice. Invest Ophth Vis Sci. 2000;41(7):1812-7. 39. Laporte S, Mismetti P, Decousus H, Uresandi
30. Aird WC. Phenotypic heterogeneity of the F, Otero R, Lobo JL, et al. Clinical predictors
endothelium: I. Structure, function, and for fatal pulmonary embolism in 15,520
mechanisms. Circ Res. 2007;100(2):158-73. patients with venous thromboembolism:
31. Moaveni DK, Lynch EM, Luke C, Sood V, findings from the Registro Informatizado
Upchurch GR, Wakefield TW, et al. Vein de la Enfermedad TromboEmbolica
wall re-endothelialization after deep vein venosa (RIETE) registry. Circulation.
thrombosis is improved with low-molecular- 2008;117(13):1711-6.
weight heparin. J Vasc Surg. 2008;47(3):616-24. 40. Tick LW, Nijkeuter M, Kramer MH, Hovens
32. Stenberg B, Bylock A, Risberg B. Effect MM, Buller HR, Leebeek FW, et al. High
of venous stasis on vessel wall fibrinolysis. D-dimer levels increase the likelihood
Thromb Haemost. 1984;51(2):240-2. of pulmonary embolism. J Intern Med.
33. Tapson VF. Acute pulmonary embolism. 2008;264(2):195-200.
N Engl J Med. 2008;358(10):1037-52. 41. Zhu T, Martinez I, Emmerich J. Venous
34. Eagleton MJ, Henke PK, Luke CE, Hawley thromboembolism: risk factors for recurrence.
AE, Bedi A, Knipp BS, et al. Southern Arterioscl Throm Vas. 2009;29(3):298-310.
Association for Vascular Surgery William J. 42. Henke PK, Ferguson E, Varma M, Deatrick
von Leibig Award. Inflammation and intimal KB, Wakefield GT, Woodrum DT. Proximate
hyperplasia associated with experimental versus nonproximate risk factor associated
pulmonary embolism. J Vasc Surg. primary deep venous thrombosis: clinical
2002;36(3):581-8. spectrum and outcomes. J Vasc Surg. 2007;
35. Henke PK, DeBrunye LA, Strieter RM, 45(5):998-1003; discussion 1003-4; quiz
Bromberg JS, Prince M, Kadell AM, et 1005-7.

43. Heit JA. The epidemiology of venous Longo C, Henke PK, Guire KE, et al.
thromboembolism in the community. D-dimer, P-selectin, and microparticles:
Arterioscl Throm Vas. 2008;28(3):370-2. novel markers to predict deep venous
44. Middeldorp S, Levi M. Thrombophilia: thrombosis. A pilot study. Thromb Haemost.
an update. Semin Thromb Hemost. 2005;94(6):1312-7.
2007;33(6):563-72. 54. Myers DD Jr, Wrobleski SK, Longo C, Bedard
45. Pabinger I, Ay C. Biomarkers and venous PW, Kaila N, Shaw GD, et al. Resolution of
thromboembolism. Arterioscl Throm Vas. venous thrombosis using a novel oral small-
2009;29(3):332-6. molecule inhibitor of P-selectin (PSI-697)
46. Bozic M, Blinc A, Stegnar M. D-dimer, other without anticoagulation. Thromb Haemost.
markers of haemostasis activation and soluble 2007;97(3):400-7.
adhesion molecules in patients with different 55. Binder L, Pieske B, Olschewski M, Geibel A,
clinical probabilities of deep vein thrombosis. Klostermann B, Reiner C, et al. N-terminal
Thromb Res. 2003;108:107-14. pro-brain natriuretic peptide or troponin
47. Wells PS, Anderson DR, Rodger M, Forgie testing followed by echocardiography for risk
M, Kearon C, Dreyer J, et al. Evaluation stratification of acute pulmonary embolism.
of D-dimer in the diagnosis of suspected Circulation. 2 005;112(11):1573-9.
deep-vein thrombosis. N Engl J Med. 56. Piazza G, Goldhaber SZ. Acute pulmonary
2003;349(13):1227-35. embolism: part II: treatment and prophylaxis.
48. Geerts W. Antithrombotic and thrombolytic Circulation. 2006;114(3):e42-7.
therapy. Chest. 2008;133 (8th ed: ACCP 57. Hong CC, Kume T, Peterson RT. Role
Guidelines):381s-451s. of crosstalk between phosphatidylinositol
49. Palareti G, Cosmi B, Legnani C, Tosetto A, 3-kinase and extracellular signal-regulated
Brusi C, Iorio A, et al. D-dimer testing to kinase/mitogen-activated protein kinase
determine the duration of anticoagulation pathways in artery-vein specification. Circ Res.
therapy. N Engl J Med. 2006;355(17):1780-9. 2008;103(6):573-9.
50. Hron G, Kollars M, Binder BR, Eichinger 58. Ganesh SK, Sharma Y, Dayhoff J, Fales HM,
S, Kyrle PA. Identification of patients at low Van Eyk J, Kickler TS, Billings EM, Nabel
risk for recurrent venous thromboembolism EG. Detection of venous thromboembolism
by measuring thrombin generation. JAMA. by proteomic serum biomarkers. PLoS One.
2006;296(4):397-402. 2007;2(6):e544.
51. Poli D, Antonucci E, Ciuti G, Abbate R, 59. Prandoni P, Bilora F, Marchiori A, Bernardi E,
Prisco D. Combination of D-dimer, F1+2 and Petrobelli F, Lensing AW, et al. An association
residual vein obstruction as predictors of VTE between atherosclerosis and venous
recurrence in patients with first VTE episode thrombosis. N Engl J Med. 2003;348(15):
after OAT withdrawal. J Thromb Haemost. 1435-41.
2008;6(4):708-10. 60. Prandoni P, Ghirarduzzi A, Prins MH, Pengo
52. Hrachovinova I, Cambien B, Hafezi- V, Davidson BL, Sorensen H, et al. Venous
Moghadam A, Kappelmayer J, Camphausen thromboembolism and the risk of subsequent
RT, Widom A, et al. Interaction of P-selectin symptomatic atherosclerosis. J Thromb
and PSGL-1 generates microparticles that Haemost. 2006;4(9):1891-6.
correct hemostasis in a mouse model of 61. Glynn RJ, Ridker PM, Goldhaber SZ,
hemophilia A. Nat Med. 2003;9(8):1020-5. Buring JE. Effect of low-dose aspirin on the
53. Rectenwald JE, Myers DD Jr, Hawley AE, occurrence of venous thromboembolism:

a randomized trial. Ann Intern Med. (LITE). Blood. 2008;112(7):2675-80.

2007;147(8):525-33. 64. Ray JG, Mamdani M, Tsuyuki RT, Anderson
62. Ageno W, Becattini C, Brighton T, Selby R, DR, Yeo EL, Laupacis A. Use of statins and
Kamphuisen PW. Cardiovascular risk factors the subsequent development of deep vein
and venous thromboembolism: a meta- thrombosis. Arch Intern Med. 2001;161(June
analysis. Circulation. 2008;117(1):93-102. 11):1405-10.
63. Chamberlain AM, Folsom AR, Heckbert 65. Glynn RJ, Danielson E, Fonseca FA, Genest J,
SR, Rosamond WD, Cushman M. High- Gotto AM Jr, Kastelein JJ, et al. A randomized
density lipoprotein cholesterol and venous trial of rosuvastatin in the prevention of
thromboembolism in the Longitudinal venous thromboembolism. N Engl J Med.
Investigation of Thromboembolism Etiology 2009;360(18):1851-61.

CHAPTER

Prevention and Treatment
Mark g. Davies
Introduction 10% of those with PE.3 The mortality rate for

PE has been estimated to be as high as 30%
Venous thromboembolic diseases comprise in studies that included autopsy-based PE di-
the spectrum from deep venous thrombosis agnosis.4 Mortality rates are lower among pa-
(DVT) to pulmonary embolism (PE), and tients with idiopathic venous thrombosis and
occur with an incidence of approximately highest among those whose thrombosis oc-
1 per 1000 annually in adult populations.1 curs in the setting of cancer. Venous throm-
Rates are slightly higher in men than in bosis increases in incidence with age, with a
women. About two-thirds of episodes mani- low rate of about 1 per 10,000 annually be-
fest as DVT, and one-third as PE with or fore the fourth decade of life, rising rapidly
without DVT. The major outcomes of ve- after age 45 years and approaching 5 to 6 per
nous thrombosis are death, recurrence, post- 1000 annually by age 80.5,6 The morbidity
thrombotic syndrome, and major bleeding impact of thrombosis on the elderly appears
due to anticoagulation. Thrombosis is also to be greater, with a steeper rise in incidence
associated with impaired quality of life, par- of PE as compared to DVT with aging.5
ticularly when postthrombotic syndrome de-
velops.2 Death occurs within 1 month of an
episode in about 6% of those with DVT and Lower Limb
There are more than a quarter of a million
hospital admissions each year in the United
States for acute lower extremity deep venous
Davies md and Alan B. Lumsden md, eds. thrombosis and pulmonary embolism.7-9 In
Cardiotext Publishing, ISBN 978-1-935395-22-5. hospital patients without prophylaxis, it is
33

estimated that the incidence of isolated calf Urologic

DVT and proximal DVT is 25% and 7%, Gynecological
respectively.9,10 When thrombosis is proxi- Any procedure >2 hours long
mal to the calf, there is a 50% likelihood of Use of oral contraceptives
pulmonary embolism. There are geographi- Pregnancy
cal differences in reports on DVT, with Presence of hypercoagulable
up to twice the reported incidence of calf disorder
DVT being reported in Europe compared to
North America.10 Commonly identified risk factors in-
clude age >40 years, prolonged immobility,
prior venous thrombotic disease, malignan-
Upper Limb cy, major surgery, obesity, varicose veins,
Upper extremity venous thrombosis repre- congestive heart failure, myocardial infarc-
sents 0.5% to 1.5% of all venous thrombo- tion, orthopedic trauma, and estrogen use.
ses. The primary form of the disease also An increasing concern during the workup
is known as effort thrombosis or Paget- for a DVT is identification of a hypercoagu-
Schroetter disease. The secondary form of lable disorder (Table 3.1). Primary disorders
the disease is most commonly a result of involve a specific defect of a hemostatic
central venous catheterization for central protein, while secondary disorders result in
venous or cardiac access. Iatrogenic causes an indirect effect on the hemostatic path-
of secondary venous thrombosis account for ways.14-17 All patients with thrombosis at an
up to 30% of symptomatic subclavian ve- early age (<45 years), with a positive family
nous thromboses.11 In addition, studies have history of thrombotic disease, thrombosis at
shown that clinically silent thrombosis may an unusual site (mesenteric vein, cerebral
occur in 20% to 30% of patients after central vein), or recurrent thromboses without pre-
venous catheter insertion.12,13 disposing factors should be screened for hy-
percoagulability. In considering who needs
prophylaxis, patients may be categorized
Risk Factors into levels of risk: low, moderate, high, and
highest (Table 3.2). In these categories, the
Multiple studies have consistently identi- incidence of calf vein DVT is 2%, 10% to
fied risk factors for the development of DVT, 20%, 20% to 40% and 40% to 80%, whereas
which are outlined in the following list: for proximal DVT, the incidence is 0.4%,
2% to 4%, and 4% to 8% and 10% to 20%,
Age >40 years respectively.
Male sex
Obesity
Presence of malignancy Pathogenesis
Presence of varicose veins
Prior deep venous thrombosis One or more of 3 conditions cause thrombi
or pulmonary embolism to form in the systemic veins: reduced blood
Type of procedure flow in the systemic veins, injury to the vein
Orthopedic wall, and the presence of hypercoagulabil-
Neurosurgical ity. These factors remain important in the

Deep Venous Thrombosis 35
Table 3.1 HypercoagulableStates
Primary
Antithrombin III deficiency Heparan cofactor II deficiency
Protein C deficiency
Protein S deficiency
Factor V Leiden
Fibrinolysis deficiencies (dysplasminogenemia, decreased plasminogen activator
or increased plasminogen activator inhibitors, dysfibrinogenemia)
Factor VII deficiency
Secondary Defects of coagulation and fibrinolysis

Malignancy
Pregnancy
Use of oral contraceptives
Nephrotic syndrome
Lupus anticoagulant
Defects of platelets
Myeloproliferative disorders
Paroxysmal nocturnal hematuria
Diabetes mellitus
Hyperlipidemia
Cushings syndrome
Heparin-induced thrombocytopenia
Defects of blood vessels and rheology
Immobilization
Postoperative care
Vasculitis
Chronic obliterative arterial disease
Homocystinemia
Hyperviscosity
Thrombocytopenic thrombotic purpura
pathogenesis of pulmonary embolism and Thrombi found in veins when blood flow is
are known as Virchows triad.18,19 reduced are composed predominantly of
fibrin and entrapped blood cells with rela-
tively few platelets and are often termed red
Lower Limb thrombi. The friable ends of these thrombi
Venous stasis is the most important feature are the source of the material that eventu-
predisposing to venous thrombosis. The ally becomes pulmonary emboli. Formation
venous sinuses of the veins are especially of venous thrombi is typically asymptomatic
vulnerable to stasis and thrombosis. Propa- and may involve the superficial or deep ve-
gation of the thrombus may then follow up- nous systems. Deep venous thrombi can
stream, or the process may spread retrograde. propagate into the superficial system.

Table 3.2 RiskStratification
LowRisk Uncomplicated minor surgery

Age 40 years
No risk factors
ModerateRisk Age 4060 years and no additional risk factors

Age 40 years, major surgery, and no additional risk factors
Minor surgery with additional risk factors
HighRisk Age >60 years and major surgery, with no additional risk factors
Age 4060 years and major surgery with additional risk factors, patients with
myocardial infarction, and medical patients with risk factors
HighestRisk Age >40 years and major surgery, plus prior thromboembolism, malignancy, or
hypercoagulable state
Major lower extremity orthopedic surgery
Hip fracture
Stroke
Multiple trauma
Spinal cord injury
shown that clinically silent thrombosis may

Upper Limb occur in 20% to 30% of patients after cen-
Primary axillosubclavian venous thrombosis tral venous catheter insertion.12,13 Although
is related to chronic venous compression and it has not been clearly delineated, there are
stenosis in the axillary-subclavian vein at the several inciting possibilities for secondary
level of the costoclavicular space. It is seen axillosubclavian vein thrombosis; they in-
most commonly in young individuals fol- clude endothelial trauma during insertion
lowing vigorous exercise or activity involving of the catheter, trauma from the indwelling
hyperabduction of the affected extremity.20,21 catheter, venous stasis and fibrin deposition
In the most common form of the disease, the upon the catheter surface, and the effects of
vein is compressed between a hypertrophied infusates on the surrounding vessel. Venous
scalene tendon and the first rib. Secondary stenosis occurs in 30% to 40% of patients
axillosubclavian venous thrombosis is gener- with central venous catheters.22,23
ally associated with instrumentation of the
central venous circulation, direct venous
trauma, or the presence of a hypercoagulable Natural History
disorder. Secondary axillosubclavian venous
thrombosis is an important clinical issue In isolated calf thrombosis, proximal prop-
that requires attention. Iatrogenic causes of agation occurs in up to 23% of untreated
secondary venous thrombosis account for patients and 10% of patients treated with
up to 30% of symptomatic subclavian ve- intravenous heparin. Propagation to a new
nous thromboses.11 In addition, studies have segment has been documented in 30% of

initially involved limbs and rethrombosis of screening method of choice for the nonin-
partially occluded or recanalized segments vasive assessment of blood flow in leg veins
in 31% of extremities. After completion of and of valve cusp movement.33 It also can dif-
a course of anticoagulation, the theoreti- ferentiate between acute and chronic venous
cal rate of recurrence is 0.9% per month.24 thrombosis. All major deep veins of the lower
Following venous thrombosis, the venous limb can be assessed, but it cannot exclude
lumen is most often reestablished. Recanali- the presence of thrombi in small veins of
zation occurs in an exponential manner over the calf. With experience, ultrasonography
6 months.25 Up to 40% of occluded segments is accurate, repeatable, and inexpensive.34
recanalized within 7 days, while 100% re- Positive tests indicating above-knee DVT
canalization occurs within 90 days.26 It ap- do not require further follow-up unless the
pears that venous valvular incompetence condition of the leg significantly worsens.
occurs after deep venous thrombosis. This Below-knee DVT requires repeat scanning
is supported by natural history studies that and is recommended to confirm that there
have demonstrated a correlation between is no antegrade propagation on therapy. Du-
segment thrombosis and subsequent valvu- plex ultrasonography sensitivity and specific-
lar incompetence.27 The development of re- ity are 94% and 96%, respectively. Overall
flux is highest within the first 6 to 12 months accuracy is considered to be 95%. Cross-
after development of a DVT.28 However, up sectional imaging with computed tomogra-
to 30% of segments will develop reflux with- phy (CT) or magnetic resonance imaging
out evidence of an initial thrombus, and this (MRI) is a reliable method of diagnosing
phenomenon appears to be related to persis- venous thrombosis, especially in the pelvic
tent proximal obstruction.29 The interval be- veins.35 Venography is recommended to vali-
tween the development of these defects and date and diagnose a DVT when duplex ul-
symptoms and signs of chronic venous insuf- trasonography has failed to rule out a DVT
ficiency may range out to 20 years.30 and should be performed prior to iliofemo-
The clinical diagnosis of DVT can be ral thrombolysis or thrombectomy. A recent
unreliable, as the disease mimics the pat- review of the evidence strongly supports the
terns of many other disorders.18,31 More than use of clinical prediction rules, particularly
half of the patients who present with the the Wells model, for establishing the pretest
classical symptoms of DVT do not have the probability of DVT or pulmonary embolism
disease.18 Clinical suspicion in combination in a patient before ordering more definitive
with diagnostic imaging should be used to testing.36 Fifteen studies support the conclu-
confirm the diagnosis.32 The proportion of sion that when a D-dimer assay is negative
patients with clinically suspected DVT, in and a clinical prediction rule suggests a low
whom the diagnosis is confirmed by objec- probability of DVT or pulmonary embolism,
tive testing, increases with the number of the negative predictive value is high enough
risk factors identified. Approximately half to justify foregoing imaging studies in many
of the patients with deep venous thrombo- patients. The evidence in 5 systematic re-
sis who develop pulmonary embolism have views regarding the use of D-dimer, in isola-
no symptoms of deep venous disease. This tion, is strong and demonstrates sensitivities
causes a delay in the administration of ap- of the enzyme-linked immunosorbent assay
propriate prophylactic and therapeutic mea- (ELISA) and quantitative rapid ELISA,
sures. Duplex ultrasonography is a popular pooled across studies, of approximately 95%.

Eight systematic reviews found that the sen-

sitivity and specificity of ultrasonography for
Lower Extremity DVT
diagnosis of DVT vary by vein; ultrasonogra- American College of Chest Physicians Ev-
phy performs best for diagnosis of symptom- idence-Based Clinical Practice Guidelines
atic, proximal vein thrombosis, with pooled (8th edition) on the treatment for venous
sensitivities of 89% to 96%. The sensitivity thromboembolic disease recommend anti-
of single-detector helical computed tomog- coagulant therapy with subcutaneous (SC)
raphy for diagnosis of pulmonary embolism low-molecular-weight heparin (LMWH),
varied widely across studies and was below monitored intravenous or subcutaneous un-
90% in 4 of 9 studies; more studies are need- fractionated heparin (UFH), unmonitored
ed to determine the sensitivity of multidetec- weight-based SC UFH, or SC fondaparinux
tor scanners.36 followed by VKA.38 For patients with a high
clinical suspicion of DVT, treatment with
anticoagulants while awaiting the outcome
Management of diagnostic tests is recommended. In acute
DVT, initial treatment with LMWH, UFH,
Prevention or fondaparinux for at least 5 days prior to
The American College of Chest Physicians initiation of VKAs and concomitant admin-
(ACCP) has recently provided the following istration of LMWH, UFH, or fondaparinux
guidelines for DVT prophylaxis.37 In pa- on the first treatment day, and discontinu-
tients admitted to the hospital with an acute ation of these heparin preparations when
medical illness, major trauma, and spinal the INR is 2 for at least 24 hours is rec-
cord injury, thromboprophylaxis with a low- ommended. For patients with DVT second-
molecular-weight heparin (LMWH), low- ary to a transient (reversible) risk factor, it is
dose unfractionated heparin (LDUH), or recommended that treatment with a VKA be
fondaparinux is recommended with grade for 3 months. For patients with unprovoked
1A weighting. For patients undergoing DVT, the ACCP recommends treatment
major general, gynecologic, and urological with a VKA for at least 3 months and that
surgery, thromboprophylaxis with LMWH, all patients are then evaluated for the risks to
LDUH, or fondaparinux is recommended. benefits of indefinite therapy. Indefinite an-
However, for patients undergoing elective ticoagulant therapy is recommended for pa-
hip or knee arthroplasty and hip fracture sur- tients with a first unprovoked proximal DVT
gery, an alternative regimen of therapy with or PE and a low risk of bleeding, when this
a vitamin K antagonist (VKA) with a goal is consistent with the patients preference,
of an international normalized ratio (INR) and for most patients with a second unpro-
target of 2.5 (range, 2.03.0) is recommend- voked DVT. The dose of VKA should be ad-
ed for 10 to 35 days. All major trauma and justed to maintain a target INR of 2.5 (INR
all spinal cord injury (SCI) patients receive range, 2.03.0) for all treatment durations.
thromboprophylaxis. Mechanical methods The ACCP recommends at least 3 months
of thromboprophylaxis are recommended of treatment with LMWH for patients with
primarily for patients at high bleeding risk venous thromboembolism (VTE) and can-
or possibly as an adjunct to anticoagulant cer, followed by treatment with LMWH or
thromboprophylaxis. VKA as long as the cancer is active. Treat-
ment with VKA for more than 3 months is

indicated for patients with recurrent venous

thromboembolism or in patients in whom
Decompression of Thoracic Outlet
there is a continuing risk factor for venous Acute upper extremity DVT can be due to
thromboembolism. Patients should be fol- thoracic outlet syndrome. The frequency of
lowed up for 3 years to ensure that there is PTS after UEDVT ranges from 7% to 46%
no recurrent thromboembolic event. Dis- (weighted mean 15%). Residual thrombosis
continuation of anticoagulant therapy in and axillosubclavian vein thrombosis appear
patients with recurrent venous thromboem- to be associated with an increased risk of
bolism is associated with an approximate PTS, whereas catheter-associated UEDVT
20% risk of recurrent venous thromboem- may be associated with a decreased risk.
bolism during the following year and a There is currently no validated, standard-
5% risk of fatal pulmonary embolism.39,40 ized scale to assess upper extremity PTS, and
In patients with a continuing risk factor, little consensus regarding the optimal man-
which is reversible, long-term therapy should agement of this condition. Quality of life is
usually be continued until the risk factor impaired in patients with upper extremity
is reversed. Anticoagulant therapy should PTS, especially after DVT of the dominant
be continued indefinitely in patients with arm.41 These patients often present with ob-
an irreversible risk factor. For prevention of structive symptoms and if acute thrombosis
postthrombotic syndrome (PTS) after proxi- is noted, common practice is to perform
mal DVT, the use of an elastic compression catheter-directed thrombolysis followed by
stocking is recommended. For DVT of the resection of the first rib.42 Decompression
upper extremity, we recommend similar of the thoracic outlet requires resection of
treatment as for DVT of the leg. Selected either the clavicle and/or the first rib with
patients with lower extremity and upper division of the scalene muscle fibers at their
extremity DVT may be considered for insertion; division of the subclavius tendon
thrombus removal, generally using catheter- and local venolysis must supplement each
based thrombolytic techniques. For exten- of these procedures.43 First rib resection has
sive superficial vein thrombosis, treatment most commonly been accomplished from
with prophylactic or intermediate doses of a transaxillary approach but may be per-
LMWH or intermediate doses of UFH for formed by supraclavicular, transclavicular,
4 weeks is recommended. and infraclavicular approaches.41 If venous
repair is considered a possibility, then either
supraclavicular anterior clavicular or infra-
Upper Extremity DVT clavicular approaches are required.
For patients with acute upper extremity
DVT (UEDVT), treatment is as described
for lower extremity DVT including treat- Pulmonary Embolism
ment with a VKA for >3 months. If the DVT For patients with a high clinical suspicion
is due to a central venous catheter and the of pulmonary embolism (PE), treatment
catheter is removed, then the duration of an- with anticoagulants while awaiting the out-
ticoagulation can be reduced for 3 months. come of diagnostic tests is recommended.
Except in unusual circumstances, routine Patients with confirmed PE, and who are
use of catheter-directed thrombolytic thera- hemodynamically compromised, should be
py is not recommended. considered for a short-course thrombolytic

therapy.38 For those with nonmassive PE, the

use of thrombolytic therapy should also be
Conclusion
considered. In acute PE, initial treatment Management of DVT has progressed signifi-
with LMWH, UFH, or fondaparinux for cantly in the last decade with the recognition
at least 5 days is recommended with subse- of the need for aggressive DVT prophylaxis,
quent initiation of vitamin K antagonists to- improvements in diagnosis, and the use of
gether with LMWH, UFH, or fondaparinux more aggressive standards on initial and
on the first treatment day, and discontinua- chronic therapy for the diagnosis of DVT.
tion of these heparin preparations when the
INR is 2 for at least 24 hours. For patients references
with PE secondary to a transient (revers- 1. White RH. The epidemiology of venous
ible) risk factor, therapy with a VKA should thromboembolism. Circulation. 2003;107:
continue for 3 months. For patients with I-4-I-8.
an unprovoked PE, it is recommended that 2. van Korlaar IM, Vossen CY, Rosendaal FR,
therapy with a VKA should continue for at Bovill EG, Cushman M, Naud S, et al. The
least 3 months and that all patients are then impact of venous thrombosis on quality of life.
evaluated for the risks to benefits of indefi- Thromb Res. 2004;114:11-8.
nite therapy. The dose of VKA should be ad- 3. Cushman M, Tsai AW, White RH, Heckbert
justed to maintain a target INR of 2.5 (INR SR, Rosamond WD, Enright P, et al. Deep
range, 2.03.0) for all treatment durations. vein thrombosis and pulmonary embolism in
two cohorts: the longitudinal investigation
of thromboembolism etiology. Am J Med.
Thrombolysis 2004;117:19-25.
Percutaneous clot removal using thromboly- 4. Heit JA, Silverstein MD, Mohr DN, Petterson
sis, mechanical thrombectomy, or a combi- TM, OFallon WM, Melton LJ. Predictors
nation of the 2 is fast becoming a treatment of survival after deep vein thrombosis and
of choice for patients presenting with acute pulmonary embolism: a population-based
iliofemoral and axillosubclavian deep venous cohort study. Arch Intern Med. 1999;159:
thrombosis.42,44 By restoring venous patency 445-53.
and preserving valvular function, catheter- 5. Silverstein M, Heit JA, Mohr D, Petterson T,
directed thrombolytic therapy potentially OFallon W, Melton L. Trends in the incidence
affords an improved long-term outcome in of deep vein thrombosis and pulmonary
selected patients with DVT. In selected pa- embolism: a 25-year population-based study.
tients with extensive acute proximal DVT Arch Intern Med. 1998;158:585-93.
(ie, iliofemoral DVT, symptoms for <14 days, 6. Tsai AW, Cushman M, Rosamond WD,
good functional status, life expectancy of >1 Heckbert SR, Polak JF, Folsom AR.
year) who have a low risk of bleeding, cath- Cardiovascular risk factors and venous
eter-directed pharmacomechanical throm- thromboembolism incidence: the longitudinal
bolysis is suggested by the ACCP.38 This investigation of thromboembolism etiology.
percutaneous therapy should include correc- Arch Intern Med. 2002;162:1182-9.
tion of any underlying venous lesions. 7. Dalen JE, Alpert JS. Natural history of
pulmonary embolism. Prog Cardiovasc Dis.
1975;17:257-70.
8. Dismuke SE, Wagner EH. Pulmonary

embolism as a cause of death; the changing 6th ed. New York: McGraw-Hill; 1989. p2207-
mortality in hospitalized patients. JAMA. 18.
1986;255:2039-42. 18. Anderson FA Jr, Wheeler HB, Goldberg
9. Salzman EW, Hirsh J. The epidemiology, RJ, Hosmer DW, Patwardhan NA, Jovanovic
pathogenosis and natural history of venous B, et al. A population-based perspective of the
thromboembolism. In: Colman RW, Hirsh hospital incidence and case fatality
J, Marder VJ, Salzman EW, eds. Hemostasis rates of deep venous thrombosisand
and Thrombosis. Philadelphia: JP Lippincott; pulmonary embolism. The Worcester
1994. p1275-96. DVT Study. Arch Intern Med. 1991;151:
10. Clagett GP, Anderson FA Jr, Levine MN, 933-8.
Salzman EW, Wheeler HB. Prevention 19. Anderson FAJ, Wheeler HB. Physician
of venous thromboembolism. Chest. practices in the management of venous
1992;102:391S-407S. thromboembolism: a community-wide survey.
11. Hill S, Berry R. Subclavian vein thrombosis: a J Vasc Surg. 1992;15:707-14.
continuing challenge. Surgery. 1990;108:1-9. 20. Hughes E. Venous obstruction in the upper
12. Smith V, Hallet J. Subclavian vein thrombosis extremity (Paget Schroetter Syndrome). Int
during prolonged catheterization for Abst Surg. 1949;88:89-127.
parenteral nutrition: early management 21. Adams J, DeWeese J. Effort thrombosis of
and long term follow-up. South Med J. the axillary and subclavian veins. J Trauma.
1983;76:603-6. 1971;11:923-30.
13. Fraschini G, Jadeja J, Lawson M, Holmes 22. Lee S, Neilberger R. Subclavian vein
FA, Carrasco HC, Wallace S. Local infusion stenosis: complication of subclavian vein
of urokinase for the lysis of thrombosis catheterization for hemodialysis. Child
associated with permanent central venous Nephrol Urol. 1991;11:212-4.
catheters in cancer patients. J Clin Oncol. 23. Schillinger F, Schillinger D, Montagnac R,
1987;5:672-8. Milcent T. Post-catheterization venous stenosis
14. Cosgriff TM, Bishop DT, Hershgold EJ, in hemodialysis: comparative angiographic
Skolnick MH, Martin BA, Baty BJ, et al. study of 50 subclavian and 50 jugular accesses.
Familial antithrombin III deficiency: its Nephrology. 1992;13:127-33.
natural history, genetics, diagnosis and 24. Meissner MH, Caps MT, Bergelin RO,
treatment. Medicine. 1983;62:209-20. Manzo RA, Strandness DE. Propagation,
15. Comp PC, Esmon CT. Recurrent venous rethrombosis and new thrombus formation
thromboembolism in patients with a partial after acute deep venous thrombosis. J Vasc
deficiency of protein S. N Engl J Med. Surg. 1995;22:558-67.
1984;311:1525-8. 25. VanRamshorst B, VanBemmelen PS,
16. Ridkar PM, Hennekens CH, Lindpaintner Honeveld H, Faber JAJ, Eikelbloom BC.
K, Stampfer MJ, Eisenberg PR, Miletich JP. Thrombus regression in deep venous
Mutation of the gene coding for coagulation thrombosis. Quantification of spontaneous
factor V and the risk of myocardial infarction, thrombolysis with duplex scanning.
stroke and venous thrombosis in apparently Circulation. 1992;86:414-9.
healthy men. N Engl J Med. 1995;332:912-7. 26. Killewich LA, Bedford GR, Beach KW,
17. Tollefsen DM. Antithrombotic deficiency. In: Strandness DE. Spontaneous lysis of deep
Scriver CR, Beaudet AL, Sly WS, Valle D, venous thrombi: rate and outcome. J Vasc
eds. The metabolic basis of inherited disease. Surg. 1989;9:89-97.

27. Meissner MH, Manzo RA, Bergelin RO, of venous thromboembolism: American
Markel A, Strandness DE Jr. Deep venous College of Chest Physicians Evidence-Based
insufficiency: the relationship between Clinical Practice Guidelines (8th ed.). Chest.
lysis and subsequent reflux. J Vasc Surg. 2008;133: 381S-453S.
1993;18:596-608. 38. Kearon C, Kahn SR, Agnelli G, Goldhaber
28. Markel A, Manzo R, Bergelin RO, Strandness S, Raskob GE, Comerota AJ. Antithrombotic
DE Jr. Valvular reflux after deep vein therapy for venous thromboembolic disease:
thrombosis: incidence and time of occurance. American College of Chest Physicians
J Vasc Surg. 1992;15:377-87. Evidence-Based Clinical Practice Guidelines
29. Caps MT, Meissner MH, Manzo R, Bergelin (8th ed.). Chest. 2008;133(6 suppl):
RO, Strandness DE. Venous valvular reflux 454S-545S.
in veins not involved at the time of acute deep 39. Schulman S, Granqvist S, Holmstrm M,
vein thrombosis. J Vasc Surg. 1995;22:524-31. Carlsson A, Lindmarker P, Nicol P, et al. The
30. Bulger CM, Jacobs C, Patel NH. duration of oral anticoagulant therapy after a
Epidemiology of acute deep vein thrombosis. second epsiode of venous thromboembolism.
Tech Vasc Interv Radiol. 2004;7(2):50-4. N Engl J Med. 1997;336:393-8.
31. Browse N. Deep vein thrombosis. Diagnosis. 40. Hyers TM, Hull RD, Weg JG. Antithrombotic
Brit J Med 1969;4:676-8. therapy for venous thromboembolic disease.
32. Hull R, Hirsh J, Sackett DL, Stoddard G. Cost Chest. 1995;108(suppl 4):335-51.
effectiveness of clinical diagnosis, venography 41. Elman EE, Kahn SR. The post-thrombotic
and non-invasive testing in patients with syndrome after upper extremity deep venous
symptomatic deep vein thrombosis. N Engl J thrombosis in adults: a systematic review.
Med. 1986;314:823-8. Thromb Res. 2006;117(8):609-14.
33. Appleman PT, DeJong TE, Lampmann 42. Molina JE, Hunter DW, Dietz CA.
LE. Deep venous thrombosis of the leg: US Protocols for Paget-Schroetter syndrome
findings. Radiology. 1987;163:743-6. and late treatment of chronic subclavian
34. Cronan JJ. Venous thromboembolic vein obstruction. Ann Thorac Surg.
disease. The role of ultrasound. Radiology. 2009;87(2):416-22.
1993;186:619-30. 43. Melby SJ, Vedantham S, Narra VR, Paletta
35. Erdman WA, Jayson HT, Redman HC, GAJ, Khoo-Summers L, Driskill M, et al.
Miller GL, Parkey RW, Peshock RW. Deep Comprehensive surgical management of
venous thrombosis of the extremities: role the competitive athlete with effort thrombosis
of MR imaging in the diagnosis. Radiology. of the subclavian vein (Paget-Schroetter
1990;174:425-31. syndrome). J Vasc Surg. 2008;47(4):
36. Segal JB, Eng J, Tamariz LJ, Bass EB. Review 809-21.
of the evidence on diagnosis of deep venous 44. Meissner MH, Wakefield TW, Ascher E,
thrombosis and pulmonary embolism. Ann Caprini JA, Comerota AJ, Eklof B, et al. Acute
Fam Med. 2007;5(1):63-73. venous disease: venous thrombosis and venous
37. Geerts WH, Bergqvist D, Pineo GF, Heit JA, trauma. J Vasc Surg. 2007;46(suppl S):
Samama CM, Lassen MR, et al. Prevention 25S-53S.

CHAPTER
Diagnostic Approach
to Venous Thromboembolism
l. Bernardo Menajovsky,
Patricia Hightower lambden, and ruth l. Bush
V enous thromboembolism (VTE) re-

fers to pathologic thrombosis that
occurs within the venous circulation. The
While mortality rates due to VTE have
declined substantially over the last few de-
cades as a result of advances in diagnostic
most common form of VTE is deep venous techniques and treatments and a better un-
thrombosis (DVT) of the lower extremities; derstanding of the disease,2 VTE and its com-
however, the most life-threatening mani- plications remain a common cause of death
festation of VTE is embolization of venous in the United States. An estimated 200,000
thrombi resulting in pulmonary embolism new cases of VTE occur in the United States
(PE). every year, including 94,000 with PE.3 This
VTE is often a silent disease that can translates into an incidence of 23 newly di-
lead to multiple complications when left un- agnosed cases per 100,000 patients per year.3
detected or inadequately treated. Potential Without treatment, PE is associated with a
complications of DVT include PE, death, mortality rate of approximately 30%, or near-
postthrombotic syndrome (PTS), and pul- ly 300,000 deaths per year.3
monary hypertension. PTS is characterized VTE is generally triggered by a com-
by signs and symptoms similar to those asso- bination of environmental and inherited
ciated with DVT and is common in patients risk factors. Accurate assessment of clinical
with VTE (~30%).1 symptoms, risk factor stratification, and ap-
propriate use of objective diagnostic tests
are pivotal in the accurate diagnosis and
treatment of VTE. The reference standard
Davies md and Alan B. Lumsden md, eds. for VTE diagnosis remains clot visualiza-
Cardiotext Publishing, ISBN 978-1-935395-22-5. tion with contrast venography or pulmonary
43

angiography. However, the invasiveness and factors, including acute infectious disease,
the risks of these modalities have led to a previous history of VTE, or advanced age
steady increase in the use of noninvasive (>75 years).6
or minimally invasive VTE testing. These Endothelial injury has been attributed
tests should optimally be used after clinical to both overt as well as subtle insults. Direct
examinations and risk assessments reveal re- trauma such as soft tissue injury or fracture,
sults highly suggestive of VTE. This chapter as well as intravenous infusion of irritating
discusses the clinical symptoms and signs substances, are more obvious factors that
indicative of possible VTE and provides an cause endothelial injury. Chemical chang-
overview of the clinical models and diagnos- es, ischemia, anoxia, and inflammation are
tic strategies available to assess for thrombo- more subtle factors that may cause endothe-
embolic disease. lial tissue damage.
In theory, during episodes of immobi-
lization or prolonged limb dependency, ve-
nous flow becomes sluggish, which may give
Pathophysiology rise to platelet and fibrin deposit, thereby
and Risk Stratification initiating venous thrombosis development.
However, results from some studies sug-
VTE can occur as an idiopathic syndrome gest that the associated risk may not be as
or may be caused by an underlying condition significant.3,7,8 In one small study, Gatt and
that predisposes a patient to thrombosis. In a colleagues observed no difference in the risk
retrospective study of 366 validated cases of of VTE in immobilized vs. mobile patients.7
VTE, approximately 48% were idiopathic In a retrospective cohort analysis, Gatt and
and 52% were secondary to an underlying colleagues evaluated 18 mobile and 8 im-
condition.4 Rudolf Ludwig Karl Virchow was mobile patients with a mean age of 85 for a
the first to identify 3 primary clinical factors duration of 10 years. The immobile patients
associated with a substantial risk of throm- were bedridden for a prolonged period (>3
bosis. Together, these factors are known as months). No difference in baseline char-
Virchows triad, and include (1) vessel wall acteristics, which included the assessment
damage due to inflammation or trauma; (2) of risk factors, was observed between the 2
changes in blood flow or volume due to im- groups. The incidence of VTE was similar
mobility, ischemia, and other conditions; between the immobile and mobile patient
and (3) hypercoagulable factors present in groups (13.9 and 15.8, respectively; P = 0.77).
the blood, including inherited and acquired While these results are not consistent with
coagulation disorders.5 previous studies9,10an inconsistency that is
The consideration of risk factors in due, in part, to a relatively small study popu-
the assessment and diagnostic evaluation of lationthey do highlight the importance of
patients with potential DVT is important; considering the added threat conferred by
however, it is also important to consider the each risk factor both independently and in
relative risk for each factor independently. combination with other risk factors.
Some reports suggest that hospitalized Similarly, an analysis that was conduct-
medical patients subjected to an extended ed using data from the American College
period of immobility may be at risk for VTE, of Surgeons National Trauma Data Bank
especially if presenting with additional risk evaluated the frequency of VTE follow-

Diagnostic Approach to Venous Thromboembolism 45
ing trauma. The results demonstrated that any force that creates an imbalance between
incidence of VTE in these patients is also platelets, clotting factors, and the fibrinolytic
relatively low.11 Data were collected from 131 system. There are many variables that con-
trauma centers. The following 6 risk factors tribute to hypercoagulability; however, the
were found to be associated with VTE: aged most common causes are hematologic disor-
40 years, lower extremity fracture with Ab- ders, malignancy, trauma, estrogen therapy,
breviated Injury Score (AIS) 3, head injury and surgical events.
with AIS 3, ventilator delays >3, venous in-
jury, and the performance of a major opera-
tive procedure. Of the 450,375 patients who
experienced trauma from 1994 to 2001, a Clinical Diagnosis
total of 1602 experienced DVT, resulting in
an incidence of 0.36%.11 of Venous
In contrast to the lower relative risk
conferred by immobilization and trauma,
Thromboembolism
a much stronger association exists between The vast majority of thrombi that originate
VTE and cancer.12 In one study, 26% of pa- within the lower extremities resolve sponta-
tients presenting with bilateral DVT were di- neously; however, propagation of fibrin net-
agnosed with cancer after the occurrence of works may lead to venous thrombosis. Such
DVT, and metastasis had occurred in 70% of thromboembolic events are classified based
these patients.12 Other reports cite as much upon the system involved (deep or superfi-
as a 2-fold increase in DVT risk among pa- cial) as well as by location. Classification of
tients with cancer and suggest that the risk thromboembolic events with regard to loca-
of recurrent thromboembolism is as much as tion refers to the thrombus in relation to the
3.5 times higher in patients with malignancy popliteal vein (distal or proximal).
vs. cancer-free patients.1,13,14 The deep venous system possesses the
Finally, hypercoagulabilityespecially ability to encompass a rather large amount
hypercoagulability due to a genetic predispo- of systemic flow. This heightened volume
sition, acquired syndromes, and certain med- capacity, as well as ease of effective collat-
ications, such as oral contraceptivescan eralization, allows many thromboembolic
increase the risk of thrombosis.15,16 Hereditary events to remain asymptomatic and undiag-
risk factors include factor V Leiden mutation; nosed. Unfortunately, the insidious nature
prothrombin G 20210A gene mutation; and of DVT allows for a great deal of vessel de-
deficiencies of protein C, protein S, and an- struction prior to diagnosis. In many cases,
tithrombin III.17 Hyperhomocysteinemia and the severity of symptomatology ineffectively
elevated levels of factors XIII and V, which represents the extent of the disease.
may be hereditary and/or acquired, are also It is important to note that 50% of pa-
risk factors.17 Additionally, ABO blood type tients who have VTE do not present with
is another VTE risk factor that was recently any symptoms.19 Clinical presentation of
noted in cancer patients and is believed to DVT most frequently involves unilateral
be associated with hypercoagulability result- lower extremity edema; however, iliofemoral
ing from influence on the levels of von Wil- obstruction may elicit bilateral edema. In-
lebrand factor (vWF) and factor VIII.18 creased intravascular blood volume, elevat-
Hypercogulability can be potentiated by ed hydrostatic pressure, and seepage of fluid

across the capillary membrane all produce DVT include immobility, dehydration, ma-
DVT-related edema. Increased tissue turgor lignancy, blood dyscrasias, estrogen therapy,
and elevated temperature of the affected postoperative status, trauma, pregnancy/
area are also secondary to edema and dila- postpartum state, congestive heart failure,
tion of superficial vessels. Mottling or cyano- and obesity.
sis of the extremity is often related to venous Wells and colleagues developed the first
stasis, increased oxygen consumption, and clinical model for the diagnosis of patients
reduction of hemoglobin. presenting with suspected DVT.20-26 This
Most individuals suffering from DVT model includes a thorough clinical exami-
report limb pain, described as aching or nation and identification of any risk factors
throbbing in nature, which is aggravated by that predispose patients to have increased
ambulation. Although considered somewhat risk of thrombosis. In accordance with this
unreliable, there are 2 techniques utilized to model, patients are first divided into 3 risk
assess limb tenderness. Homans sign entails categories (low, moderate, or high) and are
the dorsiflexion of the foot of the affected ex- further assessed through ultrasonography.23
tremity to elicit calf pain. With Lowenbergs Patients who are stratified to the high-risk
sign, calf pain is elicited by cuff inflation category, or who have abnormal ultrasonog-
over the affected limb. As mentioned previ- raphy results, are further assessed through
ously, each technique possesses a low pretest venography.23 Clinical practice guidelines
probability. for the diagnosis of DVT from the American
Long-term disability related to venous Thoracic Society concur with this strategy,
thrombosis involves the destruction of vessel recommending the use of venography as a
endothelium, which renders venous valves follow-up to inconclusive compression ul-
incompetent. Such valvular incompetence trasonography results, and the use of serial
allows for reversal of venous flow, stasis, and ultrasonography or impedance plethysmog-
recurrent thrombosis. Venous thrombosis raphy in patients with normal compression
most frequently involves the vessels within ultrasonography results.27
the calf; however, the proximal veins of the For patients presenting with PE, short-
lower extremities and the pelvic vessels are ness of breath, with or without leg pain, may
the primary etiology of pulmonary embolus, be the first symptom; however, a number of
the most common life-threatening compli- specific criteria allow for a more accurate
cation of deep venous thrombosis. Because diagnosis. Similar to the clinical model for
of the correlation of life-threatening pulmo- diagnosis of DVT, Wells and colleagues also
nary embolus with the presence of DVT, it defined a clinical algorithm for the diagno-
is of extreme importance to assess for pul- sis of PE (Figure 4.1),26 which when used
monary embolus in conjunction with DVT in conjunction with D-dimer testing, safely
evaluation. reduces the need for expensive imaging di-
Initial evaluation for DVT must include agnostics.26 This model for assessment of PE
screening for risk factors. There are many uses a point system for calculating the low,
common risk factors; however, any event moderate, or high pretest probability of PE.
that leads to the elements included within Points are assigned based on clinical symp-
Virchows triad (venous stasis, endothelial toms of DVT, including heart rate of >100
injury, and hypercoagulability) may lead beats per minute, immobilization for 3
to DVT. The most common risk factors for days or recent surgery in the past 4 weeks,

Figure 4.1 Clinical

algorithm for initial evaluation
of patients with suspected
pulmonary embolism. DVT
= deep venous thrombosis;
PTP = pretest probability;
V/Q = ventilation-perfusion.
Reprinted with permission
from Wells PS, Anderson
DR, Rodger M, Stiell I,
Dreyer JF, Barnes D, et
al. Excluding pulmonary
embolism at the bedside
without diagnostic imaging:
management of patients
with suspected pulmonary
embolism presenting to the
emergency department by
using a simple clinical model
and D-dimer. Ann Intern Med.
2001;135:98-107.
a clinical history of VTE, hemoptysis, ma- reason, it is especially important to conduct

lignancy, or the clinician determination that thorough examinations and risk stratifica-
PE is as likely or more likely than another tion when examining patients for potential
diagnosis.26 VTE.
The patient history and physical exami-
nation findings reported in the Prospective
Investigation of Pulmonary Embolism Diag-
nosis (PIOPED) trial illustrate the difficulty Objective Testing
in quickly identifying or ruling out a diag-
nosis of PE. In PIOPED, the most common for Venous
past and current physical findings included
dyspnea, pleuritic chest pain, cough, tachy-
Thromboembolism
cardia, and tachypnea.28 These symptoms Diagnostic evaluation of suspected VTE
also can be indicative of heart failure, inter- includes a clear correlation between clini-
stitial lung disease, or pneumonia. For this cal probability, test selection, and test inter-

pretation.29 However, a variety of diagnostic Compression Ultrasonography

approaches are feasible, and availability
and familiarity with particular technology Doppler compression ultrasonography with
may influence the choice of approach. Ad- real-time, B-mode imaging is employed at
ditionally, the sensitivity of certain diagnos- most institutions because of its safety, avail-
tic tests is affected by the location of the ability, reliability, and noninvasive nature.
thrombus.27 In addition to traditional tests, Benefits include detection of acute symp-
such as contrast venography for DVT and tomatic proximal DVT, as well as DVT of
pulmonary angiography for PE, newer mo- the upper extremities, and it is also capable
dalities such as D-dimer assays and mag- of identifying other pathologies.23,27,29-31 Its
netic resonance direct thrombus imaging 2-dimensional, cross-sectional representa-
(MRDTI) offer promise for better detection of lower extremity veins is also useful
tion with less invasiveness and have the in combination with venous flow detection
potential for use in detection of both DVT (duplex ultrasonography).27 However, com-
and PE. pression ultrasonography is not specific or
sensitive for the detection of DVT in patients
with asymptomatic proximal DVT or in pa-
Imaging Modalities for tients with symptomatic or asymptomatic
DVT of the calf, and it demonstrates limited
Deep Venous Thrombosis accuracy in chronic DVT cases.27 Its use is
also limited in patients who are obese or
Contrast Venography Imaging who have edema.27 Currently, a number of
ongoing large trials are in progress for the as-
Contrast venography is no longer appropri- sessment of magnetic resonance venography
ate as the initial diagnostic test in patients and computerized tomographic venography
exhibiting DVT symptoms, although it re- in the diagnosis of DVT.
mains the gold standard for confirmatory di-
agnosis of DVT. Venography is nearly 100%
specific and sensitive, and provides the abil- Diagnostic Tests for
ity to investigate the distal and proximal
venous system for thrombosis. Its use is no Pulmonary Embolism
longer widespread due to the need for ad-
ministration of a contrast medium and the Blood Gas Analysis and
increased availability of noninvasive diag-
Electrocardiogram
nostic strategies.27 However, venography is
still warranted when noninvasive testing is Arterial blood gas analysis and electrocardio-
inconclusive or impossible to perform.27 Ad- gram (ECG) are routinely used to diagnose
ditional drawbacks of venography include PE with varying rates of success. Hypoxemia
contraindication in patients with renal in- is a common feature of acute PE but is not
sufficiency and lack of accuracy in recur- present in all cases.27 Thus, while arterial
ring cases of suspected DVT due to the blood gas levels reveal the blood oxygen sat-
difficulty of visualizing an intralumenal uration level, they are not specific or sensi-
defect in veins that have been thrombosed tive for the definitive diagnosis of PE. The
previously.27 use of transcutaneous oximetry will provide

Figure 4.2 The S1Q3T3 pattern with concomitant

symptoms and signs of pulmonary embolism warrants
further investigation.
Figure 4.3 Hamptons hump is a classic finding

information regarding hypoxemia without caused by a pleural-based abnormality due to
pulmonary infarction and visible in some X-rays. Its
the need for an arterial puncture, but it is not
presence is not common and should not be used to
sensitive or specific for the diagnosis of PE.
confirm or exclude pulmonary embolism.
Hemodynamically significant PE in-
duces transient ECG abnormalities reflect-
ing right ventricular overload and/or strain32;
however, it is neither sensitive nor specific to Although chest X-ray is commonly ordered
VTE,33 although a classic S1Q3T3 pattern during the process of differential diagnosis
(Figure 4.2) might warrant consideration of of pulmonary conditions, PE patients most
PE in the presence of other signs and symp- commonly have normal chest X-ray results
toms. ECG can also be used to suggest an but sometimes present with nonspecific ra-
alternative cardiac diagnosis. Additionally, diographic findings.27 A normal chest X-ray
recent investigations have suggested that a in the presence of severe dyspnea or hypox-
simple scoring system based on ECG might emia without evidence of bronchospasm or
be useful in predicting individuals with the cardiac shunt is strongly suggestive of, but
greatest percentage of perfusion defect.34 Be- not diagnostic for, PE.27
cause neither of these methods is suggested Hamptons hump is a classic finding
as a proven diagnostic tool for the initial caused by a pleural-based abnormality due
screening or exclusion of VTE, they should to pulmonary infarction and is visible in
not be routinely used for definitive diagnosis. some X-rays; its presence, however, is not
Instead, they should be used in conjunction common and cannot be used to confirm or
with clinical examinations and other diag- exclude PE (Figure 4.3). Chest X-ray is most
nostic studies to reinforce the clinical suspi- useful to rule out other conditions that may
cion of PE.32 mimic PE, such as pneumothorax or pneu-
momediastinum.27,35
Imaging Techniques Ventilation-Perfusion Scan
Chest X-ray The ventilation-perfusion (V/Q) scan is
Chest X-ray is often used in combination used to detect areas of abnormal perfusion
with ECG to reinforce suspicion of PE.32 due to PE and has long been considered a

preferred diagnostic modality in suspected in patients with previous VTE.36 Indetermi-

PE (Figure 4.4).27,29 The ventilation com- nate scans require further diagnostic stud-
ponent of the test excludes a diagnosis of ies to accurately assess disease,37 and 33%
pneumonia and other respiratory conditions. of PIOPED participants with indeterminate
However, the V/Q scan is only diagnostic scans were later found through angiography
of PE in a minority of cases. Moreover, PE to have thromboemboli.36
frequently occurs in combination with other
lung diseases, such as pneumonia or chronic Spiral Computerized Tomographic
obstructive pulmonary disease. Because Pulmonary Arteriography
most lung diseases affect pulmonary blood Spiral (also known as helical) CT has
flow as well as ventilation, their presence demonstrated a higher degree of sensitiv-
may decrease the scans specificity.27 The ity and interobserver agreement than V/Q
PIOPED study investigators at the National scan, making this strategy a less invasive,
Heart, Lung, and Blood Institute (NHLBI) alternative diagnostic tool in patients with
reported that a high-probability V/Q scan suspected PE (Figure 4.5).38 Spiral CT has
was sensitive for the diagnosis of PE, and a demonstrated a high rate of sensitivity and
low-probability scan was sensitive for the ab- specificity in detecting PE to segmental
sence of PE.36 The V/Q scan was not useful levels,39 but it cannot accurately detect sub-
Figure 4.4 99m Tc macroaggregated albumin perfusion scan showing multiple large segmental defects.

segmental emboli.39 Although the clinical sensitivity of 85% for the detection of lobar
significance of these subsegmental emboli and segmental emboli, with less reliability
has not been established, a negative spiral in detecting peripheral subsegmental em-
CT cannot safely rule out thrombosis and boli (1 in 5).35 These results are consistent
must be confirmed with pulmonary angiog- with those of the PIOPED study, in which
raphy.39 In addition, a low sensitivity rate of interobserver agreement was 98% for lobar
70% has been reported in a previous study, PE, 90% for segmental PE, and only 66%
providing additional evidence that confirma- for subsegmental emboli.42 However, there
tory pulmonary angiography is required in has been some debate surrounding the true
patients with negative spiral CT findings.40 clinical importance of these peripheral em-
A 1-year follow-up study in patients with a boli.35 Some investigators posit that they are
normal spiral CT scan demonstrated a low caused by isolated calf vein thrombi and
2% rate of clinical PE, suggesting that ad- do not require treatment, whereas others
ditional data will be required to character- believe that they are a precursor to larger
ize the safety of a negative spiral CT without emboli.43,44
other confirmatory diagnostics.4
Pulmonary Angiography
Pulmonary angiography has been consid- Newer Diagnostic
ered a gold standard test for PE.27 It is an
invasive test and in some settings used to Techniques
detect PE in patients with indeterminate
V/Q scans. However, it is not necessary for Indirect CT Venography
diagnosis of PE in the acute setting when
the perfusion scan is normal.27 One study Indirect CT venography has been investi-
of pulmonary angiography demonstrated a gated as an adjunct to the pulmonary angi-
ography conducted during the differential
diagnosis of PE.45 This technique results
in a high rate of detection of DVTthe
underlying cause of many subsequent PE
eventsand requires no additional contrast
material.46,47
Magnetic Resonance
Direct Thrombus Imaging
MRDTI is a novel technique that has dem-
onstrated accuracy and reproducibility for
DVT diagnosis in limited studies.48 It detects
the presence of methemoglobin in clots, al-
Figure 4.5 Spiral computerized tomographic lowing visualization of thrombus without
pulmonary arteriography has demonstrated a high rate using intravenous contrast and making it
of sensitivity and specificity in detecting pulmonary
useful for detection of subacute thrombosis.49
embolism.
The major advantages of MRDTI over con-

ventional modalities include (1) early data fers limited specificity and cannot be solely
suggesting that it is highly accurate for the used to exclude a diagnosis of DVT. Such
detection of both DVT and PE, providing a results have been demonstrated in emer-
single imaging modality for the detection of gency department patients,51 patients who
VTE;49,50 (2) direct visualization of throm- are elderly, and those hospitalized for >3
bus, avoiding the pitfalls of conventional days.52 Another important issue is that some
techniques that have either identified throm- institutions do not have access to immedi-
bus as a filling defect or in terms of surro- ate D-dimer results. Nevertheless, D-dimer
gates; and (3) simultaneous imaging of the testing, especially the most specific enzyme-
legs and chest, allowing a comprehensive linked immunosorbent assay, still offers an
assessment of thrombus load, minimizing attractive bedside assay that is sensitive for
the importance of overlooked subsegmental DVT diagnosis if not specific under certain
PE, and potentially facilitating more titrated circumstances, and will still result in a re-
treatment. The safety of withholding treat- duction in the need for imaging diagnostics
ment in suspected DVT and PE on the basis when DVT can be diagnosed and treated
of negative MRDTI alone is being evaluated earlier.53 A positive D-dimer is not useful for
in ongoing outcome studies.50 Additionally, necessarily ruling in the diagnosis of VTE;
because it has proven useful in identifying rather, the value is found in the negative test,
complicated plaque in the carotid arteries in which, when it coincides with ultrasonogra-
the setting of transient and permanent cere- phy, can rule out the diagnosis.25
bral ischemia, MRDTI offers promise as a While measurement of D-dimers is
technique that is capable of detecting high- a recent addition to the diagnostic strategy
risk vessel wall disease prior to significant or of PE and has been shown to be a valu-
permanent end-organ damage.49 As costs for able tool with excellent sensitivity, there
this type of imaging decrease and institutions have been rare reports of patients with PE
gain wider access to the technology, it should but negative D-dimer tests. One investiga-
offer an attractive alternative to the invasive tion at an academic health center indicated
use of contrast venography with application that D-dimer measurement was of limited
in a wide range of vascular disease settings. utility in patients with suspected PE and
nondiagnostic lung scans or negative spiral
(helical) CT results.54 Another study of 150
Bedside D-dimer Assays patients admitted to hospital for PE who un-
derwent D-dimer measurement compared
A number of D-dimer assays have been results in patients with negative D-dimers
evaluated as diagnostic markers for VTE vs. raised D-dimers. The sensitivity of raised
and are considered to be inexpensive as well D-dimers for PE was high (96%), but the
as timely25; the predictive value of the assay finding of chest pain was statistically greater
depends on several attributes, one being the in the group with negative D-dimers (P =
prevalence of VTE in the population being 0.01). In these negative D-dimer cases, the
tested. Ruiz-Gimnez and colleagues found emboli were all distal (P = 0.0003), and the
that the VIDAS and ELISA D-dimers as a diagnostic value of ultrasound investigations
first diagnostic tool for the exclusion of DVT (echocardiography, ultrasonography of lower
are suitable approaches.22 To the contrary, limb veins) was less than in patients with
some data suggest that D-dimer testing of- higher D-dimers (P <0.0001). The authors

suggested that measurement of D-dimers by 2. Horlander KT, Mannino DM, Leeper KV.
the ELISA method may be nondiagnostic Pulmonary embolism mortality in the United
in distal PE, perhaps because of the less ex- States, 1979-1998: an analysis using multiple-
tensive thromboembolic process. They con- cause mortality data. Arch Intern Med.
cluded that in cases with negative D-dimers, 2003;163:1711-7.
a strong clinical suspicion of PE should sig- 3. Kroegel C, Reissig A. Principle mechanisms
nal a need for further investigation.55 underlying venous thromboembolism:
epidemiology, risk factors, pathophysiology
and pathogenesis. Respiration. 2003;70:
Conclusion 7-30.
4. Cushman M, Tsai AW, White RH, Heckbert
Clinical evidence indicates that patients SR, Rosamond WD, Enright P, et al. Deep
who are at moderate to high risk for develop- vein thrombosis and pulmonary embolism in
ing VTE include those with a history of can- two cohorts: the longitudinal investigation
cer, prior thrombosis, acquired syndromes, of thromboembolism etiology. Am J Med.
or genetic disorders that predispose them 2004;117:19-25.
to a hypercoagulative state, among others. 5. Nielsen H. Pathophysiology of venous
Among these patients, VTE should be high- thromboembolism. Semin Thromb Hemost.
ly suspected with the presentation of classic 1991;17 (suppl 3):250-3.
symptoms. Such cases require rapid assess- 6. Alikhan R, Cohen AT, Combe S, Samama
ment and accurate diagnosis to prevent the MM, Desjardins L, Eldor A, et al;
progression of DVT, long-term morbidity MEDENOX Study. Risk factors for venous
due to postthrombotic syndrome, and/or the thromboembolism in hospitalized patients
occurrence of potentially fatal PE. with acute medical illness: analysis of the
However, clinical presentation for 50% MEDENOX Study. Arch Intern Med.
of patients with VTE is often nonspecific, 2004;10;164:963-8.
and can be confused with a variety of other 7. Gatt M, Paltiel O, Bursztyn M. Is prolonged
conditions, including heart failure, cellulitis, immobilization a risk factor for symptomatic
hematoma, or edema due to an unrelated venous thromboembolism in elderly
condition. bedridden patients? Thromb Haemost.
New additions to the diagnostic battery 2004;91:538-43.
and increased awareness of risk stratification 8. Rosenow E. Concise review for primary-care
paradigms have the potential to allow more physicians. Mayo Clin Proc. 1995;70:45-9.
accurate identification of VTE in some pa- 9. Duggan C, Marriott K, Edwards R, Cuzick J.
tients, thereby greatly reducing incidence of Inherited and acquired risk factors for venous
mortality and morbidity. thromboembolic disease among women taking
tamoxifen to prevent breast cancer. J Clin
Oncol. 2003;21:3588-93.
references 10. Samama MM, Dahl OE, Qunlan DJ,
1. Prandoni P, Lensing AW, Cogo A, Cuppini Mismetti P, Rosencher N. Quantification of
S, Villalta S, Carta M, et al. The long- risk factors for venous thromboembolism: a
term clinical course of acute deep venous preliminary study for the development of a risk
thrombosis. Ann Intern Med. 1996;125: assessment tool. Haematologica. 2003;88:
1-7. 1410-21.

11. Knudson MM, Ikossi DG, Khaw L, Morabito of deep venous thrombosis. Fallibility of
D, Speetzen LS. Thromboembolism after clinical symptoms and signs. Arch Surg.
trauma: an analysis of 1602 episodes from 1976;111:34-6.
the American College of Surgeons National 21. Anderson DR, Wells PS, Stiell I, MacLeod
Trauma Data Bank. Ann Surg. 2004;240:490- B, Simms M, Gray L, et al. Thrombosis in
8. the emergency department: use of a clinical
12. Bura A, Cailleux N, Bienvenu B, Lger diagnosis model to safely avoid the need for
P, Bissery A, Boccalon H, Fiessinger JN, urgent radiological investigation. Arch Intern
Levesque H, Emmerich J. Incidence and Med. 1999;159:477-82.
prognosis of cancer associated with bilateral 22. Ruiz-Gimnez N, Friera A, Artieda P,
venous thrombosis: a prospective study of 103 Caballero P, Sanchez Molin P, Morales M, et
patients. J Thromb Haemost. 2004;2:441-4. al. for the Thromboembolic Disease Group.
13. Hansson PO, Sorbo J, Eriksson H. Recurrent Rapid D-dimer test combined a clinical model
venous thromboembolism after deep vein for deep vein thrombosis. Validation with
thrombosis: incidence and risk factors. Arch ultrasonography and clinical follow-up in 383
Intern Med. 2000;160:769-74. patients. Thromb Haemost. 2004;91:1237-46.
14. Schulman S, Lindmarker P, Johnsson H. A 23. Wells PS, Hirsh J, Anderson DR, Lensing AW,
comparison of six weeks with six months of Foster G, Kearon C, et al. Accuracy of clinical
oral anticoagulant therapy after a first episode assessment of deep-vein thrombosis. Lancet.
of venous thromboembolism. N Engl J Med. 1995;345:1326-30.
1995;332:1661-5. 24. Wells PS, Ginsberg JS, Anderson DR, Kearon
15. Allaart CF, Poort SR, Rosendaal FR, Reitsma C, Gent M, Turpie AG, et al. Use of a clinical
PH, Bertina RM, Brit E. Increased risk of model for safe management of patients with
venous thrombosis in carriers of hereditary suspected pulmonary embolism. Ann Intern
protein C deficiency defect. Lancet. Med. 1998;129:997-1005.
1993;341:134-8. 25. Wells PS, Anderson DR, Ginsberg J.
16. Hedenmalm K, Samuelsson E, Spigset Assessment of deep vein thrombosis or
O. Pulmonary embolism associated with pulmonary embolism by the combined use of
combined oral contraceptives: reporting clinical model and noninvasive diagnostic tests.
incidences and potential risk factors for Semin Thromb Hemost. 2000;26:643-56.
a fatal outcome. Acta Obstet Gyn Scand. 26. Wells PS, Anderson DR, Rodger M, Stiell
2004;83:576-85. I, Dreyer JF, Barnes D, et al. Excluding
17. Mazza JJ. Hypercoagulability and venous pulmonary embolism at the bedside without
thromboembolism: a review. WMJ. diagnostic imaging: management of patients
2004;103:41-9. with suspected pulmonary embolism
18. Streiff MB, Segal J, Grossman SA, Kickler presenting to the emergency department by
TS, Weir EG. ABO blood group is a potent using a simple clinical model and D-dimer.
risk factor for venous thromboembolism in Ann Intern Med. 2001;135:98-107.
patients with malignant gliomas. Cancer. 27. Tapson VF, Carroll BA, Davidson BL, Elliott
2004;100:1717-23. CG, Fedullo PF, Hales CA, et al. for the
19. Gathof BS, Picker SM, Rojo J. Epidemiology, American Thoracic Society. The diagnostic
etiology and diagnosis of venous thrombosis. approach to acute venous thromboembolism.
Eur J Med Res. 2004;9:95-103. Clinical practice guideline. Am J Resp Crit
20. Cranley JJ, Canos AJ, Sull WJ. The diagnosis Care Med. 1999;160:1043-66.

28. Stein PD, Terrin ML, Hales CA, Palevsky 36. The PIOPED Investigators. Value of the
HI, Saltzman HA, Thompson BT, et al. ventilation/perfusion scan in acute pulmonary
Clinical, laboratory, roentgenographic, and embolism. Results of the Prospective
electrocardiographic findings in patients with Investigation of Pulmonary Embolism
acute pulmonary embolism and no pre- Diagnosis (PIOPED). JAMA. 1990;263:
existing cardiac or pulmonary disease. Chest. 2753-9.
1991;100:598-603. 37. Kember PG, Euinton HA, Morcos SK.
29. Baker WF Jr. Diagnosis of deep venous Clinicians interpretation of the indeterminate
thrombosis and pulmonary embolism. Med ventilation-perfusion scan report. Br J Radiol.
Clin North Am. 1998;82(3):459-76. 1997;70:1109-11.
30. Elias A, Cadne A, Elias M, Puget J, 38. Mayo JR, Remy-Jardin M, Mller NL, Remy
Tricoire JL, Colin C, et al. Extended lower J, Worsley DF, Hossein-Foucher C, et al.
limb venous ultrasound for the diagnosis Pulmonary embolism: prospective comparison
of proximal and distal vein thrombosis of spiral CT with ventilation-perfusion
in asymptomatic patients after total hip scintigraphy. R adiology. 1997;205:447-52.
replacement. Eur J Vasc Endovasc Surg. 39. Remy-Jardin M, Remy J, Deschildre F, Artaud
2004;27:438-44. D, Beregi JP, Hossein-Foucher C, et al.
31. Stevens SM, Elliott CG, Chan KJ, Egger MJ, Diagnosis of pulmonary embolism with spiral
Ahmed KM. Withholding anticoagulation CT: comparison with pulmonary angiography
after a negative result on duplex and scintigraphy. Radiology. 1996;200:
ultrasonography for suspected symptomatic 699-706.
deep venous thrombosis. Ann Intern Med. 40. Perrier A, Howarth N, Didier D, Loubeyre P,
2004;140:985-91. Unger PF, de Moerloose P, et al. Performance
32. Costantini M, Bossone E, Renna R, of helical computed tomography in unselected
Sticchi G, Licci E, De Fabrizio G, et al. outpatients with suspected pulmonary
Electrocardiographic features in critical embolism. Ann Intern Med. 2001;135:88-97.
pulmonary embolism. Results from baseline 41. Tillie-Leblond I, Mastora I, Radenne F,
and continuous electrocardiographic Paillard S, Tonnel AB, Remy J, et al. Risk of
monitoring [in Italian]. Ital Heart J. pulmonary embolism after a negative spiral
2004;5:214-6. CT angiogram in patients with pulmonary
33. Rodger M, Makropoulos D, Turek M, disease: 1-year clinical follow-up study.
Quevillon J, Raymond F, Rasuli P, et al. Radiology. 2002;223:461-7.
Diagnostic value of the electrocardiogram 42. Stein PD, Athanasoulis C, Alavi A,
in suspected pulmonary embolism. Am J Greenspan RH, Hales CA, Saltzman HA, et
Cardiol. 2000;86:807-9. al. Complications and validity of pulmonary
34. Iles S, Le Heron CJ, Davies G, Turner JG, angiography in acute pulmonary embolism.
Beckert LE. ECG score predicts those with Circulation. 1992;85:462-9.
the greatest percentage of perfusion defects 43. Dalen JE. When can treatment be withheld in
due to acute pulmonary thromboembolic patients with suspected pulmonary embolism.
disease. Chest. 2004;125:1651-6. Arch Intern Med. 1993;153:1415-8.
35. Gupta A, Frazer CK, Ferguson JM, Kumar 44. Moser KM. Venous thromboembolism. Am
AB, Davis SJ, Fallon MJ, et al. Acute Rev Respir Dis. 1990;141:235-49.
pulmonary embolism: diagnosis with MR 45. Richman PB, Wood J, Kasper DM, Collins
angiography. Radiology. 1999;210:353-9. JM, Petri RW, Field AG, et al. Contribution of

indirect computed tomography venography to Thromb Haemost. 2003;89:773-82.

computed tomography angiography of the chest 51. Farrell S, Hayes T, Shaw M. A negative
for the diagnosis of thromboembolic disease in SimpliRED D-dimer assay result does not
two United States emergency departments. exclude the diagnosis of deep vein thrombosis
J Thromb Haemost. 2003;1:652-7. or pulmonary embolus in emergency
46. Cham MD, Yankelevitz DF, Shaham D, department patients. Ann Emerg Med.
Shah AA, Sherman L, Lewis A, et al. for 2000;35:121-5.
the Pulmonary Angiography-Indirect CT 52. Brotman DJ, Segal JB, Jani JT, Petty BG,
Venography Cooperative Group. Deep venous Kickler TS. Limitations of D-dimer testing in
thrombosis: detection by using indirect CT unselected inpatients with suspected venous
venography. Radiology. 2000;216: 744-51. thromboembolism. Am J Med. 2003;114:276-
47. Loud PA, Katz DS, Bruce DA, Klippenstein 82.
DL, Grossman ZD. Deep venous thrombosis 53. Stein PD, Hull RD, Patel KC, Olson RE,
with suspected pulmonary embolism: Ghali WA, Brant R, Biel RK, Bharadia V,
detection with combined CT venography Kalra NK. D-dimer for the exclusion of acute
and pulmonary angiography. Radiology. venous thrombosis and pulmonary embolism:
2001;219:498-502. a systematic review. Ann Intern Med.
48. Fraser DGW, Moody AR, Morgan PS, 2004;140:589-602.
Martel AL, Davidson I. Diagnosis of lower- 54. Rathbun SW, Whitsett TL, Vesely SK, Raskob
limb deep venous thrombosis: a prospective GE. Clinical utility of D-dimer in patients
blinded study of magnetic resonance with suspected pulmonary embolism and
direct thrombus imaging. Ann Intern Med. nondiagnostic lung scans or negative CT
2002;136:89-98. findings. Chest. 2004;125:851-5.
49. Moody AR. Magnetic resonance direct 55. Mansencal N, Joseph T, Vieillard-Baron
thrombus imaging. J Thromb Haemost. A, El Hajjam M, Bendaoud M, Drouin A,
2003;1:1403-9. Jondeau G, Lacombe P, Jardin F, Dubourg O.
50. Kelly J, Hunt BJ, Moody A. Magnetic Negative D-dimers and peripheral pulmonary
resonance direct thrombus imaging: a novel embolism [in French]. Arch Mal Coeur Vaiss.
technique for imaging venous thromboemboli. 2003;96:1143-8.

CHAPTER
5
Treatment of Acute
What's new in the ACCP guidelines?
Anthony J. Comerota
G uidelines for patient care offer

recommendations to physicians
for diagnosis and management of common
The method of determining the
strength and quality of the recommenda-
tions deserves mention. Recommendations
diseases that generally apply to the typical are generally accompanied by a number,
patient. This chapter addresses the most which refers to the strength of the recom-
recent edition of the American College mendation, and a letter, which refers to the
of Chest Physicians (ACCP) guidelines to quality of the evidence supporting the rec-
help clinicians manage patients with acute ommendation. The current ACCP guide-
deep venous thrombosis (DVT). The eighth lines use 2 levels for the strength of their
ACCP consensus conference recently pub- recommendations: grade 1 for strong and
lished Antithrombotic Therapy for Venous grade 2 for weak.2 They further indicate
Thromboembolic Disease1 to help physi- that statements accompanied by a grade
cians care for patients with venous disease. 1 level are recommendations and state-
The authors have made specific changes ments accompanied by a grade 2 level are
with recommendations and suggestions suggestions.
linked to objective grades, which form the The quality of evidence upon which
basis of this discussion. the strength of the recommendation is based
ranges from A for high quality, which is
consistent evidence from multiple random-
ized trials, to B for moderate quality,
Davies md and Alan B. Lumsden md, eds. which is evidence from a single randomized
Cardiotext Publishing, ISBN 978-1-935395-22-5. trial or inconsistent evidence from random-
57

ized trials. Level C is low quality, which

is suggestive evidence from nonrandomized
Whats New
trials, observational reports, or expert opin- in Guidelines?
ion. Guideline-writing committees are be-
coming increasingly aware of costs of care The 2008 ACCP recommendations for the
and patient values and preferences, as are management of extensive venous thrombosis
physicians. It stands to reason that a clear- and pulmonary embolism are a major depar-
thinking, well-informed patient will agree ture from previous guideline versions. The
with treatment recommendations that follow authors acknowledged that extensive venous
strong guidelines (grades 1A) whereas when thrombosis is clearly associated with the de-
physicians are faced with atypical clinical velopment of the postthrombotic syndrome.
circumstances or weak guideline recom- Moreover, they recognized the existing evi-
mendations (ie, suggestions), cost of care dence supporting the value of venous throm-
and patient values and preferences should be bectomy (grade 2B) and catheter-directed
considered in addition to the risks and ben- thrombolysis (grade 2B) as components of a
efits of the treatment. strategy of thrombus removal, especially for
patients with iliofemoral DVT.
The morbidity of the postthrombotic
Magnitude syndrome was a driving force for revising the
of the Problem ACCP recommendations as was evidence
that early removal of thrombus prevents or
Venous disease is one of the most common reduces postthrombotic morbidity. Studies
disorders afflicting the populations of devel- have shown that patients with postthrom-
oped and developing countries. New studies botic venous insufficiency suffer reduced
show that acute venous thrombosis result- quality of life (QOL),7,8 and the development
ing in fatal pulmonary embolism kills more of the postthrombotic syndrome is directly
people than acute myocardial infarction or related to the severity of acute DVT. This is
acute stroke.3 Over 1 million people per year particularly true in patients with iliofemoral
will suffer an acute venous thromboembolic DVT, who appear to be a clinically relevant
event in the United States alone. The post- subset of patients with acute DVT and suf-
thrombotic morbidity that follows is sub- fer the most severe postthrombotic morbid-
stantial and is proportional to the extent of ity.6,9,10 The available evidence clearly shows
venous thrombosis.4-6 that a strategy of thrombus removal is the
Acute DVT often leads to chronic post- preferred management for patients with ilio-
thrombotic morbidity, and the severity of femoral DVT and offers them the best long-
postthrombotic venous disease correlates term outcome.11
directly with the extent of the acute DVT. The committee went on to refine anti-
Extensive DVT results in severe chronic coagulant therapy for patients with venous
morbidity unless patients are treated with a thromboembolism, addressing the needs
strategy to eliminate the acute clot. of patients with malignancy and idiopathic
venous thromboembolic disease. They also
dispelled the myth of bed rest for patients
with acute DVT, highlighting the benefit of
early anticoagulation, early compressions of

Treatment of Acute Deep Venous Thrombosis 59
the affected leg, and the significant benefit symptoms compared to patients treated with
of continued use of 30- to 40-mm Hg com- anticoagulation alone. Other, nonrandom-
pression stockings. ized series also reported favorable outcomes
of contemporary venous thrombectomy.
Long-term observational results from 10 re-
Treatment of Iliofemoral ports with the mean of 41 months of follow-up
Venous Thrombosis demonstrated a 76% patency, with 8 reports
The writing committee for the 2008 ACCP demonstrating functional venous valves in
guidelines recognized the excess morbid- the femoropopliteal segment in 63%.11
ity of this particular distribution of disease. Since venous thrombectomy is infre-
It is important to understand the anatomy quently performed, the committee sug-
of lower extremity venous drainage, which gested that catheter-directed thrombolysis
functionally resembles a funnel, with distal is usually preferable to operative venous
veins draining into larger but progressively thrombectomy (grade 2C).1
fewer veins as blood moves cephalad. The
common femoral and iliac veins represent
the spout of the funnel, which is the single Catheter-Directed Thrombolysis
common channel of lower extremity venous The recommendation for catheter-directed
drainage. If this channel is obstructed, it will thrombolysis (CDT) for acute iliofemoral
affect the entire leg, with adverse functional DVT in patients with a low risk of bleeding,
consequences on all distal veins. symptoms <14 days, good functional status,
The greatest change in guidelines is and a life expectancy 1 year is also pro-
the recommendation for consideration of posed to reduce acute symptoms and post-
a strategy of thrombus removal in patients thrombotic morbidity (grade 2B).
with iliofemoral DVT. This is a reversal of Rationale A small randomized trial of
the statements from guidelines published in catheter-directed lytic therapy vs. anticoagu-
2004.12 lation demonstrated significantly better pa-
tency and preservation of valve function in
patients treated with CDT vs. anticoagula-
Venous Thrombectomy tion.16 Large single-center series and multi-
The 2008 ACCP guidelines recommend center venous registries demonstrate an 80%
considering venous thrombectomy for acute to 90% success rate, with progressively lower
iliofemoral DVT in patients with symptoms bleeding complications over time.17-19 A case-
for <7 days, good functional status, and a life controlled cohort study, which followed the
expectancy 1 year (grade 2B). National Venous Registry, demonstrated sig-
Rationale This is based upon level 1 data nificantly improved quality of life (QOL) in
emanating from a large randomized study patients with iliofemoral DVT treated with
by Plate et al.13-15 Patients were followed up CDT compared to those treated with antico-
and reported at 6 months, 5 years, and 10 agulation.20 The improved QOL was directly
years following randomization to venous related to lytic success.
thrombectomy or anticoagulation. Patients The committee suggests correction of
randomized to thrombectomy demonstrated underlying venous lesions using balloon an-
improved patency, lower venous pressures, gioplasty and stents (grade 2C). While there
less leg swelling, and fewer postthrombotic are no objective data supporting this state-

ment, the collective clinical observations low-molecular-weight heparin (LMWH),

and expert opinion suggest that residual subcutaneous fondaparinux, subcutaneous
(uncorrected) venous lesions increase the weight-adjusted UFH, or subcutaneous mon-
likelihood of rethrombosis. Alternatively, itored UFH is recommended (grade 1A).
correction of focal lesions in the proximal Rationale The writing committee em-
system is associated with good long-term phasizes the importance of both early and
outcome. This suggestion applies to both ve- long-term anticoagulation for the treat-
nous thrombectomy and CDT. ment of acute DVT. Early anticoagulation,
Another suggestion new to this edition otherwise termed initial treatment, is de-
of the guidelines is for the use of pharmacom- signed to prevent thrombus extension and
echanical thrombolysis to reduce treatment embolization, and, when patients are treated
times, shorten hospital and intensive care properly, reduces the risk of early and late
unit stays, and reduce costs (grade 2C). Phar- recurrence. The objective of long-term an-
macomechanical techniques complement ticoagulation is to avoid recurrence. Slow or
catheter-directed thrombolysis, often result- delayed therapeutic anticoagulation results
ing in better patient outcomes in addition to in a significantly increased risk of recurrent
facilitating treatment. Current pharmacom- of venous thromboembolism compared to
echanical techniques include pulse-spray, early and immediate therapeutic anticoagu-
isolated segmental, and ultrasound-acceler- lation.24 If early anticoagulation with UFH
ated thrombolysis. While randomized trials falls below a therapeutic level, patients have
comparing pharmacomechanical techniques a 15-fold risk of recurrence.25
with catheter-directed thrombolysis alone are For isolated distal DVT (calf vein
lacking, single-center reports,21,22 multicenter thrombosis), 3 months of anticoagulation is
registries,23 and expert opinion suggest that recommended (grade 2B).
thrombus removal can be achieved more rap- Rationale A randomized trial of isolated
idly with lower doses of plasminogen activa- calf vein thrombosis treated with 5 days of
tors using pharmacomechanical techniques. anticoagulation with heparin followed by 3
The committee recommends the same months of vitamin K antagonists (VKA) vs.
intensity and duration of anticoagulation fol- 5 days of heparin followed by placebo dem-
lowing thrombectomy and CDT as patients onstrated significant benefit to 3 months of
who do not undergo these treatments (grade oral VKA. Nonrandomized trial observation
1C). This is a uniformly strong opinion by of thrombus propagation in patients not an-
the experts that underscores the need to ticoagulated would support this approach.
avoid recurrence, although proper trials of If a patient with isolated calf DVT were at
intensity and duration of anticoagulation fol- high risk for bleeding with anticoagulation,
lowing interventional therapy have not been calf compression and ambulation with ultra-
performed. sound surveillance appear to be an appropri-
ate approach.
Anticoagulation for Acute Venous For patients with proximal DVT, a min-
imum of 3 months of anticoagulation is rec-
Thromboembolic Disease ommended over shorter periods (grade 1A).
For patients with acute DVT, early thera- For patients with proximal DVT who are at
peutic anticoagulation with intravenous un- low risk for bleeding, long-term anticoagula-
fractionated heparin (UFH), subcutaneous tion is recommended (grade 1A).

Rationale Numerous studies evaluating occur at periodic intervals to evaluate their

the appropriate duration of anticoagulation risks of anticoagulation vs. their benefits.
for proximal DVT have been performed. As Rationale A randomized trial has dem-
a general observation, studies have demon- onstrated that in patients with unprovoked
strated that the longer the course of therapy, recurrent DVT, indefinite anticoagulation
the less risk of recurrence. Levine and col- significantly reduced recurrence compared
leagues26 randomized patients with veno- to 6 months of therapy (relative risk [RR],
graphically proven proximal DVT to 4 weeks 8.0; P <0.001).31 There was a trend toward
vs. 3 months of anticoagulation. They dem- reduced mortality; however, there was an in-
onstrated that 3 months of anticoagulation creased risk of a major bleed from indefinite
was significantly better than 4 weeks. Schul- anticoagulation (RR 0.3; P = 0.0084). The
man et al27 randomized patients to 6 weeks above data underscore the benefit of long-
vs. 6 months of oral anticoagulation for term anticoagulation with regard to recur-
acute DVT and demonstrated significant re- rence. However, reassessment of a patients
duction in recurrence in the patients treated bleeding risk over time is crucial to maintain
for 6 months. The bleeding complications benefit vs. risk.
were no different between the 2 groups. Ke- In patients with malignancy, LMWH is
aron and colleagues28 randomized patients recommended as the treatment of choice for
to 3 months vs. indefinite anticoagulation the first 3 to 6 months following diagnosis
and demonstrated significant long-term of acute DVT (grade 1A). Subsequent treat-
benefit to indefinite anticoagulation. There ment with VKA or LMWH is recommended
was a trend to more major bleeding in pa- for an indefinite period or until the cancer is
tients with long-term anticoagulation vs. resolved (grade 1C).
the 3-month cohort. Although Ridker et al29 Rationale A randomized trial evaluating
demonstrated long-term low intensity warfa- the LMWH dalteparin vs. VKA with Cou-
rin to be significantly better than placebo for madin demonstrated a significant reduction
treatment of idiopathic venous thromboem- in recurrent venous thromboembolism (RR
bolism, Kearon and colleagues30 compared 0.48; P = 0.002). There was no difference in
conventional long-term anticoagulation to bleeding complications nor was there a dif-
low-intensity long-term anticoagulation and ference in mortality. Since cancer is a major
showed no difference in bleeding complica- thrombotic risk, recurrence rates are high.
tions; the study did show a significant benefit Therefore, the recommendations are for an-
to conventional doses compared to subthera- ticoagulation until cancer is resolved.
peutic doses of VKA. Therefore, patients
with venous thromboembolic disease who
are at low risk for bleeding should be con- Early Ambulation and Compression
sidered for long-term anticoagulation (grade Early ambulation in patients with acute
1A). The patient should be reevaluated at DVT is now recommended in preference to
specified time intervals for the ongoing risk- initial bed rest (grade 1A).
benefit ratio. Rationale Bed rest and immobiliza-
In patients with recurrent, unprovoked tion are known risk factors for DVT and
DVT, long-term, indefinite anticoagulation thrombus propagation. Randomized trials
is recommended (grade 1A). For patients of early ambulation and leg compression
on long-term therapy, reassessment should have demonstrated reduced pain, edema,

and postthrombotic morbidity compared to compression, the addition of IPC is suggest-

patients treated with bed rest and anticoagu- ed (grade 2B).
lation.32-36 Application of a snug wrap from Rationale IPC has been shown to in-
the base of the toes to the upper thigh at the crease venous velocity, reduce edema, in-
time of diagnosis of acute DVT combined crease TcPO2, increase popliteal artery
with ambulation and anticoagulation is the blood flow, and increase endothelial nitric
method described by Partsch et al.37 This oxide synthase. These basic effects of IPC
has been shown to be effective in the early have translated into improved healing of
management of patients with acute proxi- venous leg ulcers in clinical trials. Random-
mal DVT. ized trials in patients with persistent venous
For patients who have symptomatic ulcers have demonstrated significantly in-
proximal DVT, elastic compression stock- creased healing42,43 and more rapid healing
ings of 30 to 40 mm Hg are recommended of venous leg ulcers when IPC was used in
(grade 1A). Stockings should be applied as addition to standard wound care and com-
soon as available and worn from the time the pression wraps. Compression pressures and
patient awakens in the morning until going cycles have varied in the studies reported;
to bed at night. therefore, IPC prescription for the treatment
Rationale Two randomized trials treat- of postthrombotic syndrome and venous ul-
ing patients after a first episode of acute cers has not been standardized. With our
symptomatic proximal DVT demonstrated understanding of the pathophysiology of
significant reduction (50% reduction) of chronic venous disease (high ambulatory
postthrombotic symptoms in patients wear- venous pressures) and the improved effec-
ing compression stockings compared to tiveness of high-pressure, rapid-inflation de-
those treated without compression.38,39 A vices,44 it appears that patients would benefit
Cochrane review of compression following from higher inflation pressures, short infla-
acute DVT also concluded that compres- tion time, and cycles of 2 to 3 minutes.
sion stockings substantially reduced the inci-
dence of the postthrombotic syndrome after
2 years.40 Conclusion
The 2008 ACCP guidelines have suggested
Intermittent Pneumatic Compression new approaches for patients with iliofemo-
For patients with severe edema of the leg ral DVT with their recommendations for
due to postthrombotic syndrome, a course of a strategy of thrombus removal to avoid
intermittent pneumatic compression is sug- postthrombotic morbidity. The data sup-
gested (grade 2B). porting operative venous thrombectomy
Rationale In a crossover study 41 in pa- and catheter-based techniques were used to
tients with severe postthrombotic syndrome, support such a recommendation. The com-
intermittent pneumatic compression (IPC) mittee also recognized the need for early
of 40 mm Hg was proven more effective therapeutic anticoagulation. Furthermore,
than placebo pressures. Patients uniformly they underscored the importance of long-
preferred therapeutic pressures to placebo. term anticoagulation for patients with un-
In patients with venous ulcers resistant provoked DVT and for cancer patients who
to healing with wound care and standard suffer DVT. Finally, the myth of bed rest

was dispelled with their recommendations Arch Intern Med. 2002;162(10):1144-8.

for early ambulation and the importance of 8. Kahn SR, Kearon C, Julian JA, Mackinnon
good compression from the time of diagnosis B, Kovacs MJ, Wells P, et al. Extended
and long-term treatment with 30 to 40 mm Low-intensity Anticoagulation for
Hg ankle-gradient compression stockings Thromboembolism (ELATE) Investigators.
were highlighted. Predictors of the post-thrombotic syndrome
during long-term treatment of proximal
deep vein thrombosis. J Thromb Haemost.
references 2005;3(4):718-23.
1. Kearon C, Kahn SR, Agnelli G, Goldhaber 9. ODonnell TF, Browse NL, Burnand KG,
SZ, Raskob G, Comerota AJ. Antithrombotic Thomas ML. The socioeconomic effects of
therapy for venous thromboembolic an iliofemoral venous thrombosis. J Surg Res.
disease: ACCP evidence-based clinical 1977;22(5):483-8.
practice guidelines (8th ed). Chest. 10. Akesson H, Brudin L, Dahlstrom JA, Eklof
2008;133(6):454S-545S. B, Ohlin P, Plate G. Venous function
2. Guyatt G, Gutterman D, Baumann MH, assessed during a 5 year period after acute
Addrizzo-Harris D, Hylek EM, Phillips B, ilio-femoral venous thrombosis treated
et al. Grading strength of recommendations with anticoagulation. Eur J Vasc Surg.
and quality of evidence in clinical guidelines: 1990;4(1):43-8.
report from an American College of Chest 11. Comerota AJ, Gravett MH. Iliofemoral venous
Physicians task force. Chest. 2006;129(1): thrombosis. J Vasc Surg. 2007;46(5):1065-76.
174-81. 12. Buller HR, Agnelli G, Hull RD, Hyers TM,
3. Heit JA, Cohen AT, Anderson FA Jr, on Prins MH, Raskob GE. Antithrombotic
behalf of the VTE Impact Assessment therapy for venous thromboembolic
Group. Estimated annual number of disease: the Seventh ACCP Conference on
incident and recurrent, non-fatal and fatal Antithrombotic and Thrombolytic Therapy.
venous thromboembolism (VTE) events Chest. 2004;126(3 suppl):401S-28S.
in the US. ASH Annual Meeting Abstracts 13. Plate G, Einarsson E, Ohlin P, Jensen R,
2005;106(11):910. Qvarfordt P, Eklof B. Thrombectomy with
4. Strandness DE, Jr., Langlois Y, Cramer temporary arteriovenous fistula: the treatment
M, Randlett A, Thiele BL. Long-term of choice in acute iliofemoral venous
sequelae of acute venous thrombosis. JAMA. thrombosis. J Vasc Surg. 1984;1(6):867-76.
1983;250(10):1289-92. 14. Plate G, Akesson H, Einarsson E, Ohlin
5. Beyth RJ, Cohen AM, Landefeld CS. Long- P, Eklof B. Long-term results of venous
term outcomes of deep-vein thrombosis. Arch thrombectomy combined with a temporary
Intern Med. 1995;155(10):1031-7. arterio-venous fistula. Eur J Vasc Surg.
6. Delis KT, Bountouroglou D, Mansfield 1990;4(5):483-9.
AO. Venous claudication in iliofemoral 15. Plate G, Eklof B, Norgren L, Ohlin P,
thrombosis: long-term effects on venous Dahlstrom JA. Venous thrombectomy for
hemodynamics, clinical status, and quality of iliofemoral vein thrombosis10-year results
life. Ann Surg. 2004;239(1):118-26. of a prospective randomised study. Eur J Vasc
7. Kahn SR, Hirsch A, Shrier I. Effect of Endovasc Surg. 1997;14(5):367-74.
postthrombotic syndrome on health-related 16. Elsharawy M, Elzayat E. Early results
quality of life after deep venous thrombosis. of thrombolysis vs anticoagulation in

iliofemoral venous thrombosis. A randomised 24. Brandjes DP, Heijboer H, Buller HR, de
clinical trial. Eur J Vasc Endovasc Surg. RM, Jagt H, ten Cate JW. Acenocoumarol
2002;24(3):209-14. and heparin compared with acenocoumarol
17. Bjarnason H, Kruse JR, Asinger DA, alone in the initial treatment of proximal-vein
Nazarian GK, Dietz CA Jr, Caldwell MD, thrombosis. N Engl J Med. 1992;327(21):
et al. Iliofemoral deep venous thrombosis: 1485-9.
safety and efficacy outcome during 5 years of 25. Hull RD, Raskob GE, Rosenbloom D, Panju
catheter-directed thrombolytic therapy. J Vasc AA, Brill-Edwards P, Ginsberg JS, et al.
Interv Radiol. 1997;8(3):405-18. Heparin for 5 days as compared with 10 days
18. Comerota AJ, Kagan SA. Catheter-directed in the initial treatment of proximal venous
thrombolysis for the treatment of acute thrombosis. N Engl J Med. 1990;322(18):
iliofemoral deep venous thrombosis. 1260-4.
Phlebology. 2000;15:149-55. 26. Levine MN, Hirsh J, Gent M, Turpie AG,
19. Mewissen MW, Seabrook GR, Meissner Weitz J, Ginsberg J, et al. Optimal duration
MH, Cynamon J, Labropoulos N, Haughton of oral anticoagulant therapy: a randomized
SH. Catheter-directed thrombolysis for lower trial comparing four weeks with three
extremity deep venous thrombosis: report of months of warfarin in patients with proximal
a national multicenter registry. Radiology. deep vein thrombosis. Thromb Haemost.
1999;211(1):39-49. 1995;74(2):606-11.
20. Comerota AJ, Throm RC, Mathias SD, 27. Schulman S, Rhedin AS, Lindmarker P,
Haughton S, Mewissen M. Catheter-directed Carlsson A, Larfars G, Nicol P, et al. A
thrombolysis for iliofemoral deep venous comparison of six weeks with six months of
thrombosis improves health-related oral anticoagulant therapy after a first episode
quality of life. J Vasc Surg. 2000;32(1): of venous thromboembolism. Duration of
130-7. Anticoagulation Trial Study Group. N Engl J
21. Martinez J, Comerota AJ, Kazanjian S, Med. 1995;332(25):1661-5.
DiSalle RS, Sepanski DM, Assi Z. The 28. Kearon C, Gent M, Hirsh J, Weitz J, Kovacs
quantitative benefit of isolated, segmental, MJ, Anderson DR, et al. A comparison of
pharmacomechanical thrombolysis three months of anticoagulation with extended
for iliofemoral DVT. J Vasc Surg. anticoagulation for a first episode of idiopathic
2008;48(6):1532-7. venous thromboembolism. N Engl J Med.
22. Lin PH, Zhou W, Dardik A, Mussa F, 1999;340(12):901-7.
Kougias P, Hedayati N, et al. Catheter-direct 29. Ridker PM, Goldhaber SZ, Danielson E,
thrombolysis versus pharmacomechanical Rosenberg Y, Eby CS, Deitcher SR, et al.
thrombectomy for treatment of symptomatic PREVENT Investigators. Long-term, low-
lower extremity deep venous thrombosis. Am J intensity warfarin therapy for the prevention
Surg. 2006;192(6):782-8. of recurrent venous thromboembolism.
23. Parikh S, Motarjeme A, McNamara T, N Engl J Med. 2003;348(15):1425-34.
Raabe R, Hagspiel K, Benenati JF, et al. 30. Kearon C, Ginsberg JS, Kovacs MJ,
Ultrasound-accelerated thrombolysis for the Anderson DR, Wells P, Julian JA, et al.
treatment of deep vein thrombosis: initial Extended Low-Intensity Anticoagulation
clinical experience. J Vasc Interv Radiol. for Thrombo-Embolism Investigators.
2008;19(4):521-8. Comparison of low-intensity warfarin

therapy with conventional-intensity warfarin compression in the treatment of deep vein

therapy for long-term prevention of recurrent thrombosis. Dis Mon. 2005;51(2-3):135-40.
venous thromboembolism. N Engl J Med. 38. Brandjes DP, Buller HR, Heijboer H, Huisman
2003;349(7):631-9. MV, de Rijk M, Jagt H, et al. Randomised trial
31. Schulman S, Granqvist S, Holmstrom M, of effect of compression stockings in patients
Carlsson A, Lindmarker P, Nicol P, et al. The with symptomatic proximal-vein thrombosis.
duration of oral anticoagulant therapy after a Lancet. 1997;349(9054):759-62.
second episode of venous thromboembolism. 39. Prandoni P, Lensing AW, Prins MH, Frulla
The Duration of Anticoagulation Trial Study M, Marchiori A, Bernardi E, et al. Below-
Group. N Engl J Med. 1997;336(6):393-8. knee elastic compression stockings to
32. Aschwanden M, Labs KH, Engel H, prevent the post-thrombotic syndrome: a
Schwob A, Jeanneret C, et al. Acute deep randomized, controlled trial. Ann Intern Med.
vein thrombosis: early mobilization does 2004;141(4):249-56.
not increase the frequency of pulmonary 40. Kolbach DN, Sandbrink MW, Hamulyak
embolism. Thromb Haemost. 2001;85(1):42-6. K, Neumann HA, Prins MH. Non-
33. Blattler W, Partsch H. Leg compression and pharmaceutical measures for prevention
ambulation is better than bed rest for the of post-thrombotic syndrome. Cochrane
treatment of acute deep venous thrombosis. Database Syst Rev. 2004;(1):CD004174.
Int Angiol. 2003;22(4):393-400. 41. Ginsberg JS, Magier D, Mackinnon B, Gent
34. Junger M, Diehm C, Storiko H, Hach- M, Hirsh J. Intermittent compression units for
Wunderle V, Heidrich H, Karasch T, et al. severe post-phlebitic syndrome: a randomized
Mobilization versus immobilization in the crossover study. CMAJ. 1999;160(9):1303-6.
treatment of acute proximal deep venous 42. Smith PC, Sarin S, Hasty J, Scurr JH.
thrombosis: a prospective, randomized, Sequential gradient pneumatic compression
open, multicentre trial. Curr Med Res Opin. enhances venous ulcer healing: a randomized
2006;22(3):593-602. trial. Surgery. 1990;108(5):871-5.
35. Partsch H, Blattler W. Compression and 43. Kumar S, Samraj K, Nirujogi V, Budnik
walking versus bed rest in the treatment J, Walker MA. Intermittent pneumatic
of proximal deep venous thrombosis with compression as an adjuvant therapy in venous
low molecular weight heparin. J Vasc Surg. ulcer disease. J Tissue Viability. 2002;12(2):
2000;32(5):861-9. 42-4, 46, 48.
36. Schellong SM, Schwarz T, Kropp J, Prescher 44. Malone MD, Cisek PL, Comerota AJ,
Y, Beuthien-Baumann B, Daniel WG. Bed rest Jr., Holland B, Eid IG, Comerota AJ.
in deep vein thrombosis and the incidence of High-pressure, rapid-inflation pneumatic
scintigraphic pulmonary embolism. Thromb compression improves venous hemodynamics
Haemost. 1999;82 suppl 1:127-9. in healthy volunteers and patients who are
37. Partsch H. Immediate ambulation and leg post-thrombotic. J Vasc Surg. 1999;29(4):593-9.

CHAPTER
Anticoagulation Therapy
ruth l. Bush, Brandi Huf, and CleAnn Toner
T he main aim of venous thromboembo-

lism (VTE) therapy is to prevent exten-
sion of thrombosis and embolization to the
pathways of the coagulation cascade: the
contact activation pathway (intrinsic path-
way) and the tissue factor pathway (extrinsic
lungs. Other long-term goals include reduc- pathway).3 These 2 pathways meet to form
tion in the incidence of recurrent VTE, the final common pathway. Of the 2 initial
prevention of postthrombotic syndrome, pathways, the tissue factor pathway is the
and avoidance of pulmonary hypertension.1 primary pathway. Figure 6.1 gives a sim-
Anticoagulants have been the mainstay of plified diagram of the coagulation cascade
VTE therapy since Barritt and Jordan dem- and the site of action of the most commonly
onstrated the efficacy of heparin and warfa- used commercially available anticoagulant
rin in reducing morbidity and mortality in drugs.
patients with acute PE.2 Since then, a vast
array of drugs have emerged, including low-
molecular-weight heparin (LMWH), direct Oral Agents
thrombin inhibitors, and factor Xa inhibitors.
When looking at agents for anticoagu- Vitamin K AntagonistWarfarin
lation, an understanding of the coagulation While there are many ongoing studies to
cascade needs to be achieved. There are 2 find a new anticoagulant medication, war-
farin remains the only oral anticoagulant
available on the market in the United States.
Indications for warfarin include VTE (deep
Davies md and Alan B. Lumsden md, eds. venous thrombosis and pulmonary embo-
Cardiotext Publishing, ISBN 978-1-935395-22-5. lism), atrial fibrillation, mechanical and
67

Figure 6.1 The coagulation cascade and available anticoagulants. Adapted from Weitz JI, Bates SM. New
anticoagulants. J Thromb Haemost. 2005;3:1843-1853. Used with permission.
bioprosthetic heart valve replacement, myo- are more likely to affect the metabolism of
cardial infarction (MI), and hypercoagu- warfarin, including amiodarone, azole anti-
lable conditions.4 fungals, Flagyl, and Bactrim.
Vitamin K antagonists (VKAs) exert Patients should be advised to moni-
their anticoagulation effect by inhibiting the tor their diet when initiating warfarin due
enzyme vitamin K oxide reductase, block- to the adverse affect it can have on the pa-
ing the conversion of vitamin K epoxide to tients therapy. Stressing a diet consistent in
vitamin K. This inhibits the carboxylation vitamin K foods can have a considerable im-
and activation of the vitamin Kdependent pact on stabilizing patients early on in their
coagulation factors II, VII, IX, and X, and therapy. See Figure 6.2 for a list of vitamin K
proteins C and S.5 Warfarin is a racemic foods divided into low, moderate, and high
mixture of 2 active isomers, the R and S vitamin K content. In our facility, we watch
enantiomers. Of the 2 enantiomers, the S closely for changes in consumption of dark
enantiomer is 2.7 to 3.8 times more potent leafy green vegetables, green tea, nutritional
than the R enantiomer. The S enantiomer supplements, and multivitamin use. On the
is metabolized by the CYP2C9 enzyme of other side of the spectrum, 2 foods we ad-
the cytochrome P450 system, whereas the vise patients to stay away from are cranber-
R enantiomer is metabolized by CYP 1A2 ries and grapefruit, along with their juices.
and 3A4.5 Therefore, medications that are Grapefruit can inhibit warfarin metabolism,
metabolized through the CYP2C9 enzyme and the mechanism behind the interaction

Anticoagulation Therapy 69
Figure 6.2 Vitamin K foods. Adapted from USDA National Nutrient Database, http://www.nal.usda.gov/fnic
/foodcomp/Data/SR17/wtrank/sr17a430.pdf.

with cranberries is largely unknown; how- In patients at high risk for developing
ever, it is thought to also be involved in me- another event and in need of immediate
tabolism. anticoagulation, heparin/LMWH is used
Contraindications to using this medica- alongside warfarin for at least 4 to 5 days
tion are typically based on risk vs. benefit. If and pending 2 therapeutic INR levels. The
the beneficial use of warfarin is outweighed therapeutic INR range for most indications
by a higher risk of bleeding, the patient can- continues to be 2.0 to 3.0 (target 2.5) and for
not be placed on warfarin. Disease states high-risk mechanical valves 2.5 to 3.5 (target
and processes included in this consideration 3.0).5,10,11
are certain blood disorders, active bleeding, Monitoring is usually done within the
any disorder causing the patient to be more first few days of therapy initiation and then
prone to bleed, and malignant hypertension. is spaced out depending on the patients re-
If a patient cannot be properly monitored sponse and stabilization of warfarin. If a pa-
due to personal/psychological issues or non- tient develops a sub/supratherapeutic INR
compliance, it is more often than not safer due to diet, noncompliance, interacting
for the patient to be taken off warfarin. As medication, hold for surgery, or for any other
with all medications, if the patient develops reason, it is prudent to monitor more closely
an allergic reaction to warfarin, other op- and restart the cycle of monitoring. A patient
tions regarding anticoagulation will need should never go more than 4 weeks without
to be evaluated. Warfarin is teratogenic and having an INR drawn.5 Dose adjustment pro-
cannot be used during pregnancy. When a tocols vary depending on the institution.12,13
patient already on warfarin is scheduled for
any type of surgery or procedure where there
is a high risk of bleeding, the patient will Parenteral
need to hold warfarin for a sufficient period
of time prior to the procedure with or with- Anticoagulants
out an alternative anticoagulant.4
Warfarin dosing is highly individualized Heparins
and must take into account comorbid disease First discovered in 1916, unfractionated
states. The recommended initiation dose for heparin (UFH) exerts its anticoagulant
warfarin is 5 mg, which was shown to achieve effect by binding to antithrombin.14 The
therapeutic INR as fast as the 10-mg loading resulting heparin-antithrombin complex in-
dose.6 This was contested in an outpatient hibits the activity of factor Xa and thrombin.
study that demonstrated faster achievement Only one-third of the molecules of UFH
of target INR with a 10-mg loading dose.7 have the required sequence for binding to
However, the study also demonstrated in- antithrombin.15 UFH also contains mol-
creased rates of bleeding associated with the ecules that bind to other plasma proteins
higher loading dose. A lower starting dose that are present in variable amounts in each
of 2 to 3 mg daily may be used in patients individual. Thus, UFH has an unpredict-
with disease states or illnesses causing an in- able effect in different people.16 UFH is me-
creased response to warfarin, including the tabolized by the liver and has side effects of
elderly population.5,8 Table 6.1 is an example bleeding, heparin-induced thrombocytope-
of a warfarin dose adjustment protocol used nia (HIT), and osteoporosis.16 The dosing
in our facility for initiation of warfarin.9 of UFH is based on 1 of 2 schedules and

Table 6.1 WarfarinInitialDosingProtocol
INR Dayl Day2 Day3 Day4 Day5
<1.5 5.0 mg 5.0 mg 7.5 mg 10.0 mg 10.0 mg
1.51.99 2.5 mg 2.5 mg 5.0 mg 7.5 mg 7.5 mg
2.02.49 1.0 mg 2.5 mg 5.0 mg 5.0 mg
2.52.9 0.0 mg 1.0 mg 2.5 mg 2.5 mg
3.03.5 0.0 mg 0.0 mg 1.0 mg 1.0 mg
>3.5 0.0 mg 0.0 mg 0.0 mg 0.0 mg
requires intravenous administration and fre- Fractionated heparin, also known as

quent monitoring of the aPTT, or activated low-molecular-weight heparin (LMWH),
partial thromboplastin time, to ensure that is formed by enzymatic cleavage of UFH.
a steady therapeutic level of anticoagulation LMWH retains the ability to inhibit factor
is achieved. Xa and thrombin, but its smaller size de-
Heparin is indicated in venous and/ creases nonspecific binding to other plasma
or arterial thromboembolism for both pre- proteins. Compared to UFH, LMWH has in-
vention and treatment. Other indications creased bioavailability, a longer half-life, and
include disseminated intravascular coagula- a more predictable dose response. In addi-
tion and surgeries that carry a high risk of tion, LMWH is associated with a lower inci-
clotting, as well as procedures or laboratory dence of heparin-induced thrombocytopenia
tests requiring anticoagulation. Contrain- (HIT), potentially lower risk of bleeding, and
dications to using this medication include lower risk of osteoporosis.16 It has a dose- de-
thrombocytopenia, heparin-induced throm- pendent renal clearance, which needs to be
bocytopenia, and bleeding.17 Heparin can taken into consideration in patients with
be given a few different ways when begin- renal insufficiency. Other considerations
ning treatment. When used to treat venous include dosing difficulties in obese patients
thromboembolism (VTE), for example, dif- and problems with reversibility in cases of
ferent routes and dosing administration can bleeding. There are currently 8 LMWHs
be used (Table 6.2). available in the market, each with its own
The activity of heparin is measured by molecular weight and dosing regimen.
the degree of inhibition of activated factor There is insufficient evidence to differentiate
X (Xa), which can be monitored using the among these LMWHs based on efficacy and
aPTT, with a therapeutic range being 1.5 to safety.16 LMWHs do not require monitoring
2.5 times the normal (or 0.30.7 anti-factor to ensure therapeutic anticoagulation. The
Xa units/mL).22,23 As with warfarin, the dose utility of anti-factor Xa assay is controversial,
of heparin is adjusted based on the institu- since its correlation with anticoagulant effect
tions nomogram or dose adjustment protocol. and bleeding risk is unclear.25
The therapeutic aPTT is dependent on the Early studies demonstrated that
reagent used during testing, as they do vary.24 LMWH had better efficacy and safety when

Table 6.2 HeparinDosing
Route Weight-Based Non-Weight-Based
IV 80 units/kg bolus + 5,000 units bolus +

18 units/kg/h infusion18 32,000 units/day infusion19
SQ 333 units/kg bolus + 5,000 units bolus + the

250 units/kg twice daily20 weight-based daily dose21
compared to UFH, particularly for reducing induced thrombocytopenia when regional

mortality at 3- to 6- month follow-up.26 The anesthesia is needed,31 and allergies to the
more recent evidence has shown a lower drug, heparin, or pork.30-32 If a patient is al-
magnitude of benefit of LMWH over UFH.27 lergic to sulfites or benzyl alcohol, tinzapa-
The majority of the evidence is with respect rin therapy is contraindicated.32
to therapy of DVT, rather than PE. However, Enoxaparin is used most often in the
a 2004 meta-analysis provided convincing hospital setting, and dosing is listed here.
evidence that LMWH is safe and efficacious For specific dosing on dalteparin and tinza-
in patients with noncritical PE.28 Cost-effec- parin, please refer to their package insert.31-32
tiveness analyses have shown a benefit or at Treatment dosing for enoxaparin is 1 mg/
least equivalency for LMWH in comparison kg twice daily or 1.5 mg/kg daily. In patients
to UFH regardless of treatment setting.29 with creatinine clearance (CrCl) of less than
The low-molecular-weight heparins or equal to 30 mL/min, dosing is decreased
available in the United States are enoxapa- to 1 mg/kg daily. Prophylactic dosing for
rin (Lovenox), dalteparin (Fragmin), and surgeries is normally 40 mg daily or 30 mg
tinzaparin (Innohep). Since their FDA in- twice daily. In renal dysfunction, the dose is
dications vary, they will be listed separately. decreased to 30 mg daily.30
Enoxaparin is indicated in the prophylaxis of Low-molecular-weight heparin moni-
venous thromboembolism, as well as in the toring is usually not performed; however,
treatment of venous thromboembolism and high-risk patients can be monitored using
ST segment elevation myocardial infarction. anti-factor Xa levels. High-risk patients in-
The medication is also used in patients with clude those who have received heparin with-
unstable angina and non-Q-wave MI for in the past 6 months and women who are
prevention of ischemic embolisms during pregnant.22
treatment of these disease states.30 Daltepa- Fondaparinux is an indirect factor Xa
rin is not indicated in the treatment of ve- inhibitor. It has a linear pharmacokinetic
nous thromboembolism unless the patient profile, allowing for weight-based daily dos-
has cancer and a history of blood clots; how- ing and negating the need for continuous
ever, it can be used for prophylaxis similar monitoring.33 It does not bind to platelet
to enoxaparin.31 Tinzaparin is indicated only factor 4 and theoretically does not cause
for therapy in patients with venous thrombo- thrombocytopenia, and it has renal clear-
embolism.32 ance. In 2 large, multicenter clinical trials,
Contraindications to using LMWHs fondaparinux was as effective and safe as
as a class include hemorrhage, heparin- UFH for the treatment of PE, and as effec-

tive and safe as enoxaparin for the treatment Table 6.3 FondaparinuxDosing
of DVT.34,35 The drug is also used for prophy-
Dose
laxis of venous thromboembolism and is con- Weight(kg) (mg)
traindicated in patients with CrCl <30 mL/
min, hemorrhage, bacterial endocarditis, Prophylaxis 2.5 mg 2.5
thrombocytopenia, and allergy to the drug Treatment <50 kg 5.0
itself, and for prophylaxis in patients weigh-
50100 kg 7.5
ing less than 50 kg.36 Fondaparinux is dosed
based on body weight for treatment and at a >100 kg 10.0
fixed dose for prophylaxis (Table 6.3).
As with LMWHs, fondaparinux is cus-
tomarily not monitored.22
initial dose should be 44 kg, and the maxi-
mum infusion dose should be 16.5 mg/h.37
Direct Thrombin Inhibitors The dose of lepirudin does need to be re-
Direct thrombin inhibitors also have the ad- nally adjusted starting at a CrCl of 60 mL/
vantages of predictable dose response and min.37 Monitoring of lepirudin is done with
reduced incidence of thrombocytopenia. the aPTT. A baseline value should be ob-
However, they do not have an antidote for tained, and lepirudin should not be started
cases of severe bleeding. Argatroban and if the baseline aPTT is more than 2.5 times
lepirudin are 2 intravenous direct thrombin the normal aPTT control.37 Target range for
inhibitors that are FDA approved for the pre- aPTT during lepirudin treatment should be
vention and/or treatment of VTE in patients 1.5 to 2.5 times the normal aPTT control,
with heparin-induced thrombocytopenia.33 with the value taken 4 hours after the initial
Ximelagatran is an oral direct thrombin in- dose of lepirudin.37
hibitor that has been withdrawn from the Argatroban is dosed initially at 2 mcg/
market because of an increased risk of he- kg/min, as a continuous infusion. The dose
patic toxicity. may be adjusted up to 10 mcg/kg/min to an
These medications bind directly to aPTT of 1.5 to 3 times the baseline.38 The
thrombin to inhibit the activation of fibrin. aPTT levels need to be drawn 1 to 3 hours
Lepirudin and argatroban are indicated for after the dose is given. No dosage adjustment
patients with HIT.37,38 Bivalirudin is indi- is necessary in renal impairment. However,
cated for patients undergoing percutaneous adjustments are needed in hepatic impair-
transluminal coronary angioplasty (PTCA) ment to 0.5 mcg/kg/min as an initial dose,
who have unstable angina.39 Contraindica- and then adjusted according to the aPTT.38
tions to the direct thrombin inhibitors in- Initial dose of bivalirudin is an IV bolus
clude hypersensitivity to the products37-39 and of 0.75 mg/kg, followed with an infusion of
major bleeding.38,39 Antidotes are not avail- 1.75 mg/kg/h during the PTCA.39 An activat-
able for the direct thrombin inhibitors.10 ed clotting time (ACT) should be drawn 5
Lepirudin is dosed 0.4 mg/kg body minutes after the bolus, and an extra 0.3 mg/
weight (up to 110 kg) IV as a bolus dose, kg should be given to achieve the target ACT
then 0.15 mg/kg body weight/h IV continu- of 300 to 350 seconds if needed. Dose adjust-
ous infusion for 2 to 10 days.37 If a patients ments are necessary in renal impairment. In
body weight exceeds 110 kg, the maximum patients with CrCl of 30 to 59 mL/min, the

infusion rate should be 1.75 mg/kg/h; CrCl researched to attempt to alleviate some of
less than 30 mL/min, reduce rate to 1 mg/ the current drawbacks.
kg/h.39 When looking at developing agents
for anticoagulation, another look to the co-
agulation cascade provides a graphical un-
New Anticoagulants derstanding of their mechanism of action
Current medications on the market for (Figure 6.3).
anticoagulation have many drawbacks. As mentioned at the beginning of the
Vitamin K antagonists require intensive chapter, there are 2 pathways of the coagula-
monitoring and have multiple drug interaction cascade: the contact activation pathway
tions. Heparin usually requires admission and the tissue factor pathway.3 These 2 path-
to the hospital, it is administered paren- ways meet to form the final common path-
terally, and there is the possibility for the way. The first step in this pathway to initiate
development of heparin-induced thrombo- coagulation is the binding of tissue factor to
cytopenia. Low-molecular-weight heparins factor VII, which activates factor VIIa/tissue
are also parenterally administered, have factor complex. New agents that are being
the potential for developing HIT, and cost studied target this step to inhibit the initia-
considerably more than the other alterna- tion of coagulation, including Tifacogin and
tives. Several new medications are being NAPc2, which are both parenteral medica-
Figure 6.3 New anticoagulants. Adapted from Weitz JI, Bates SM. New anticoagulants. J Thromb Haemost.
2005;3:1843-1853. Used with permission.

tions.10 The propagation step of the coagu- references

lation cascade is where the final common 1. McRae S, Ginsberg J. Initial treatment of
pathway begins at factor Xa.3 With inhibition venous thromboembolism. Circulation.
of factor Xa, propagation is terminated. This 2004;110(9 suppl 1):I3-9.
can be done indirectly by binding to the 2. Barritt DW, Jordan SC. Anticoagulant drugs
plasma cofactors (antithrombin) or directly in the treatment of pulmonary embolism: a
by binding directly to the enzyme. This can controlled trial. Lancet. 1960;1:1309-12.
also be achieved by inhibiting factor IXa, 3. Haines ST, Zeolla M, Witt DM. Venous
thus preventing the activation of X to Xa. thromboembolism. In: DiPiro JT, Talbert RL,
The indirect factor Xa inhibitors in develop- Yee GC, Matzke GR, Wells BG, Posey LM,
ment include idraparinux, SSR12517E, and eds. Pharmacotherapy: A Pathophysiologic
SR123781A, all of which are parenteral.10 Approach. 6th ed. New York: McGraw-Hill/
The direct factor Xa inhibitors still under de- Appleton and Lange; 2005. p373-413.
velopment are otamixaban, apixaban, rivar- 4. Coumadin [package insert]. Princeton, NJ:
oxaban, LY-517717, YM-150, DU-176b, and Bristol-Myers Squibb Company; 2009. http://
PRT054021.10 Of these, otamixaban is paren- packageinserts.bms.com/pi/pi_coumadin.pdf.
teral, and the rest are oral medications. The 5. Ansell J, Hirsh J, Hylek E, Jacobson A,
only factor IXa inhibitor under development Crowther M, Palareti G; American College
is RB006, which is IV administration. An- of Chest Physicians. Pharmacology and
other target in the coagulation cascade is the management of the vitamin K antagonists.
final step, the formation of fibrin. Thrombin Chest. 2008;133:160S-98S.
converts fibrinogen to fibrin and activates 6. Crowther MA, Ginsberg JB, Kearon C,
platelets.3 By inhibiting thrombin, the final Harrison L, Johnson J, Massicotte MP, et al. A
step of the coagulation cascade is terminated. randomized trial comparing 5-mg and 10-mg
Again, this inhibition can be done indirectly warfarin loading doses. Arch Intern Med.
or directly. The indirect thrombin inhibitors 1999;159:46-8.
act on the heparin cofactor II. There was an 7. Kovacs MJ, Rodger M, Anderson DR, Morrow
oral indirect thrombin inhibitor in develop- B, Kells G, Kovacs J, et al. Comparison
ment, odiparcil.10 However, its development of 10-mg and 5-mg warfarin initiation
was stopped at phase II. Of the direct throm- nomograms together with low-molecular-
bin inhibitors being developed, 2 are paren- weight heparin for outpatient treatment
teral: flovagatran and pegmusirudin.10 The of acute venous thromboembolism. A
other is oral, dabigatran etexilate.10 Factor Va randomized, double-blind, controlled trial.
is activated by thrombin, which then helps Ann Intern Med. 2003;138:714-9.
convert more prothrombin into thrombin. 8. Garcia D, Regan S, Crowther M, Hughes
Activated protein C degrades factor Va to act RA, Hylek EM. Warfarin maintenance
as a natural anticoagulant.3 Two of the new dosing patterns in clinical practice. Chest.
medications being developed act on factor 2005;127;2049-56.
Va. Drotrecogin alfa is a recombinant form 9. Scott and White Healthcare Anticoagulation
of activated protein C, which will degrade Guidelines 2009-2010.
factor Va, and ART-123 converts thrombin 10. Antithrombotic and thrombolytic therapy:
into an activator of protein C.10 American College of Chest Physicians Evi-
denced-Based Clinical Practice Guidelines,
8th ed. Chest. 2008;133(6):67S-968S.

11. Kearon C, Kahn SR, Agnelli G, Goldhaber S, 21. Prandoni P, Carnovali M, Marchiori A;
Raskob GE, Comerota AJ; American College Galilei Investigators. Subcutaneous adjusted-
of Chest Physicians. Antithrombotic therapy dose unfractionated heparin vs fixed-dose
for venous thromboembolic disease: the low-molecular-weight heparin in the initial
eighth ACCP conference on antithrombotic treatment of venous thromboembolism. Arch
and thrombolytic therapy. Chest. Intern Med. 2004;164(10):1077-83.
2008;133:454S545S. 22. Hirsh J, Bauer KA, Donati MB, Gould M,
12. Ebell M. A systematic approach to managing Samama MM, Weitz JI; American College of
warfarin doses. Family Practice Management. Chest Physicians. Parenteral anticoagulants:
2005:77-83. http://www.aafp.org/fpm American College of Chest Physicians
/20050500/77asys.html. Evidence-Based Clinical Practice Guidelines
13. Franke CA, Dickerson LM, Carek PJ. Improv- (8th ed.). Chest. 2008;133(6 suppl):
ing anticoagulation therapy using point-of- 141S-59S.
care testing and a standardized protocol. Ann 23. Basu D, Gallus A, Hirsh J, Cade J. A
Fam Med. 2008;6 suppl 1:S28-32. prospective study of the value of monitoring
14. McLean J. The thromboplastic action of heparin treatment with the activated partial
cephalin. Am J Physiol. 1916;41:250-7. thromboplastin time. N Engl J Med.
15. Choay J, Lormeau JC, Petitou M, Sina P, 1972;287(7):324-7.
Fareed J. Structural studies on a biologically 24. Brill-Edwards P, Ginsberg JS, Johnston M,
active hexasaccharide obtained from heparin. Hirsh J. Establishing a therapeutic range
Ann NY Acad Sci. 1981;370:644-9. for heparin therapy. Ann Intern Med.
16. McRae S, Ginsberg J. Initial treatment of 1993;119(2):104-9.
venous thromboembolism. Circulation. 2004; 25. Bounameaux H, de Moerloose P. Is laboratory
(9 suppl 1):I3-9. monitoring of low-molecular-weight heparin
17. Heparin Sodium Injections [package insert]. therapy necessary? No. J Thromb Haemost.
Schaumburg, IL: APP Pharmaceuticals, 2004;2:551-4.
LLC, 2008. http://www.apppharma.com 26. The Columbus Investigators. Low-molecular-
/our-products/alphabetical/product-55.html. weight heparin in the treatment of patients
18. Raschke RA, Reilly BM, Guidry JR, Fontana with venous thromboembolism. N Engl J
JR, Srinivas S. The weight-based heparin Med. 1997;337:657-62.
dosing nomogram compared with a standard 27. Mismetti P, Quenet S, Levine M, Merli G,
care nomogram. A randomized controlled Decousus H, Derobert E, et al. Enoxaparin
trial. Ann Intern Med. 1993;119(9):874-81. in the treatment of deep vein thrombosis
19. Cruickshank MK, Levine MN, Hirsh J, with or without pulmonary embolism: an
Roberts R, Siguenza M. A standard heparin individual patient data meta-analysis. Chest.
nomogram for the management of heparin 2005;128:2203-10.
therapy. Arch Intern Med. 1991;151(2):333-7. 28. Quinlan DJ, Mcquillan A, Eikelboom JW.
20. Kearon C, Ginsberg JS, Julian JA, Douketis Low-molecular-weight heparin compared
J, Solymoss S, Ockelford P, et al; Fixed-Dose with intravenous unfractionated heparin for
Heparin (FIDO) Investigators. Comparison treatment of pulmonary embolism: a meta-
of fixed-dose weight-adjusted unfractionated analysis of randomized, controlled trials. Ann
heparin and low-molecular-weight heparin for Intern Med. 2004;140:175-83.
acute treatment of venous thromboembolism. 29. OBrien JA, Caro JJ. Direct medical cost of
JAMA. 2006;296(8):935-42. managing deep vein thrombosis according

to the occurrence of complications. 35. Buller HR, Davidson BL, Decousus H,

Pharmacoeconomics. 2002;20:603-15. Gallus A, Gent M, Piovella F, et al; Matisse
30. Lovenox [package insert]. Greenville, NC: Investigators. Fondaparinux or enoxaparin
Sanofi-Aventis U.S. LLC, 2008. http:// for the initial treatment of symptomatic
products.sanofi-aventis.us/lovenox/lovenox deep vein thrombosis. Ann Intern Med.
.html. 2004;140:867-73.
31. Fragmin [package insert]. New York, NY: 36. Arixtra [package insert]. Research Triangle
Pfizer Inc, 2007. http://www.eisai.com Park, NC: GlaxoSmithKline, 2008. http://
/package_inserts/Fragmin_PI.pdf. us.gsk.com/products/assets/us_arixtra.pdf.
32. Innohep [package insert]. Summit, NJ: Leo 37. Refludan [package insert.] Montville, NJ:
Pharma A/S, 2008. Berlex, 2004. http://www.bayer.ca/files
33. Prandoni P, Lensing AW, Pesavento R. New /REFLUDAN-PM-ENG-29MAY2007-113596
strategies for the treatment of acute venous .pdf?#.
thromboembolism. Semin Thromb Hemost. 38. Argatroban [package insert]. Research
2006;32(8):787-92. Triangle Park, NC: GlaxoSmithKline,
34. The Matisse Investigators. Subcutaneous 2009. http://us.gsk.com/products/assets/
fondaparinux versus intravenous us_argatroban.pdf.
unfractionated heparin in the initial treatment 39. Angiomax [package insert]. Bedford, OH:
of pulmonary embolism. N Engl J Med. BenVenue Laboratories, 2005. http://www
2003;349:1695-702. .angiomax.com/Files/SalesAidRef/PI.pdf.

CHAPTER
Pharmacological Thrombolysis
Indications, Techniques, and outcomes
Mark g. Davies
C urrent therapy for extensive deep ve-

nous thrombosis (DVT) in both the
upper and lower extremities is evolving into
Physiology
Fibrinolysis refers to the dissolution of the
fibrin network that forms the supporting lat-
a more aggressive lytic-based approach with ticework of a thrombus. Formation of fibrin
the goal of rapid clearance of thrombus bur- is a stimulus for activation of fibrinolysis.
den. Lytic therapy can now be achieved by Plasminogen must be converted to plasmin
catheter-directed lysis, ultrasound-assisted for fibrinolysis to occur.1 Plasminogen (glu-
catheter-directed lysis, mechanical throm- plasminogen) binds to endothelial cells and
bectomy, and pharmacomechanical lysis is converted to a form (lys-plasminogen)
(power-pulse spray). The single common that is more efficiently activated. Plasmin
element in these techniques is a throm- degrades cross-linked fibrin, non-cross-
bolytic agent. This chapter is focused on linked fibrin, and fibrinogen. Degradation
thrombolysis and its outcomes in DVT and of non-cross-linked fibrin and fibrinogen
will complement the subsequent chapters on results in the production of fibrin degrada-
mechanical thrombolysis, power-pulse spray, tion products A, B, D, and E. Degradation
and ultrasound-assisted lysis. of cross-linked fibrin is slower because of the
presence of cross-linkages, which results in
noncovalently bound fragments (DD and
EE, the D-dimers). These latter fragments
of fibrin can be considered markers of true
Davies md and Alan B. Lumsden md, eds. fibrin degradation. Endothelial cells synthe-
Cardiotext Publishing, ISBN 978-1-935395-22-5. size the plasminogen activators as single-
79

chain proteins and then secrete them. These which retains clottability but forms clots
single-chain proteins then assemble into with reduced tensile strength that stimulate
functional complexes and act as serine prote- plasminogen activation by tPA more than
ases. There are 2 forms of plasminogen acti- fibrin clots.4 It accumulates after treatment
vators (PA): the urokinase type (uPA), which with tPA but not with tPA given with 2 anti-
activates plasminogen in the fluid phase, plasmin. It does not accumulate with use of
and tissue PA (tPA), which is most active more fibrin-specific agents.5
when bound to fibrin.2,3 Normal endothelial
cells express tPA. However, if stimulated by
a variety of cytokines and other factors, they Pharmacology
preferentially synthesize uPA and down-
regulate tPA synthesis. In addition to these Thrombolysis is the pharmacological dis-
2 fibrinolytic enzymes, endothelial cells solution of fibrin thrombus by exogenously
also secrete 2 PA inhibitors, PAI-1 and PAI- delivered agents. The characteristics of these
2. Both are serine protease inhibitors and agents are shown in Table 7.1. These agents
form equimolar complexes with active uPA form plasmin, which leads to the degradation
or tPA molecules. PAI-1 requires the pres- of fibrin, fibrinogen, factor V, and factor VII.
ence of vitronectin in the extracellular ma- Plasminogen is found free in the circulation
trix to maintain its active conformation and and within the thrombus, which allows for
is therefore inactive outside the matrix. The activation of plasmin within a thrombus by
thrombin-thrombomodulin pathway also regional perfusion. Streptokinase binds plas-
is involved in the regulation of fibrinolysis. minogen to form an active complex, which
Thrombin activatable fibrinolysis inhibitor then activates another plasminogen to form
(TAFI), a carboxypeptidase, is a plasma pro- plasmin. Use of streptokinase is associated
tein that is a substrate, like protein C, for the with an increased incidence of complications
thrombin thrombomodulin complex. TAFIa such as allergic reactions or hemorrhages and
suppresses glu-plasminogen but not lys-plas- has fallen out of favor. The success of uroki-
minogen activation by tPA in the presence nase was due to its nonantigenic properties
of a fibrin analogue that had been exposed and its direct activation of plasminogen with-
to plasmin (clotted fragment X). Through out the formation of an intermediate complex.
the activation of protein C and TAFI, the As a nonspecific activator, however, it can, in
thrombin/thrombomodulin complexes can high doses, induce systemic fibrinolysis. rtPA
down-regulate coagulation and fibrinolysis. or alteplase is a recombinant version of the
The general schema of fibrinolysis is shown naturally occurring tPA protein; its activity is
in Figure 7.1. Bleeding, the most serious enhanced by the presence of fibrin. The final
complication of thrombolytic therapy with product is recombinant plasminogen activa-
tissue-type plasminogen activator (tPA), is tor rPA or reteplase. rPA is a recombinantly
thought to result from lysis of fibrin in hemo- derived mutant lacking the 3 nonprotease do-
static plugs and from the systemic lytic state mains of rtPA. A comparison of the efficacy,
caused by unopposed plasmin. One mecha- safety, and costs associated with catheter-
nism by which systemic plasmin can impair directed thrombolysis (CDT) with urokinase
hemostasis is by partially degrading fibrino- (UK) and the recombinant agents alteplase
gen to fragment X, a high-molecular-weight (tissue plasminogen activator [tPA]) and re-
clottable fibrinogen degradation product, teplase (recombinant plasminogen activator

Pharmacological Thrombolysis 81
Figure 7.1 Control of intravascular coagulation.The endothelium coordinates the procoagulant and anticoagulant
pathways to maintain vascular fluidity. It synthesizes several factors: protein S, urokinase-like plasminogen activator
(uPA), tissue plasminogen activator (tPA), and plasminogen activator inhibitors (PAI). uPa and PAI are localized on
the membrane by uPAR. Several pathways involved in fibrinolysis and anticoagulation reside on the membrane of the
endothelial cells (plasminogen activator/inhibitors and the protein C pathway). Clot is formed by the activation of
tissue factor, which leads to thrombin activity that cleaves fibrinogen to fibrin and is quickly associated with activated
platelets to form a secondary hemostatic plug. At this time the plasmin-activating system is inhibited at various levels
by the cells. Once fibrinolysis is activated, plasmin is formed from plasminogen by the actions of tPA and uPA whose
activity is in turn modulated by PAI-1 and 2 antiplasmin. Activated plasmin breaks fibrin into fibrin degradation
products, one of which is termed fragment X and can lead to clot instability in the presence of long-term infusions of
exogenous plasminogen activators.
[RPA]) in the treatment of symptomatic deep eral third-generation thrombolytic agents

venous thrombosis (DVT) reveals that each have been developed.8 They are either con-
of the agents is safe and effective, but that jugates of plasminogen activators with mono-
the new recombinant agents are significantly clonal antibodies against fibrin, platelets, or
less expensive than urokinase.6 First-gener- thrombomodulin; mutants, variants, and
ation thrombolytics (streptokinase and uro- hybrids of alteplase (reteplase, tenecteplase,
kinase) have no fibrin-binding capabilities and pamiteplase) and prourokinase (ame-
and cause systemic plasminogen activation diplase); or new molecules of animal (vam-
with concomitant destruction of haemostatic pire bat) or bacterial (Staphylococcus aureus)
proteins.7 A primary driving force behind origin. These variations may lengthen the
the development of the second-generation drugs half-life, increase resistance to plasma
plasminogen activator tissue plasminogen protease inhibitors, or cause more selec-
activator (tPA or alteplase) was its ability to tive binding to fibrin. Compared with the
bind to fibrin and target thrombolysis. Sev- second-generation agent (alteplase), third-

Table 7.1 Pharmacology
Streptokinase Urokinase tPA rPA
Size(kD) 45,00050,000 20,00034,000 63,00065,000 39,000
Source -hemolytic Fetal renal culture Recombinant Recombinant

Streptococcus
Activity SK-plasminogen Direct PA Direct PA Direct PA

complex
Metabolism Liver Liver Liver Liver
1/2life 1883 minutes 15 minutes 4 minutes 15 minutes
Fibrinselective Low Low High Very high
Affinity-free High High Low Very low

plasminogen
Therapeutic Wide Wide Narrow Wide

window
Cost Low High High High
Availability Yes No Yes Yes
Advantages Expense No allergy No allergy No allergy
Disadvantages Allergy No available Expense Expense
generation thrombolytic agents such as mon- which can occur months to years after the
teplase, tenecteplase, reteplase, lanoteplase, initial event.9 Conventional therapy with
pamiteplase, and staphylokinase result in a heparin has been effective in diminishing
greater angiographic patency rate in patients and preventing the propagation of throm-
with acute myocardial infarction, although, bus;10 however, it does not diminish the de-
thus far, mortality rates have been similar velopment of postthrombotic syndrome.11
for those few drugs that have been studied In fact, it has been estimated that 30% to
in large-scale trials. Bleeding risk, however, 40% of existing thrombi will demonstrate
may be greater. propagation even at therapeutic levels of
heparin.12 A pooled analysis of 13 studies by
Comerota13 showed that 4% of patients who
Indications received heparin therapy had significant or
complete lysis of thrombus compared with
Systemic anticoagulation has been the main- 45% who were randomized to systemic strep-
stay of therapy to prevent the occurrence of tokinase therapy. The ideal therapy should
pulmonary embolism. However, the long- dissolve existing thrombus, restore flow, and
term effects of venous thromboembolism preserve venous valve function, which will
have largely gone unnoticed, specifically, the diminish the incidence of venous hyperten-
development of postthrombotic syndrome, sion. Catheter-directed venous thrombolysis

Figure 7.2 Algorithm for thrombolysis of venous thrombotic disease.In the patient presenting with a diagnosed
deep venous thrombosis in the lower extremity, the acuity of the lesion, the extent of the thrombosis, and the impact
of the thrombosis on distal circulation must be considered. In the absence of contraindications, thrombolysis can be
instituted. The use of an adjunct inferior vena cava (IVC) filter is institutional-dependent. Lysis of the clot may reveal a
lesion that will require intervention, either endoluminally or surgically. With resolution or no documented improvement,
thrombolysis may be stopped and a standard heparin/oral anticoagulation protocol instituted.
has the potential to meet these goalsin of postthrombotic syndrome and its associ-
particular, the retention of venous valvular ated complications.
function, which can prevent the formation Percutaneous clot removal using throm-

bolysis, mechanical thrombectomy, or a com- Relative

bination of the 2 is fast becoming a treatment Major
of choice for patients presenting with acute Recent (<10 days) major surgery,
iliofemoral and axillosubclavian deep venous obstetric delivery, or organ
thrombosis.14,15 By restoring venou s patency biopsy
and preserving valvular function, catheter- Active peptic ulcer or GI
directed thrombolytic therapy potentially pathology
affords an improved long-term outcome in Recent major trauma
selected patients with DVT. In selected pa- Uncontrolled hypertension
tients with extensive acute proximal DVT Minor
(ie, iliofemoral or axillosubclavian DVT, Minor surgery or trauma
symptoms for <14 days, good functional sta- Recent CPR
tus, life expectancy of >1 year) who have a High likelihood of left heart
low risk of bleeding, catheter-directed phar- thrombus
macomechanical thrombolysis is suggested Bacterial endocarditis
by the American College of Chest Physicians Hemostatic defects
(ACCP).16 An algorithm for thrombolysis of Pregnancy
venous thrombotic disease is shown in Fig- Diabetic hemorrhagic retinopathy
ure 7.2. This percutaneous therapy should Agent Specific
include correction of any underlying venous Streptokinase
lesions. Patients suitable for catheter-directed Known allergy
thrombolysis would include young, active Recent streptococcal infection
individuals who have an acute (<14 days) il- Previous therapy within 6 months
iofemoral and axillosubclavian deep venous
thrombosis. Patients who have signs/symp-
toms of phlegmasia cerulea dolens, regard-
less of their clinical condition or age, should Techniques
be considered for thrombolytic therapy. The
contraindications for the use of a throm- Catheter-Directed Thrombolysis
bolytic agent in any given scenario can be
considered to fall into general and disease- Although a variety of access sites have been
specific categories. The general contraindi- utilized, the most common site reported is
cations can be broken down into absolute, the popliteal vein.17 Earlier techniques em-
relative, and minor. ployed the internal jugular vein or common
femoral vein; however, traversal of the oc-
cluded vein segment can be difficult due to
retrograde passage through the venous valves,
Contraindications which may make further manipulations
to Thrombolysis problematic. Using a popliteal approach, the
Absolute patient is typically placed in a prone position
Active internal bleeding on the fluoroscopy table, and the popliteal
Recent cerebrovascular accident vein is localized using ultrasound guidance.
(<2 months) Access is obtained with a small needle under
Intracranial pathology ultrasound guidance and a 4F or 5F sheath is

D Figure 7.3 Case of a femoropopliteal DVT with spon-

taneous hematoma. A.Venogram showing a proximal
femoral DVT (indicated by arrow). B. Infusion catheter in
place (indicated by arrow). C. Final venogram showing
a patent IVC and iliofemoral system. D. Spontaneous
pelvic hematoma, which occurred 18 hours after com-
mencing the infusion.
introduced. Baseline venography is then per- Low-dose tPA continuous infusion at 0.5 to
formed. After passage of a 0.035-in guidewire 1.0 mg/h is administered through the infu-
through the occluded venous segment, a di- sion sheath with intravenous heparin at a
agnostic catheter is advanced past the occlud- rate of 500 U/h. Others recommend giving
ed segment. This catheter is then exchanged a lacing bolus of tPA 3 to 5 mg followed by
for a multiple side hole infusion catheter. It a continuous tPA infusion at 0.05 mg/kg/h.
is important that the total volume of throm- Adding intermittent pneumatic compres-
bus is traversed with an infusible segment of sion to CDT for DVT treatment of the leg
catheter so that the thrombus can come into resulted in better early and late outcomes
direct contact with the thrombolytic agent. compared with CDT alone and was not asso-

ciated with an increased risk of symptomatic strates that a larger proportion of patients
pulmonary emboli.18 with acute deep venous thrombosis (86%)
vs. chronic (68%) were able to achieve grade
II or III lysis. This is important because the
Monitoring degree of lysis was found to be predictive of
Currently, there is no need to monitor the early and continued patency. For instance,
changes in coagulation induced by low-dose 75% of limbs with complete lysis remained
thrombolytics after the physician has estab- patent at 1 year compared to 32% of limbs
lished a baseline coagulation profile. Many in which there was <50% lysis. A presenta-
centers will follow fibrinogen and seek to tion of acute deep venous thrombosis (<10
maintain a target of >200 mg/dL. The pa- days) was predictive of a better lysis grade
tient does require monitoring of the infu- when compared with chronic deep venous
sion site to ensure that there is no exit site thrombosis, although a significant percent-
bleeding or disruption of the thrombolytic age of patients with chronic iliofemoral
delivery system. Progress in thrombolysis deep venous thrombosis were found to have
may be assessed by interval duplex ultraso- significant lysis (grade IIIII). The cathe-
nography or more commonly by repeat ve- ter-directed group demonstrated a much
nography. Repeat imaging is performed at higher degree of significant lysis (83% vs.
8- to 12-hour intervals, and infusions may be 20%) when compared with the 19% of pa-
continued as long as 72 hours, although the tients who had a pedal infusion. This point
mean infusion time is generally 56 hours.19 underscores the necessity of having the
The incidence of major clinical hemorrhage thrombus in direct contact with the throm-
after fibrinolysis for DVT is between 6% and bolytic agent. Adjunctive procedures with
30%, a 3-fold increase compared to standard stent placement and/or venoplasty were per-
heparin therapy. Infusion of thrombolysis is formed in 33% of affected limbs, which indi-
associated with increased risk of local he- cates that an underlying lesion often needs
matoma formation at the site of catheter in- to be treated in conjunction with throm-
sertion and a lower risk of distant bleeding bolysis. Major bleeding complications were
(Figure 7.3). Hypersensitivity reactions are noted in 4% of patients enrolled. Subgroup
low with both tPA or rPA. analysis reveals that the best results occurred
in patients with acute iliofemoral deep ve-
nous thrombosis with no prior history of
Outcomes deep venous thrombosis. Complete lysis oc-
curred in 65% of patients with 96% patency
In the National Multicenter Venous Reg- at l year. Conversely, the worst results were
istry,17 catheter-directed thrombolysis was seen in patients with chronic femoral-pop-
performed on acute (66%), chronic (45%), liteal deep venous thrombosis. The groups
and a mixture of acute and chronic throm- were not large enough to establish statistical
bus (19%). The degree of lysis achieved was significance, but the investigators believed
categorized as grade I (<50% lysis), grade II these results could serve as a guide to patient
(50%90% lysis), and grade III (complete selection for catheter-directed thrombolysis.
lysis). Continued patency was followed at 3, Patients most likely to achieve complete lysis
6, and 12 months with duplex ultrasonog- and increased patency were those with an
raphy. Analysis of the registry data demon- acute iliofemoral deep venous thrombosis

and no prior history of deep venous throm- eligibility criteria has improved the safety
bosis. Follow-up data are limited because a and acceptability of this treatment. The
significant proportion of patients were lost optimum drug, dose, and route of adminis-
to follow-up at 12 months.17 Following treat- tration have yet to be determined. A more
ment, patients receiving catheter-directed recent review of articles on catheter-directed
thrombolysis reported better overall physi- thrombolysis (CDT) from PubMed and the
cal functioning, less stigma, less health dis- Cochrane library was recently performed.23
tress, and fewer postthrombotic symptoms CDT reduced clot burden and DVT recur-
compared to those patients treated with rence and may prevent the formation of
anticoagulation alone.20 Within the throm- postthrombotic syndrome. Indications for
bolysis group, successful lysis correlated with its use include younger individuals with a
health-related quality of life.20,21 long life expectancy and few comorbidities,
The results of the current Cochrane limb-threatening thromboses, and proximal
database included 12 studies.22 Complete iliofemoral DVTs. These results suggest that
clot lysis occurred significantly more often the outcomes of CDT in DVT management
in the treatment group in early follow-up are encouraging in selected patient cohorts,
(relative risk [RR] 0.24; 95% confidence but further evidence is required to establish
interval [CI] 0.070.82) and in late follow- longer-term benefits and cost-effectiveness.
up (RR 0.37; 95% CI 0.25 0.54). A similar There is a marked lack of randomized
effect was also seen for any degree of im- controlled trials comparing CDT-related
provement in venous patency. Significantly mortality and long-term outcomes com-
less postthrombotic syndrome occurred in pared to anticoagulation alone. The current
those receiving thrombolysis (RR 0.66; 95% ATTRACT trial sponsored by the NIH will
CI 0.470.94). Leg ulceration was reduced, hopefully answer many of the questions re-
although the data were limited by small lated to catheter-directed thrombolysis.
numbers (RR 0.53; 95% CI 0.122.43). Ve-
nous function was improved at late follow- references
up, but not significantly (RR 0.43; 95% CI 1. Gertler JP, Abbott WM. Prothrombotic and
0.063.17). Out of 668 patients, those re- fibrinolytic functions of normal and perturbed
ceiving thrombolysis had significantly more endothelium. J Surg Res. 1992;52:89-95.
bleeding complications (RR 1.73; 95% CI 2. Henkin K, Marcotte P, Yang H. The
1.042.88). Two strokes occurred in the plasminogen-plasmin system. Prog Cardiovasc
treatment group (RR 1.70; 95% CI 0.21 Dis. 1991;34:135-62.
13.70). The incidence of bleeding appears to 3. Robbins K. The Plasminogen-Plasmin
have reduced over time with the introduc- Enzyme System. New York: Lippincott; 1995.
tion of stricter selection criteria. There was 4. Schaeerf AV, Leslie BA, Rischke JA, Stafford
no significant effect on mortality detected AR, Fredenburgh JC, Weitz JI. Incorporation
in either early or late follow-up. Data on of fragment X into fibrin clots renders
occurrence of pulmonary embolism (PE) them more susceptible to lysis by plasmin.
and recurrent DVT were inconclusive. The Biochemistry. 2006;45(13):4257-65.
conclusions were that thrombolysis appears 5. Weitz JI. Limited fibrin specificity of tissue-
to offer advantages in terms of reducing type plasminogen activator and its potential
postthrombotic syndrome and maintaining link to bleeding. J Vasc Interv Radiol. 1995;6
venous patency after DVT.22 Use of strict (Pt 2 suppl):19S-23S.

6. Grunwald MR, Hofmann LV. Comparison 16. Kearon C, Kahn SR, Agnelli G, Goldhaber S,
of urokinase, alteplase, and reteplase for Raskob GE, Comerota AJ; American College
catheter-directed thrombolysis of deep of Chest Physicians. Antithrombotic therapy
venous thrombosis. J Vasc Interv Radiol. for venous thromboembolic disease: American
2004;15(4):347-52. College of Chest Physicians Evidence-Based
7. Longstaff C, Williams S, Thelwell C. Fibrin Clinical Practice Guidelines (8th ed.). Chest.
binding and the regulation of plasminogen 2008;133(6 suppl):454S-545S.
activators during thrombolytic therapy. 17. Mewissen M, Seabrook G, Meissner M,
Cardiovasc Hematol Agents Med Chem. Cynamon J, Labropoulos N, Haughton S.
2008;6(3):212-23. Catheter-directed thrombolysis for lower
8. Verstraete M. Third-generation thrombolytic extremity deep venous thrombosis: report
drugs. Am J Med. 2000;109(1):52-8. of a nation multicenter registry. Radiology.
9. Strandness DE Jr, Langlois Y, Cramer M, 1999;211:39-49.
Randlett A, Thiele BL. Long-term sequelae 18. Ogawa T, Hoshino S, Midorikawa H, Sato
of acute venous thrombosis: a clinical review. K. Intermittent pneumatic compression of
JAMA. 1983(250):1289-92. the foot and calf improves the outcome of
10. Johnson BF, Manzo RA, Bergelin RO, catheter-directed thrombolysis using low-dose
Strandness DE Jr. Relationship between urokinase in patients with acute proximal
changes in the deep venous system and the venous thrombosis of the leg. J Vasc Surg.
development of the post-thrombotic syndrome 2005;42(5):940-4.
after an acute episode of lower limb deep 19. Mewissen M, Seabrook GR, Meissner MH,
venous thrombosis: a one-to-six year follow-up. Cynamon J, Labropoulos N, Haughton SH.
J Vasc Surg. 1995;21:307-12. Catheter-directed thrombolysis for lower
11. Meissner MH, Manzo RA, Bergelin RO, extremity deep venous thrombosis: report
Markel A, Strandness DE Jr. Deep venous of a nation multicenter registry. Radiology.
insufficiency: the relationship between 1999;211:39-49.
lysis and subsequent reflux. J Vasc Surg. 20. Comerota AJ. Quality-of-life improvement
1993;18:596-608. using thrombolytic therapy for iliofemoral
12. Strandness DE. Thrombus propagation deep venous thrombosis. Rev Cardiovasc Med.
and level of anticoagulation. J Vasc Surg. 2002;3 (suppl 2):S61-7.
1990;12:497-8. 21. Comerota AJ, Throm RC, Mathias SD,
13. Comerota A, Aldridge S. Thrombolytic Haughton S, Mewissen M. Catheter-directed
therapy for deep vein thrombosis: a clinical thrombolysis for iliofemoral deep venous
review. Can J Surg. 1993;36:359-64. thrombosis improves health-related quality of
14. Meissner MH, Wakefield TW, Ascher E, life. J Vasc Surg. 2000;32(1):130-7.
Caprini JA, Comerota AJ, Eklof B, et al. Acute 22. Watson LI, Armon MP. Thrombolysis for acute
venous disease: venous thrombosis and venous deep vein thrombosis. Cochrane Database
trauma. J Vasc Surg. 2007;46 (suppl S):25S- Syst Rev. 2004; Oct 18(4):CD002783.
53S. 23. Gogalniceanu P, Johnston CJ, Khalid U, Holt
15. Molina JE, Hunter DW, Dietz CA. PJ, Hincliffe R, Loftus IM, et al. Indications
Protocols for Paget-Schroetter syndrome for thrombolysis in deep venous thrombosis.
and late treatment of chronic subclavian Eur J Vasc Endovasc Surg. 2009;38(2):
vein obstruction. Ann Thorac Surg. 192-8.
2009;87(2):416-22.

CHAPTER
8
Pharmacomechanical Thrombolysis
for Acute Deep Venous Thrombosis
Indications, Techniques, and Clinical Data
elina Quiroga and Mark H. Meissner
T he primary goals of treatment of acute

deep venous thrombosis are the preven-
tion of pulmonary embolism, recurrent deep
Early strategies focused on the systemic ad-
ministration of thrombolytic agents, which
were 3.7 times more effective in restoring
venous thrombosis (DVT), and postthrom- venous patency than anticoagulation alone.2
botic syndrome (PTS). Although standard Unfortunately, systemic thrombolysis was
anticoagulation is effective in preventing associated with prolonged infusion times as
recurrent venous thromboembolism, it pro- well as significant rates of bleeding and par-
tects imperfectly against postthrombotic tial thrombolysis.3
syndrome, which may develop in 29.6% of Efforts to reduce infusion times, im-
patients by year 5.1 As an adjunct to anticoag- prove thrombolytic efficiency, and reduce
ulation, early thrombus removal, including bleeding complications led to the develop-
thrombolytic therapy and venous thrombec- ment of catheter-directed thrombolytic strat-
tomy, likely plays a role in reducing the inci- egies. Although these strategies were also
dence of postthrombotic syndrome. associated with significant infusion times
As venous thrombolytic strategies can and bleeding complications, they were criti-
rapidly restore venous patency, they have the cal in establishing the optimal patient pop-
potential to preserve valve function and the- ulation to be considered for thrombolysis
oretically prevent postthrombotic syndrome. and in developing modern interventional
techniques for the treatment of acute DVT.
Such strategies established the importance
of thrombolysis via an ultrasound-guided
Davies md and Alan B. Lumsden md, eds. antegrade puncture of the popliteal artery;
Cardiotext Publishing, ISBN 978-1-935395-22-5. the frequency of underlying iliac obstruc-
89

tive lesions and need for treatment; and being alternatives. Initial venography is per-
the poor results of lytic therapy for chronic formed to document the extent of thrombus
thrombosis.4 and a wire is advanced across the thrombus.
The limitations of both systemic and If caval thrombus is suspected, some also
catheter-directed thrombolysis have led to recommend a computed tomography (CT)
a variety of pharmacomechanical strategies venogram to document thrombus extent
directed toward increasing lytic efficiency prior to bringing the patient to the angiog-
while reducing procedural times. Theo- raphy suite. Further aspects of the procedure
retically, this would translate into reduced are more device-specific.
bleeding complications, decreased hospital The AngioJet rheolytic thrombectomy
and intensive care unit stays, and reduced catheter (MEDRAD Inc., Warrendale, PA)
cost. is an over-the-wire system using a high-
velocity (350/450 km/h) saline jet directed
backward from the tip of the catheter. A
Pharmacomechanical low-pressure zone is created via a Venturi
Thrombolytic Strategies effect, causing fragmentation of the throm-
bus, which is aspirated through the efflu-
The isolated use of mechanical devices is ent lumen (Figure 8.1). In the power-pulse
rarely successful, and their use in combi- mode, the effluent port is closed with a stop-
nation with thrombolytic drugs is usually cock and a dilute lytic agent, usually tPA at
required.5 However, use of a mechanical a dose of 0.4 mg/cc is used in place of the
device to speed thrombolysis is theoretically saline jet. After infusion, the lytic agent is
attractive since it has the potential to reduce allowed to dwell for 20 to 25 minutes prior
thrombolytic infusion times and thereby to aspiration on withdrawal of the catheter
decrease bleeding complications, hospital with the effluent port open. A variety of
resource utilization, and cost while improv- catheter sizes are available, allowing treat-
ing efficacy. Although there is little solid ment of different vessels.6
data that this has been achieved, use of such The Trellis-8 infusion catheter (Covi-
devices would optimally allow the entire ve- dien) has proximal and distal occlusion bal-
nous system to be cleared of thrombus in a loons with a drug-infusion port between the
single session in the angiography suite. balloons, and a wire that oscillates within the
Amongst several mechanical throm- catheter (Figure 8.2). This allows the throm-
bolytic devices, 3 have been widely utilized bolytic drug to be isolated between balloons
in the venous systemthe AngioJet power in direct contact with the thrombus. The
pulse, the Trellis-8 infusion catheter, and the oscillating wire macerates the thrombus, in-
EKOS EndoWave. The mechanical throm- creasing exposure to the thrombolytic agent
bolytic devices share many technical details. and theoretically reducing the dose required
If inflow to the popliteal vein is patent, ul- for lysis. After treatment, the thrombolytic
trasound-guided puncture of the popliteal agent and clot fragments are aspirated. Sev-
vein using a micropuncture set is preferred eral thrombolytic agents had been used, in-
by many interventionalists, with more distal cluding tenecteplase, with doses between 5
access via the posterior tibial vein, small sa- and 10 mg and tPA, with a recommend dose
phenous vein, or great saphenous vein (with of 5 to 10 mg. The current available cath-
deep system access through a perforator) eters are 80 and 120 cm in length, and the

Pharmacomechanical Thrombolysis for Acute Deep Venous Thrombosis 91
Figure 8.1 AngioJet rheolytic thrombectomy catheter. Image Courtesy of MEDRAD Interventional.
Adjunctive procedures are a critical

component of pharmacomechanical throm-
bolysis. The Venous Registry reported a low
incidence (1%) of symptomatic pulmonary
embolism with catheter-directed thromboly-
sis, and it is generally accepted that routine
placement of an IVC filter is not required in
this setting. However, there is not yet con-
sensus regarding placement and use of IVC
filters in pharmacomechanical thromboly-
sis. Pulmonary embolism has been reported
Figure 8.2 Trellis-8 Peripheral Infusion System. Image in association with mechanical thrombo-
courtesy of Covidien.
lytic devices, although the precise incidence
is unknown. Some have suggested the rou-
tine placement of an IVC filter when using
the AngioJet catheter.9 However, given the
treatment zone varies between balloons of relatively short period of risk during the ac-
10, 15, and 30 cm.7 tual use of the mechanical devices, use of
The EKOS EndoWave (EKOS Corpo- a retrievable filter is likely more appropri-
ration, Bothell, WA) device utilizes ultra- ate than a permanent filter if this is deemed
sound transducers to alter the structure of appropriate.
the thrombus, increasing its permeability Thrombus maceration using an angio-
and exposing plasminogen receptors (Figure plasty balloon is also used by some to assist
8.3). The ultrasound forces the drug into the with lytic exposure. Thrombolysis will un-
thrombus and may help to hold the drug cover an underlying iliac vein lesion in 45%
within the thrombosed segment. tPA at a to 60% of patients presenting with a left-side
dose of 0.5 to 1.0 mg/h is among the most iliofemoral DVT.4,10 Such lesions most com-
commonly use thrombolytic agents.8 monly occur at the crossing of the left com-

treatment. If follow-up catheter-directed

thrombolysis is required, current guidelines
suggest tPA infusion at a rate of 0.5 to 1.0
mg/h. High-volume infusions are generally
preferred in the venous system, and follow-
up venography is indicated at 12 to 24 hours.5
Systemic anticoagulation, usually with un-
fractionated heparin, is utilized during the
procedure, usually at a reduced dose to avoid
bleeding complications, and conventional
anticoagulation therapy is required after the
procedure to prevent rethrombosis. The du-
Figure 8.3 EKOS EndoWave device. Image Courtesy ration of anticoagulation should be based
of EKOS. on the patients underlying risk factors and
should follow current ACCP guidelines.12 As
prevention of the postthrombotic syndrome
is the primary goal of pharmacomechani-
mon iliac vein by the right common iliac cal thrombolysis, all patients should be dis-
artery (the May-Thurner or Cockett syn- charged with knee-high 30 to 40mm Hg
drome), but may occur on the right side as compression stockings that should be contin-
well.11 Failure to treat such lesions with self- ued for at least 2 years.12,13
expanding stents is associated with a high
rate of recurrent thrombosis and thrombolyt-
ic failure.4 Current studies suggest that when
an iliac vein stenosis is uncovered, self-ex- Results of
pandable metallic stents improve 1-year ve-
nous patency rates from 53% to 74% among Pharmacomechanical
limbs treated with metallic stents. Although
the data regarding infrainguinal stents are
Thrombolysis
limited, their use is not currently indicated. Unfortunately, the outcomes after phar-
The Venous Registry reported that 4 of 5 macomechanical thrombolysis are entirely
infrainguinal stents occluded early after derived from case series, and there have
placement.4 Although lacking data, it also been no randomized clinical trials com-
seems likely that a stented and freely reflux- paring pharmacomechanical thrombolysis
ing femoropopliteal venous segment may be with either catheter-directed thrombolysis
more detrimental hemodynamically than a or conventional anticoagulation. Further-
chronically obstructed segment. more, such series tend to focus on surro-
The AngioJet and Trellis system theo- gate outcomes such as degree of lysis rather
retically allow treatment in a single setting, than clinically important outcomes such as
although placement of an infusion catheter quality of life or objectively defined post-
for completion of thrombolysis is not infre- thrombotic syndrome. Many outstanding
quently required. Although the EndoWave questions regarding pharmacomechanical
catheter may speed the lytic process, it does thrombolysis are expected to be answered
not have the potential for single-setting by the ATTRACT trial, which will compare

Pharmacomechanical Thrombolysis for Acute Deep Venous Thrombosis 93
these pharmacomechanical strategies with the potential to minimize treatment times,

conventional anticoagulation. The currently with corresponding decreases in cost and
available data, although weak, do however bleeding complications. Unfortunately, such
show a promising trend favoring pharma- approaches have not yet been evaluated in
comechanical thrombolysis. rigorous randomized trials using clinically
As noted above, degree of lysis and relevant, objective measures of quality of life
bleeding are the most common outcomes and postthrombotic syndrome. Fortunately,
reported in pharmacomechanical series. the potential value of this approach is cur-
The percentage of lysis is usually based on rently being evaluated in the ATTRACT
the Society of Interventional Radiology clas- trial, which will evaluate both quality of life
sification: grade I, less than 50% thrombus and objectively determined postthrombotic
removal; grade II, 50% to 95%; and grade syndrome among patients randomized to
III, >95% to 100% thrombus removal. Grade pharmacomechanical techniques in com-
I or II lysis was achieved in 79% of patients parison to conventional anticoagulation.
receiving catheter-directed thrombolysis; in
93% of patients using the Trellis-8 device; references
and in 91% treated with EndoWave. Mean 1. Johnson BF, Manzo RA, Bergelin RO,
infusion times of 76 34 minutes and 22 Strandness DE Jr. Relationship between
11 minutes for the AngioJet and Trellis-8 changes in the deep venous system and the
devices, respectively, have been reported.10,14 development of the postthrombotic syndrome
Major bleeding complications were reported after an acute episode of lower limb deep vein
in 11.0% of patients treated with catheter- thrombosis: a one- to six-year follow-up. J Vasc
directed thrombolysis and 3.8% of patients Surg. 1995;21:307-13.
treated with EndoWave.4,8 There were no 2. Goldhaber SZ, Buring JE, Lipnick RJ,
reports of major bleeding complications Hennekens CH. Pooled analyses of
in patients treated with either the AngioJet randomized trials of streptokinase and
or Trellis-8 devices, although more rigor- heparin in phlebographically documented
ous methodological evaluations are clearly acute deep venous thrombosis. Am J Med.
needed. 1984;76:393-7.
3. Forster A, Wells P. Tissue plasminogen
activator for the treatment of deep venous
Conclusion thrombosis of the lower extremity. Chest.

2001;119:572-9.
Thrombolytic therapy for acute treatment 4. Mewissen MW, Seabrook GR, Meissner
of DVT is an attractive adjunct to conven- MH, Cynamon J, Labropoulos N, Haughton
tional anticoagulation, as it has the theoreti- SH. Catheter-directed thrombolysis of lower
cal ability to rapidly restore venous patency extremity deep venous thrombosis: report of
and preserve valvular function. However, a national multicenter registry. Radiology.
early experiences with systemic and cath- 1999;211:39-49.
eter-directed thrombolysis were character- 5. Vedantham, Thorpe PE, Cardella JF. Quality
ized by prolonged infusion times, significant improvement guidelines for the treatment of
rates of partial lysis, and high rates of bleed- lower extremity deep vein thrombosis with
ing complications. Pharmacomechanical use of endovascular thrombus removal. J Vasc
approaches are attractive in that they have Interv Radiol. 2009;20:S227-39.

6. Kasirajan K, Gray B. Percutaneous angiojet 11. Raju S, Neglen P. High prevalence of

thrombectomy in the management of nonthrombotic iliac vein lesions in chronic
extensive deep venous thrombosis. J Vasc venous disease: a permissive role in
Interv Radiol. 2001;12:179-85. pathogenicity. J Vasc Surg. 2006;44(1):
7. McLafferty R. Endovascular management of 136-43.
deep venous thrombosis. Perspect Vasc Surg 12. Kearon C, Kahn SR, Goldhaber S.
Endovasc Ther. 2008;20;87-91. Antithrombotic therapy for venous
8. Parikh S, Motarjeme A. Ultrasound-accelerat- thromboembolic disease: American College
ed thrombolysis for the treatment of deep vein of Chest Physicians Evidence-Based Clinical
thrombosis: Initial clinical experience. J Vasc Practice Guidelines (8th ed.). Chest. 2008;
Interv Radiol. 2008;19:521-8. 133 (6 suppl):454S-545S.
9. Lin PH, Zhou W, Dardik A. Catheter-direct 13. Brandjes D, Bller HR, Heijboer H,
thrombolysis versus pharmacomechanical Huisman MV, de Rijk M, Jagt H, et al.
thrombectomy for treatment of symptomatic Randomised trial of effect of compression
lower extremity deep venous thrombosis. Am J stockings in patients with symptomatic
Surg. 2006;192:782-8. proximal-vein thrombosis. Lancet.
10. OSullivan GJ, Semba CP, Bittner CA, Kee 1997;349:759-62.
ST, Razavi MK, Sze DY, et al. Endovascular 14. Hilleman DE, Razavi MK. Clinical and
management of iliac vein compression (May- economic evaluation of the Trellis-8 infusion
Thurner) syndrome. J Vasc Interv Radiol. catheter for deep vein thrombosis. J Vasc
2000;11:823-36. Interv Radiol. 2008;19:377-83.

CHAPTER
9
endovascular Intervention for lower
extremity Deep Venous Thrombosis
Techniques, Devices, and outcomes
erin H. Murphy and Frank r. Arko III
A nticoagulation remains the gold

standard for calf vein DVT.1-3 How-
ever, strategies of early intervention with
ultrasound-accelerated thrombolysis, and
percutaneous mechanical thrombectomy.
clot removal have been advocated for cases

of more proximal iliofemoral DVT. These Catheter-Directed
procedures are almost exclusively performed
endovascularly, with little downtime for the
Thrombolysis
patient and early clinical improvement. Early Catheter-directed thrombolysis (CDT) al-
clot removal has been shown to dramatically lows infusion of thrombolytics directly into
reduce the long-term morbidity of proximal the venous thrombosis, limiting systemic
DVT by preventing permanent venous val- drug exposure. Thrombolytic agents used
vular damage and thereby the debilitating with CDT include urokinase (ImaRx Thera-
sequelae of postthrombotic syndrome.4-7 peutics, Tucson, AZ), tissue plasminogen
Multiple techniques and devices are activator (Activase, Genentech, South San
available for endovascular DVT intervention. Francisco, CA), recombinant tissue plasmin-
The majority of available treatment options ogen activator (Retavase, PDL BioPharma,
may be separated by mechanism of action Fremont, CA), or tenecteplase (Genentech).
and include catheter-directed thrombolysis, Most commonly, patients treated with
CDT undergo percutaneous access in the
operating room with an initial venogram to
determine thrombus extent. A small infu-
Davies md and Alan B. Lumsden md, eds. sion catheter is placed just proximal to the
Cardiotext Publishing, ISBN 978-1-935395-22-5. location of thrombus and secured in place
95

externally. Patients are monitored in the brin composition. Increased permeability

intensive care unit, and thrombolytics are results in augmented lytic dispersion within
slowly administered through the catheter. the thrombus.15-17 In fact, a 65% increase
The patient undergoes repeat venography in the number of fibrin strands exposed to
in the operating room to assess clot lysis thrombolytic drugs has been demonstrated
once every 24 hours until complete lysis is to occur with a 44% ultrasound-mediated
achieved. reduction in the diameter of fibrin strands.16
This option has proven effective in This translates to increased thrombus up-
proximal DVT, resulting in early clot resolu- take of recombinant tissue plasminogen ac-
tion, prevention of PE, prevention of recur- tivator by 48%, 84%, and 89% at 1, 2, and
rent DVT, preservation of valve function, 4 hours, respectively.17 Furthermore, the ul-
and improved quality of life over treatment trasound waves penetrate past valves, allow-
with isolated anticoagulation. Results have ing for thrombus removal behind the valve,
demonstrated 60% to 90% clot resolution in which may be inaccessible with other PMT
proximal DVT8-11 with degree of clot remov- devices.
al correlating directly with improvements in The device consists of an infusion/as-
long-term patency and reduced incidence of piration catheter, an ultrasound core wire,
postthrombotic syndrome.11 and a drive unit. The catheter, available in
Unfortunately, this therapy is still as- treatment lengths of 6 to 50 cm, contains
sociated with significant bleeding complica- a central lumen that accommodates the
tions in 11% to 43% of patients.8-14 Ouriel et 0.035-ultrasound core wire and normal sa-
al. demonstrated insertion site bleeding in line infusate used for central cooling. In a
22% to 44%, transfusion requirements in triangular distribution around the central
12% to 22%, and intracranial hemorrhage in lumen are 3 separate infusion channels con-
0.6% to 3.0% of patients undergoing CDT taining microinfusion pores for drug deliv-
with urokinase and recombinant tissue plas- ery and thermocouples to monitor changes
minogen activator, respectively.14 Further in temperature and flow patterns. The ul-
limitations to the widespread use of this trasound core wire has transducers (2.2
technique include prolonged lytic infusion MHz) located at 1-cm intervals. When the
times of 36 to 72 hours, prolonged ICU stay, drive unit is activated, ultrasound waves are
and expensive drug costs.8-14 delivered to the core wire and transmitted
through the catheter, penetrating thrombus
and allowing lytic dispersion.
Ultrasound-Accelerated Access is obtained with a 5F introducer
Thrombolysis sheath and the lesion is crossed with a 0.035-
in guidewire. The catheter is positioned such
The EKOS EndoWave and EkoSonic sys- that the treatment zone extends through the
tems use low-power, high-frequency ul- length of the thrombosed venous segment.
trasound (2 MHz) in combination with After positioning, the guidewire is exchanged
catheter-directed thrombolysis to achieve for the ultrasound core wire. The 3 separate
clot disruption. Ultrasound waves generated drug-infusion lumens are primed with un-
by the unit do not directly macerate the clot fractionated heparin. The control unit is
but rather create microstreams that increase activated and delivers ultrasound energy via
thrombus permeability via alteration of fi- the core wire while the thrombolytic agent

Endovascular Intervention for Lower Extremity Deep Venous Thrombosis 97
of choice is administered through micro-

pores located throughout the length of the
treatment zone on each of the 3 catheters.
Normal saline is infused through the central
lumen continuously during the procedure
to dissipate heat production. The drive unit
automatically adjusts power according to
changing vessel conditions, reducing power
as flow is restored. The procedure is contin-
ued until complete lysis is achieved (Figure
9.1).
Early evaluation of the EKOS Endo-
Wave system in 53 patients, including 32
A
patients with lower extremity DVT, demon-
strated greater than 90% clot lysis in 70% of
patients and at least partial thrombus reso-
lution in 91%. Importantly, at least partial
lysis was achieved in 96% of acute DVT
(< 14 days), 100% of subacute DVT (15 to
28 days), and 77.8% of chronic (> 28 days)
and acute on chronic DVT. Median lytic
infusion time was 22 hours, and bleeding
complications were low (3.8%). Further-
more, median infusion times and median
total drug dosages administered were lower
with ultrasound-mediated thrombolysis
compared to standard CDT when using UK,
tPA, or rtPA. The median dosages and infu- B
sion times were similar for CDT and US-me-

diated thrombolysis when tenecteplase was Figure 9.1 The EKOS EndoWave and EkoSonic
used.18 The EKOS EkoSonic Endovascular systems. The microstreams alter fibrin composition even
System with MACH4e is similar to the ear- behind the valves, as seen in B. Image Courtesy of
lier version but this system allows for faster EKOS.
clot lysis.
removal while limiting thrombolytic dosages

and bleeding complications. PMT addition-
Percutaneous ally offers a treatment option for patients
with absolute contraindications for lytic
Mechanical therapy, as the AngioJet, a PMT device dis-
Thrombectomy cussed next, is the only device that can be

used without the addition of lytics. PMT has
Percutaneous mechanical thrombectomy further been shown to be more cost effective
(PMT) offers the benefit of early thrombus than alternative treatment regimens when

considering the lower thrombolytic dosages With the aspiration port clamped, in-
administered and decreased length of ICU fusate is released into the thrombosed ve-
stay compared to CDT,19 and the decreased nous segment during a slow pull-back of
long-term morbidity from postthrombotic the catheter, effectively lacing the clot with
syndrome compared to traditional antico- thrombolytic drug. After 10 minutes, the as-
agulation. piration function of the catheter is turned
on. The catheter is then advanced through
AngioJet Rheolytic the thrombosed segment a second time, re-
moving macerated thrombus through the
Thrombectomy System aspiration ports as the catheter is advanced.
Power-Pulse Spray Technique This process may be repeated if there is re-
The AngioJet catheter system is comprised maining thrombus burden at the end of the
of a single-use catheter, a single-use pump first pass. Alternatively, as is often our prefer-
set, and a drive unit. The catheter, which is ence, the patient may then undergo catheter-
available in working lengths of 60, 100, and directed thrombolysis in the intensive care
120 cm, contains a central lumen for in- unit (ICU) overnight and return to the oper-
fusate and a larger lumen encompassing the ative room the following day for reevaluation
central channel, the guidewire, and aspirate with venography and possible repeat throm-
from the thrombus. The drive unit generates bectomy or venous stenting, if indicated.
10,000 psi of pulsatile infusion flow, which Success in thrombus removal, restora-
is released from the catheter in retrograde- tion of venous patency, and preservation of
directed high-velocity saline jets. These jets valvular function have been demonstrated
create a localized low-pressure zone (Ber- with the use of the AngioJet power-pulse
noullis principle) at the catheter tip, mac- spray technique. While Kasirajan reported
erating thrombus and redirecting flow and only 24% of patients had >90% clot resolu-
debris into outflow channels directed behind tion, 35% had 50% to 90% resolution, and
the catheter tip for aspiration and removal 41% had less than 50% resolution,21 im-
(see Figure 8.1 on page 91). proved results have been demonstrated with
Access for the AngioJet system requires the addition of lytics to the infusate, as dis-
a 6F introducer sheath. The AngioJet cath- cussed previously. Bush et al. reported com-
eter is then advanced over a 0.035 guidewire plete thrombus resolution in 65% of patients,
through the thrombus load. While this sys- with at least partial resolution seen in all of
tem was originally intended for use without the remaining patients.22 Lin et al. demon-
adjunctive thrombolytics, it has been dem- strated that PMT with the AngioJet system
onstrated that the addition of lytics to the was at least as effective as CDT in treating
infusion solution results in decreased treat- lower extremity DVT. They showed com-
ment time and improved results. We recom- plete clot lysis in 75% of patients treated with
mend that thrombolytics be routinely used AngioJet vs. 70% in patients treated with
except when contraindicated, as is our prac- CDT (P = NS) with similar patency at 1-year
tice. While thrombolytic choice and dose follow-up of 64% and 68%, respectively. In
vary according to surgeon preference, we addition, they demonstrated reduced ICU
have experienced good results using 10 mg stay, reduced total in-hospital length of stay,
of tenecteplase in 50 mL of sodium chloride and reduced costs in the PMT cohort.19 In
infusing solution.20 our series, we demonstrated a 90% venous

patency restoration and maintenance of ve- tial venogram, with the goal of minimizing
nous valvular function in 88% at a mean treatment length of nonthrombosed vein.
follow-up of 6 months.20 The drive unit is attached to the sinusoidal
This therapy is associated with a low dispersion wire, which creates catheter os-
incidence of hemorrhagic complications. cillatation at 500 to 3500 rpm, causing dis-
Isolated case reports of pancreatitis resulting persion of lytics within the thrombus load
from massive hemolysis with use of the An- and mechanical clot disruption. Aspiration
gioJet system have been reported but appear of thrombus debris and lytic remaining in
to be rare occurences.23 the isolated segment completes treatment of
the isolated venous segment (Figure 9.2).
Trellis-8 Infusion System Access for the Trellis-8 infusion sys-
tem requires and 8F introducer sheath. A
Pharmacomechanical Thrombectomy 0.035 Glidewire (Terumo, Sommerset, NJ)
The Trellis-8 Periphperal Infusion System is used to cross the thrombosed venous seg-
incorporates the use of both chemical throm- ment and the Trellis-8 catheter is advanced
bolysis and mechanical thrombectomy. The over the Glidewire. With proximal and dis-
Trellis device consists of a single-use cath- tal balloons inflated, 5 to 10 mg of lytics are
eter, a dispersion wire, and an integral drive infused within the thrombus. After 10 min-
unit. The catheter contains proximal and utes, the dispersion wire is inserted into the
distal occlusion balloons that allow infu- catheter. Catheter vibration between the oc-
sion of thrombolytics to an isolated segment clusion balloons aids in clot maceration and
of thrombosed vein. Catheters are available increases the thrombus surface area exposed
in lengths of 80 or 120 cm, with varied dis- to the lytics. The dispersion wire may fur-
tances between occlusion balloons allowing ther be advanced and retracted once a min-
treatment of 10-, 15-, or 30-cm venous seg- ute during the treatment interval to further
ments. Selection of which catheter to use ensure mixing of the lytics with the throm-
will depend on the location and length of bus. After 5 to 15 minutes, the distal balloon
the thrombosed segment determined on ini- is deflated and the catheter aspirated via a
Figure 9.2 The Trellis-8 Peripheral Infusion System. Image courtesy of

Covidien.

side port to remove macerated thrombus rather as complementary procedures.26 The

and a substantial portion of the remaining initial device achieves significant clot bur-
lytics. The proximal balloon is left inflated den reduction, paving an easier path for the
during aspiration to prevent embolization second intervention. Use of a second PMT
of clot. After aspiration, with both balloons device can be performed in the same setting,
deflated, the system may be removed or ad- often with lower doses of thrombolytics.20,26
vanced into adjacent thrombosed segments, Alternatively, overnight CDT therapy may
repeating the procedure until thrombus load be sufficient to eliminate residual thrombus
is resolved. after debulking of clot with PMT.15,20,21 This
Hilleman et al. reported success with significantly reduces the time required for
the Trellis-8 infusion system for the treat- effective CDT and thereby reduces the as-
ment of proximal lower extremity DVT in sociated bleeding risks with this treatment
135 patients. They demonstrated superior modality.15,20,21 Patients should then undergo
clot lysis with Trellis-8 compared to con- a second evaluation with intravascular ultra-
ventional catheter-directed thrombolysis sonography and/or completion venography
with 93% achieving grade II (50%99% clot to evaluate vessel conditions. Underlying
resolution) or III lysis (100% clot resolution) venous stenosis may require venoplasty and
verus 79%, respectively. They additionally stenting (Figure 9.3).
demonstrated that patients receiving phar-
macomechanical lysis required lower lytic
dose, was more cost effective and associated Postprocedure
with significantly lower rates of hemorrhage
(0% vs. 8.5%, P <0.001).24 Arko et al. further
Anticoagulation
demonstrated 80% of patients experienced Patients should be anticoagulated after inter-
complete clot resolution with this technique ventions for DVT with unfractionated hepa-
in a single setting with venous patency main- rin or low-molecular-weight heparin and
tained in 88% of patients treated with this transitioned to oral warfarin for 6 months
device at a mean follow-up of 6 months.20 (goal international normalized ratio 23).
OSullivan demonstrated grade II or III lysis Patients with recurrent DVT or hypercoag-
in 96% of patients in a single setting, with ulable disorders may require a longer dura-
100% assisted primary patency at 30 days.25 tion of anticoagulation. Patients with venous
stents require lifelong aspirin therapy.
Adjunctive Procedures
Conclusion
Recalcitrant thrombus after initial treatment
with PMT may require further therapy. Percutaneous mechanical thrombectomy
While small residual thrombus may respond and ultrasound-accelerated thrombolysis are
to venoplasty and stenting, larger amounts at least as effective as catheter-directed throm-
of residual thrombus may require use of a bolysis with reduced ICU and hospital stays
second PMT device or overnight catheter- and decreased overall costs. Further, use of a
directed thrombolysis.15,20,21,26 Use of an ad- second PMT device, adjunctive CDT, and/or
junctive device or CDT should not be re- stenting of underlying venous stenoses may
garded as a failure of the first device, but be needed to achieve optimal clinical results.

Figure 9.3 A. Following an IVC filter placement the left iliac vein occlusion is treated with ultrasound-facilitated
thrombolysis in the prone position. B. PTA and stenting of May-Thurner syndrome results in an excellent venographic
result with long-term patency.
With widespread implementation of these arterio-venous fistula. Eur J Vasc Sur.

advanced treatment options for DVT, we can 1990;4:483-9.
achieve a significant reduction in long-term 6. Plate G, Eklf B, Norgren L, Ohlin P,
morbidity after proximal DVT. Dahlstrm JA. Venous thrombectomy for
iliofemoral vein thrombosis10-year results
of a prospective randomized study. Eur J Vasc
references Endovasc Surg. 1997;14:367-74.
1. Kearon C. Natural history of venous 7. Comerota AJ, Aldridge SA. Thrombolytic
thromboembolism. Circulation. 2003;107: therapy for acute deep vein thrombosis. Semin
I22-30. Vasc Surg. 1992;5(2):76-84.
2. Buller HR, Sohne M, Middledorp S. 8. Blum A, Roche E. Endovascular management
Treatment of venous thromboembolism. J of acute deep vein thrombosis. Am J Med.
Thromb Haemost. 2005;3:1554-60. 2005;118 (suppl):31S-36S.
3. Fifth ACCP consensus conference 9. Comerota AJ, Throm RC, Mathias SD,
on antithrombotic therapy. Chest. Haughton S, Meiwissen M. Catheter-directed
1998;114:439S-769S. thrombolysis for iliofemoral deep venous
4. Plate G, Einarsson E, Ohlin P, Jensen R, thrombosis improves health-related quality of
Qvarfordt P, Eklf B. Thrombectomy with life. J Vasc Surg. 2000;32:130-7.
temporary arteriovenous fistula: the treatment 10. Semba CP, Razavi MK, Kee ST, Sze DY, Dake
of choice in acute iliofemoral venous MD. Thrombolysis for lower extremity deep
thrombosis. J Vasc Surg. 1984;1:867-76. venous thrombosis. Tech Vasc Interv Radiol.
5. Plate G, Akesson H, Einarsson E, Ohlin 2004;7:68-78.
P, Eklf B. Long-term results of venous 11. Mewissen MW, Seabrook GR, Meissner
thrombectomy combined with a temporary MH, Cynamon J, Labropoulos N, Haughton

SH. Catheter-directed thrombolysis for lower 19. Lin PH, Zhou W, Dardick A, Mussa F,
extremity deep venous thrombosis: report of Kougias P, Hedayati N, et al. Catheter-directed
a national multicenter registry. Radiology. thrombolysis versus pharmacomechanical
1999;211:39-49. thrombectomy for treatment of symptomatic
12. Lieberman S, Safadi R, Aner H, Verstandig lower extremity deep venous thrombosis. Am J
A, Sasson T, Bloom AI. Local thrombolysis Surg. 2006;192:782-8.
for the treatment of patients with proximal 20. Arko F, Davis CM, Murphy EH, Smith ST,
deep vein thrombosis of the leg. Harefuah. Timaran CH, Modrall JG, et al. Aggressive
2002;141:424-9. percutaneous mechanical thrombosis: Early
13. AbuRahma AF, Pekins SE, Wulu JT, Ng clinical results. Arch Surg. 2007;142:513-8.
HK. Iliofemoral deep vein thrombosis: 21. Kasirajan K, Grey B, Ouriel K. Percutaneous
conventional therapy versus lysis and AngioJet thrombectomy in the management
percutaneous transluminal angioplasty and of extensive deep venous thrombosis. J Vasc
stenting. Ann Surg. 2001;233:752-60. Interv Radiol. 2001;12(2):179-85.
14. Ouriel K, Grey B, Clair DG, Olin J. 22. Bush RL, Lin PH, Bates JT, Mureebe L,
Complications associated with the use Zhou W, Lumsden AB. Pharmacomechanical
of urokinase and recombinant tissue thrombectomy for treatment of symptomatic
plasminogen activator for catheter directed lower extremity deep venous thrombosis:
peripheral aterial and venous thrombolysis. safety and feasibility study. J Vasc Surg.
J Vasc Interv Radiol. 2000;11:295-8. 2004;40:965-70.
15. McLafferty RB. Endovascular management of 23. Piercy KT, Ayerdi J, Geary RL, Hansen
deep venous thrombosis. Perspect Vasc Surg KJ, Edwards MS. Acute pancreatitis: a
Endovasc Ther. 2008;20:87-91. complication associated with rheolytic
16. Braaten JV, Goss RA, Francis CW. Ultrasound mechanical thrombectomy of deep venous
reversibly disaggregates fibrin fibers. Thromb thrombosis. J Vasc Surg. 2006;44(5):1110-3.
Haemost. 1997;78:1063-8. 24. Hilleman DE, Pharm D, Razavi MK. Clinical
17. Francis CW, Blinc A, Lee S, Cox C. and economic evaluation of the Trellis-8
Ultrasound accelerates transport of infusion catheter for deep vein thrombosis.
recombinant tissue plasminogen activator J Vasc Interv Radiol. 2008;19:377-83.
into clots. Ultrasound Med Biol. 1995;21: 25. OSullivan GJ, Lohan DG, Gough N,
419-24. Cronin CG, Kee ST. Pharmacomechanical
18. Parikh S, Motarjeme A, McNamara T, Raabe thrombectomy of acute deep vein thrombosis
R, Hagspiel K, Benenati JF, et al. Ultrasound- with the Trellis-8 isolated thrombolysis
accelerated thrombolysis for the treatment catheter. J Vasc Interv Radiol. 2007;715-24.
of deep vein thrombosis: initial clinical 26. McLafferty RB. Endovascular management of
experience. J Vasc Interv Radiol. 2008;19: deep venous thrombosis. Perspect Vasc Surg
521-8. Endovasc Ther. 2008;20:87-91.

CHAPTER
10
Surgical Thrombectomy
Bo eklf
T he options for early removal of an acute

thrombus in the proximal veins of the
leg are (1) catheter-directed thrombolysis
Rationale for Early
Thrombus Removal
(CDT), (2) percutaneous pharmacomech-
anical thrombectomy (PMT), and (3) sur- When deep venous thrombosis (DVT) oc-
gical thrombectomy (TE). In this chapter, curs, the goals of therapy are (1) to prevent
we propose that if CDT or PMT fails or is the extension or recurrence of the deep ve-
contraindicated, surgical TE is a valid alter- nous thrombus and fatal pulmonary embo-
native, primarily in acute iliofemoral vein lism (PE) and (2) to minimize the early and
thrombosis (IFVT). The techniques and relate sequelae of DVT. Antithrombotic thera-
sults of this procedure are presented. py can accomplish the former goal, but con-
tributes little to the second. Particularly in
proximal DVT (ie, IFVT), progressive swell-
ing of the leg can lead to phlegmasia cerulea
dolens (literally, painful blue swelling), and
to increased compartmental pressure, which
can progress to venous gangrene and limb
loss. Later, the development of severe post-
thrombotic syndrome (PTS) can result from
Davies md and Alan B. Lumsden md, eds. persistent obstruction of the venous outflow
Cardiotext Publishing, ISBN 978-1-935395-22-5. and/or loss of valvular competence; and pul-
103

monary embolism can lead to chronic pul- Pathophysiological

monary hypertension.
Most clinicians seem to focus on the Changes Associated
initial treatment (ie, preventing PE, the
propagation of the thrombus, or DVT re- with DVT
currence), with little attention to the limb The clinical outcome after DVT can be
sequelae, possibly swayed by the medical lit- categorized into 4 basic patient subgroups:
erature, which demonstrates, in randomized (1) those with neither detectable obstruc-
clinical trials (RCTs), that antithrombotic tion nor valvular incompetence, (2) those
therapy, particularly low molecular weight with obstruction alone, (3) those with val-
heparin (LMWH), can achieve these goals. vular incompetence alone, and (4) those
There is a lack of appreciation that the end- with both outflow obstruction and distal
points of these RCTs have not been preven- valvular incompetence. Noninvasive testing4
tion of postthrombotic sequelae. However, has shown that less than 20% of those with
2 RCTs have compared the late outcome documented DVT have neither significant
several years after acute, symptomatic proxi- obstruction nor reflux. In a 5-year follow-up
mal DVT in patients who wore compression of 20 patients treated with anticoagulation
stockings with those who did not.1,2 alone in a Swedish prospective, randomized
Both studies, one performed with study in acute IFVT5 there was residual iliac
custom-made thigh-length stockings and vein obstruction in 70% of patients and val-
one using calf-length ready-made stockings, vular incompetence and pathological mus-
showed that consequent wearing of stockings cle pump function in all patients, creating
could reduce the frequency of a postthrom- severe venous hypertension and contributing
botic syndrome to one-half. Partsch et al. per- to the severity of the postthrombotic sequel-
formed an RCT in 53 patients with proximal ae. However, the overall outcome of DVT,
DVT, comparing bed rest without compres- in terms of disturbed venous physiology, de-
sion with walking exercises using either com- pends to a great degree on the location of the
pression stockings or bandages, all undergoing thrombosed segments and the extent of in-
anticoagulation with LMWH.3 At follow-up volvement (ie, single or multiple segments).
after 2 years, a significantly better outcome The former relates to the likelihood of re-
could be found in the mobile group than in canalization of the thrombosed segment,
the bed rest group, as judged by the Prandoni which, in the lower extremity, decreases as
scale (median score 5.0 vs. 8.0, P <0.01). one moves proximally. Venographic stud-
Furthermore, there is apparently also a ies, which were only commonly performed
lack of appreciation of the different natural in the 1950s and 1960s, showed that close to
history of IFVT compared to more distal 95% of popliteal or tibial thromboses recana-
DVT, as well as a lack of understanding of lize completely,6 and at least 50% of femoral
the different pathophysiologic changes that venous thromboses recanalize.7
occur with time following DVT, and their In contrast, the minority of iliofemoral
relative contribution to the pathogenesis thromboses (less than 20%) completely re-
and severity of the postthrombotic sequelae. canalize and form a normal, unobstructed
These are reviewed in this chapter to provide lumen.8 However, while it is true that only
a selective basis for the decision to opt for 20% of the iliac vein segments involved in
early thrombus removal. IFVT completely recanalize spontaneously,

Surgical Thrombectomy 105
another 60% partly recanalize, or at least de- matosclerosis, and ulceration), but these too
velop adequate enough collaterals that they can be managed, in the compliant patient,
do not test positive for obstruction on non- with elastic stockings and elevation. Howev-
invasive study. Thus, in reality, only about er, those with both obstruction and valvular
20% remain significantly occluded enough incompetence have severe postthrombotic
or have such poor collaterals to produce sequelae, so severe, in fact, that conservative
symptomatic venous outflow obstruction de- management is difficult even in a compliant
tectable on noninvasive physiologic testing, patient.
if time (3 to 6 months) is allowed for com- The importance of the proximal ob-
plete resolution. Nevertheless, a significant struction for the development of distal valvu-
degree of obstruction persists for these 3 to 6 lar incompetence has been carefully studied
months in the majority of patients with ilio- by D. Eugene Strandness Jr. and his group.
femoral venous thrombosis. What is not well In a series of articles, they have shown the
appreciated is that persisting proximal ob- following: from 20% to 50% of initially un-
struction, even if it is ultimately relieved by involved distal veins become incompetent
partial recanalization or collateral develop- by 2 years; the combination of reflux and
ment, can lead to progressive breakdown of obstruction, as opposed to either alone, cor-
distal valves, resulting in reflux. This is im- related with the severity of symptoms and
portant because, if the distal venous valves was present in 55% of symptomatic patients;
(and particularly those in the popliteal vein) 25% of all venous segments developed reflux
remain competent, the postthrombotic se- in time, of which 32% were documented
quelae are relatively modest and controllable to not have been previously involved with
by conservative measures.9 The severity of thrombosis; and finally, in a study of the pos-
postthrombotic sequelae correlates with the terior tibial veins located below a popliteal
level of ambulatory venous pressure (AVP), segment involved with thrombosis, 55% of
as shown by Nicolaides et al.10 It is impres- the distal veins became incompetent if the
sive how steadily the frequency of ulceration segment remained obstructed, compared
climbs with increasing levels of AVP. It sug- to 7.5% of those below a popliteal vein that
gests that anything that significantly reduces recanalized.12 These changes in the distal
AVP will reduce the severity of the PTS, and veins occur early enough that they cannot
vice versa. Nicolaides and Sumner11 have be blamed on proximal valvular reflux, al-
also shown that the highest levels of AVP are though this also comes into play with time.
found in patients with both obstruction and Thus, it would appear that the early relief
reflux. of obstructing thrombus by thrombolysis
These 2 observations underscore a or thrombectomy should prevent more ex-
major point to be made in regard to the tensive postthrombotic sequelae if only by
disturbed pathophysiology associated with protecting the distal veins against progres-
iliofemoral venous thrombosis and the sive valvular incompetence. This is a point
pathogenesis of postthrombotic sequelae. that has not been well recognized not only
Obstruction alone is rarely sufficient enough in terms of basic pathogenesis but in judging
to cause venous claudication and mostly the results of thrombolysis and thrombecto-
causes increased swelling with activity. Val- my, where critics have largely ignored the res-
vular incompetence alone causes most of toration of proximal patency while focusing
the stasis sequelae (pigmentation, lipoder- on the presence or absence of valve reflux.

Another aspect of the importance of more likely to be present in proximal DVT

early thrombus removal is the new data ap- (ie, IFVT), but in bedridden patients there
pearing on the inflammatory response to may be no apparent swelling and the first
DVT. Thomas Wakefield and his group in signs may be those of PE (pleuritic chest
Ann Arbor have shown that the leukocyte pain, shortness of breath, and/or hemopty-
adhesion moleculeP-selectinactivates sis). Duplex scanning should interrogate not
the leukocytes emigrating into the venous only the femoral and popliteal veins, but the
wall, creating an inflammation that destroys iliac and calf veins as well; otherwise, 30%
the venous wall and the valves. John Har- of DVT will be missed.
ris and his group from Stanford13 showed, Venography is mainly used now when
in another experimental model, that if the catheter-directed interventions or surgical
thrombus was removed early, the inflamma- TE are indicated, to guide and monitor them.
tory changes were reversible. Standard ascending venography using foot
Finally, while modern venous recon- vein injection may miss isolated iliac vein
structive valvuloplasty can achieve good thrombosis unless an adequate contrast load
long-term results in primary venous disease is infused. In cases of IFVT, with extension
with severe reflux, the results of vein segment into the iliac vein without visualization of the
transfer and autologous vein transplantation upper end of the thrombus, a femoral veno-
in secondary (ie, postthrombotic) venous gram from the contralateral side is performed
disease are much less promising.14 There- to visualize the inferior vena cava (IVC) and
fore, there are less suitable late interventions determine the upper extent of thrombus ex-
for the venous derangements typically seen tension. An alternative to the conventional
in patients with severe PTS. The common venogram is a CT venogram, particularly if
mistake is for clinicians to treat all DVT pa- this is indicated for diagnosis of PE.
tients with anticoagulation only, and belat-
edly refer those who complain of severe pain
and swelling during follow-up for consider- Indications
ation of thrombolysis or thrombectomytoo
late for them to achieve their goals.
for Intervention
The conclusions from this and all the Early clot removal has clear benefit in 2
other studies cited are clear: early removal categories of patients, with IFVT falling at
of the thrombus conveys significant benefits, the 2 ends of the clinical spectrum: (1) in
and the earlier the removal, the better the active healthy patients with good longevity
outcome. in order to prevent or mitigate potentially
severe late postthrombotic sequelae and (2)
in those with massive swelling and phlegma-
Diagnosis sia cerulea dolens in order to mitigate early
morbidity and prevent progression to venous
Whether DVT is suspected on the basis of gangrene. Older patients with significant in-
pain, on the basis of discoloration or swell- tercurrent disease and serious comorbidities
ing of one leg, or in search of a source for who are unlikely to be active and live a long
pulmonary embolism, the diagnosis can be life, or those with distal thrombosis, should
confirmed with accuracy by duplex scanning be treated by anticoagulant therapy. Late
using color flow imaging. Clinical signs are PTS is not likely to be an issue with them.

However, even these patients, if faced with presented 6 patients, 5 of whom had an ex-
the threat of venous gangrene, may deserve cellent result with rapid disappearance of leg
prompt clot removal. swelling, very little late morbidity, and mini-
In terms of the choice of method of clot mum leg edema. There was no PE prior or
removal, CDT is an appropriate choice for subsequent to surgery. Mahorner claimed
removing obstructing thrombus and thereby that this method restores vein function with
preserving valve function, although the lat- preservation of the vein lumen and vein
ter has been presumed rather than proven. valves. In a follow-up paper in 1957,17 he re-
If CDT cannot be achieved, the clot reported 16 patients where TE was performed
moval or dissolution is unsuccessful or does in 14 legs and 2 arms with excellent results
not progress satisfactorily, or the concomi- in 12, good in 2, and poor in 2 patients. The
tant anticoagulation is contraindicated (eg, enthusiasm for TE created by Mahorner re-
IFVT in young women in the peripartum ceived strong support by the report by Haller
period, or in certain postoperative or trauma and Abrams in 1963.18 They presented 45
patients), then surgical TE or PMT is an ap- patients with IFVT who underwent TE. In
propriate choice. 34 patients with short history (<10 days), ex-
cellent bidirectional flow was established in
31 patients (91%). At follow-up after an aver-
age of 18 months, 26 out of these 31 patients
Historical Background of (84%) had normal legs, and where ascend-
ing venography was permitted in 13 patients,
Venous Thrombectomy normal patency of the deep venous system
in the United States was demonstrated in 11 (85%).

However, enthusiasm quickly subsided
The history of venous TE in the United after Lansing and Davis presented their
States is quite interesting and reveals mis- 5-year follow-up of Haller and Abramss pa-
conceptions that underscore its current in- tients in 1968.19 Of Haller and Abrams 34
frequent use. John Homans, an advocate for patients with short history, only 17 patients
division of the femoral vein to prevent PE, (50%) were interviewed, but 16 patients were
first suggested thrombectomy in a paper found to have swelling of the leg requiring
entitled Exploration and division of the stockings and one patient had developed
femoral and iliac veins in the treatment of an ulcer. Ascending venography in the su-
thrombophlebitis of the leg,15 presented at pine position was performed in 15 patients,
the New England Surgical Society in 1940. showing patent veins but the involved area
In this paper, Homans discussed indications of the deep venous system was found to be
for TE with or without ligation of the femo- incompetent in all cases and there were
ral vein, the technique, the complications, no functioning valves. Unfortunately, the
and the importance to prevent reflux. flaws in Lansing and Daviss study were not
However, the modern era of TE in the recognized and discussed widely enough:
United States started with Howard Mahorn- (1) they studied only half of the original co-
ers paper New management for thrombosis hort, and likely those with symptoms bring-
of deep veins of extremities in 1954,16 where ing them back to the vascular clinic; (2) the
he advocated TE followed by restoration of outstanding late patency of the thrombec-
vein lumen and regional heparinization. He tomized veins was completely ignored; and

(3) incompetence of the valves in the femo- pressure (PEEP) added during manipulation
ral and popliteal veins cannot reliably be as- of the thrombus to prevent perioperative PE.
sessed from an ascending venographic study The involved leg, contralateral groin, and
in the supine position. abdomen are prepared. A cell saver is used to
In a 1969 paper, Iliofemoral venous minimize blood loss. A longitudinal incision
thrombosis: reappraisal of thrombectomy,20 is made in the groin to expose the great sa-
William Edwards argued with Lansings phenous vein (GSV), which is followed to its
results and concluded that venous TE of- confluence with the common femoral vein
fers an effective and safe method of restor- (CFV), which is dissected up to the ingui-
ing flow in the deep venous system; when nal ligament. The superficial femoral artery
the thrombus is less than 10 days in dura- is cleared off 3 to 4 cm below the femoral
tion and is of the iliofemoral segment, TE bifurcation for construction of the arteriove-
is recommended; venograms at operation to nous fistula (AVF). In primary IFVT, where
determine the patency of the deep venous the thrombus originates in the iliac vein
system will aid in complete removal of the with subsequent distal progression of the
thrombus and give a basis for later compari- thrombus, a longitudinal venotomy is made
son and evaluation of long-term patency. In in the CFV, and a venous Fogarty TE cath-
the discussion, Lansing repeated his findings eter is passed upward through the thrombus
from the 5-year follow-up, still questioning into the IVC. The balloon is inflated and
the value of TE, but Haller, who was never withdrawn, these maneuvers being repeated
consulted about the follow-up report, stated until no more thrombotic material can be
that, at a recent visit to Louisville, he had extracted. With the balloon left inflated in
studied 17 patients in whom total removal the common iliac vein, a suction catheter is
of the thrombus had been possible and none introduced to the level of the internal iliac
had significant residual edema. Despite this vein to evacuate thrombi from this vein.
rebuttal, the impact of Lansings study, com- Backflow is not a reliable sign of throm-
bined with Karp and Wylies subsequent bus clearance since a proximal valve in the
1-page report21 in the Surgical Forum of 10 external iliac vein may be present in 25% of
patients in whom 8 had reocclusion of the cases, preventing retrograde flow in a cleared
femoral vein before discharge (even though vein. On the other hand, backflow can be
all had phlegmasia cerulea dolens and ex- excellent from the internal iliac vein and its
tensive thrombosis) was profound: only a few tributaries despite a remaining occlusion of
series on TE were subsequently published the common iliac vein. Therefore, an intra-
from the United States, in spite of the fact operative completion venogram is mandato-
that they all showed very good clinical re- ry. An alternative is the use of an angioscope,
sults in >75% of patients. which enables removal of residual thrombus
material under direct vision, or intravascular
ultrasound (IVUS). In early cases, the distal
Surgical Thrombectomy thrombus is usually readily extruded through
the venotomy by manual massage of the leg
The first TE for IFVT was performed by distally, starting at the foot. The Fogarty
Lwen, in Germany in 1937.22 Surgery today venous catheter, with a soft flexible tip, can
is performed under general intubation an- sometimes be advanced in retrograde fash-
esthesia with 10-cm positive end-expiratory ion without significant trauma. The aim is to

remove all fresh thrombi from the leg. The If phlegmasia cerulea dolens is the in-
venotomy is closed with continuous suture dication, because of the threat of impend-
and an AVF created using the saphenous ing venous gangrene, we start the operation
vein, anastomosing it end-to-side to the su- with fasciotomy of the calf compartments
perficial femoral artery (for illustrations, see to release the pressure and improve the cir-
Elkf et al12). An intraoperative venogram is culation immediately. If there is extension
performed through a catheter inserted in a of the thrombus into the IVC, the cava is
branch of the AVF. After a satisfactory com- approached transperitoneally through a
pletion venogram the wound is closed in lay- subcostal incision. The IVC is exposed by
ers and a closed suction drain is placed in the deflecting the ascending colon and duo-
wound to evacuate blood and lymphatic fluid denum medially. Depending on the veno-
that may accumulate after the operation. graphic findings relative to the top of the
In IFVT secondary to ascending thrombus, the IVC is controlled, usually just
thrombosis from the calf, the thrombus in below the renal veins. The IVC is opened
the femoral vein may be old and adherent to and the thrombus is removed by massage,
the venous wall. In such cases, the chance of especially of the iliac venous system; then if
preserving valve function has already been the femoral segment is involved, the opera-
lost and the opportunity to restore patency tion is continued in the groin as described
significantly diminished. A femoral segment previously. As an alternative, a retrievable
without functioning valves will lead to distal caval filter can be introduced before the TE
valve dysfunction in time, much as will fail- to protect against fatal PE. Heparin is con-
ure to achieve proximal patency. However, tinued at least 5 days postoperatively, and
a patent iliac venous outflow plus a compe- warfarin is started the first postop day and
tent profunda collateral system will most of continued routinely for 6 months. The pa-
the time achieve normal venous function. tient is ambulated the day after the operation
Therefore, if iliac patency is established but wearing a compression stocking and is usu-
the thrombus in the femoral vein is too old ally discharged after a week, to return after 6
to remove, it is preferable to ligate the femo- weeks for closure of the fistula.
ral vein. If normal flow in the femoral vein The objectives of a temporary AVF are
cannot be reestablished, we recommend ex- to increase blood flow in the thrombecto-
tending the incision distally and exploring mized iliac segment to prevent immediate
the orifices of the deep femoral branches. rethrombosis, to allow time for healing of
These are isolated, and venous flow is re- the endothelium, and to promote develop-
stored with a small Fogarty catheter. The ment of collaterals in case of incomplete
femoral vein is then ligated distal to the clearance or immediate rethrombosis of
profunda branches. In a 13-year follow-up the iliac segment. Usually the AVF is per-
after femoral vein ligation in this setting, formed between the saphenous vein and
Masuda et al23 found excellent clinical and the superficial femoral artery. More distally
physiological results without PTS. Finally, if placed AVFs have not been functional, in
there is evidence of iliac vein compression our experience.
on the completion venogram, which can A new percutaneous technique for fis-
occur in about 50% of left-sided IFVT, we tula closure was developed by Endrys in Ku-
recommend intraoperative endovenous iliac wait.24 Through a puncture of the femoral
angioplasty and stenting. artery on the opposite, surgically untouched

side, a catheter is inserted and positioned at vative group, respectively.27,28 Femoropopli-

the fistula level. Prior to release of a coil, an teal valvular competence at 6 months was
arteriovenogram can be performed to evalu- 52% in the surgical group compared with
ate the patency of the iliac and caval veins, 26% in the conservatively treated group, as
which is of prognostic value. More than monitored by descending venography with
10% of patients have been shown to have re- Valsalvaa significant difference.26 After 5
maining significant stenosis of the iliac vein years, combining the results of all functional
despite initial successful surgery. A percuta- tests, 36% of the surgical patients had nor-
neous, transvenous angioplasty and stenting mal venous function compared with 11% of
can be performed under the protection of the conservatively treated group.27 These dif-
the AVF, which is closed 4 weeks later. ferences were not statistically significant due
to loss of patients. At 10 years, using duplex
scanning, popliteal reflux was found in 32%
in the surgical group compared with 67% in
Complications and the conservative group.28
Results of Surgical
New Developments
Thrombectomy
One reason that induced surgeons to aban-
to Improve TE
don thrombectomy in the 1960s was the high There has been a significant improvement
mortality associated with early thombecto- of the results of surgical TE with the un-
my. In our series of more than 200 patients, derstanding to immediately restore iliac
mortality was less than 1%. There was no vein outflow in patients with iliac vein ob-
case of fatal PE in the perioperative period. struction as the major cause of their DVT.
In an RCT from Sweden,25 perfusion lung Control of iliac vein outflow immediately
scans were positive on admission in 45% of after TE can be achieved by intraoperative
all patients, with additional defects seen after venogram, angioscopy, or IVUS. In remain-
1 and 4 weeks in the conservatively treated ing obstructions, intraoperative endovenous
group, in 11% and 12% respectively, and in angioplasty and stenting are recommended.
the thrombectomized group in 20% and 0%, Schwarzbach et al. report excellent results
respectively. Tables with results from the lit- in 18 of 20 patients who maintained patency
erature are available in Rutherfords sixth after 21 months of followup.29 To improve
edition of Vascular Surgery.12 In the RCT patency and preservation of the valves in
from Sweden where TE was combined with the deep veins of the leg Blttler et al. has
a temporary AVF, 13% of patients had early combined TE of the iliac vein with throm-
rethrombosis of the iliac vein.26 In the Swed- bolysis applied under ischemic conditions
ish RCT, iliac vein patency at 6 months was to the leg veins.30 None of the 33 patients
76% in the surgical group compared with experienced clinically apparent recurrence
35% in the conservative group, as demon- within the first year. Clinical signs of the
strated by venography.26 This significant dif- postthrombotic syndrome were absent in all
ference was upheld after 5 and 10 years, with but 1 patient.
77% and 83% patency in the surgical group, With improved outflow through the
respectively, vs. 30% and 41% in the conser- iliac system by angioplasty and stenting

and increased inflow from the improved 200633the following recommendations are
distal patency by regional thrombolysis, the suggested:
temporary AVF may not be necessary. An
excellent review of contemporary venous Catheter-directed thrombolysis should
thrombectomy is published by Comerota be considered for proximal DVT,
and Gale 2006.31 especially iliofemoral thrombosis in
active patients at low risk for bleeding,
where the risk of the postthrombotic
Evidence-Based syndrome is higher than for more
Recommendations distal DVT (grade B). Systemic
thrombolysis should be avoided
The ACCP recommendations from 2004 because it is less effective, and because
(Bller, Agnelli, Hull, Hyers, Prins, Ras- the longer duration of therapeutic
kob)32 for treatment of acute venous throm- infusion required increases the risk of
boembolism concerning early thrombus hemorrhagic complications.
removal are the following: Surgical venous thrombectomy
should be considered for patients with
In patients with DVT or PE the symptomatic iliofemoral DVT who are
routine use of systemic thrombolytic not candidates for catheter-directed
treatment is not recommended (grade thrombolysis (grade C).
1A). Data concerning the short- and
In selected DVT patients, such as long-term effects of catheter-based
those with massive iliofemoral DVT mechanical intervention on the vessel
at risk of limb gangrene secondary to wall, venous valve, and pulmonary
venous occlusion, IV thrombolysis is vasculature are lacking and are
suggested (grade 2C). required before its role can be clearly
In patients with DVT, we recommend defined. This technique needs further
against the routine use of catheter- short- and long-term evaluation and
directed thrombolysis (grade 1C). eventually randomized controlled
In DVT patients confining catheter- trials before any recommendations can
directed thrombolysis to selected be made.
patients, such as those requiring limb
salvage is suggested (grade 2C). The ACCP recommendations from 2008
In patients with DVT, we recommend (Kearon, Kahn, Agnelli, Goldhaber, Raskob,
against the routine use of venous Comerota)34 are more in favor of early throm-
thrombectomy (grade 1C). bus removal:
In selected patients, such as patients
with massive iliofemoral DVT at risk In selected patients with extensive
of limb gangrene secondary to venous acute proximal DVT (eg, iliofemoral
occlusion, venous thrombectomy is DVT, symptoms for <14 days, good
suggested (grade 2C). functional status, life expectancy
>1 year) who have a low risk of
In the international consensus statement bleeding, we suggest that CDT may
guidelines according to scientific evidence be used to reduce acute symptoms

and postthrombotic morbidity if syndrome remain the primary objectives

appropriate expertise and resources are in treating these patients. More and more
available. data now show that early removal of throm-
After successful CDT in patients with bus provides a better outcome for patients
acute DVT, we suggest correction compared to anticoagulation alone. Multi-
of underlying venous lesions using ple treatment modalities such as catheter-
balloon angioplasty and stents (grade directed thrombolysis and percutaneous me-
2C). chanical thrombectomy are now available in
We suggest pharmacomechanical addition to surgical thrombectomy. In proper
thrombolysis (eg, with inclusion hands, percutaneous techniques for throm-
of thrombus fragmentation and/ bus removal are becoming the first-line treat-
or aspiration) in preference to CDT ment with durable outcomes being reported.
alone to shorten treatment time if More clinical trials are needed to assist in
appropriate expertise and resources are determining optimal patient selection to
available (grade 2C). improve immediate success, limit complica-
We suggest that they should not be tions, and provide long-term freedom from
treated with PMT alone (grade 2C). disease. Figure 10.1 suggests a treatment al-
In selected patients with acute gorithm for acute iliofemoral thrombosis.
iliofemoral DVT (eg, symptoms for
<7 days, good functional status, and
life expectancy >1 year), we suggest Addendum
that operative venous thrombectomy Acute Venous Thrombosis: Thrombus Removal
may be used to reduce acute symptoms with Adjunctive Catheter-Directed Thromboly-
and postthrombotic morbidity if sis: The ATTRACT Trial This is the NIH-
appropriate expertise and resources are sponsored prospective randomized trial with
available (grade 2B). If such patients the primary objective to determine if the ini-
do not have a high risk of bleeding, we tial adjunctive use of pharmacomechanical
suggest that CDT is usually preferable catheter-directed thrombolysis in symptom-
to operative venous thrombectomy atic patients with acute proximal DVT re-
(grade 2C). duces the occurrence of the postthrombotic
In patients who undergo any of syndrome (PTS) over 24-months follow-up.
these interventions, we recommend Three methods of initial rtPA will be used:
the same intensity and duration of (1) Trellis-8 peripheral infusion system; (2)
anticoagulant therapy afterwards as AngioJet rheolytic thrombectomy system; or
for comparable patients who do not (3) catheter-directed rtPA infusion. The con-
undergo intervention (grade 1C). trol arm will be treated with conventional
anticoagulation and compression stockings.
The primary efficacy outcome is cumulative
Conclusion incidence of PTS within 24 months, using
the Villalta PTS scale. Six hundred ninety-
Acute iliofemoral deep venous thrombo- two patients will be included in 30 to 50
sis remains a severely debilitating problem. centers in the United States, and the study,
Prevention of extension of thrombus, pul- which started in October 2008, is scheduled
monary embolism, and postthrombotic to finish in March 2013.

Figure 10.1 Treatment algorithm for acute iliofemoral thrombosis. CDT = catheter-
directed thrombolysis; PMT = percutaneous mechanical thrombectomy.
references 5. kesson H, Brundin L, Dahlstrm JA, et

1. Brandjes DP, Bller HR, Heijboer H, al. Venous function assessed during a 5
Huisman MV, de Rijk M, Jagt H, et al. year period after acute iliofemoral venous
Incidence of the postthrombotic syndrome thrombosis treated with anticoagulation. Euro
and the effects of compression stockings in J Vasc Surg. 1990;4:43-48.
patients with proximal venous thrombosis. 6. Arenander E. Varicosity and ulceration of the
Lancet. 1997;349:759-62. lower limb: a clinical follow-up study of 247
2. Prandoni P, Lensing AW, Prins MH, Frulla patients examined phlebographically. Acta
M, Marchiori A, Bernardi E, et al. Below- Chir Scand. 1957;12:135-44.
knee elastic stockings to prevent the post- 7. Thomas ML, McAllister V. The radiological
thrombotic syndrome. Ann Intern Med. progression of deep venous thrombosis.
2004;141:249-56. Radiology. 1971;99:37-40.
3. Partsch H, Kaulich M, Mayer W. Immediate 8. Mavor GE, Galloway JMD. Iliofemoral venous
mobilisation in acute venous thrombosis thrombosis: pathological considerations and
reduces postthrombotic syndrome. Int surgical management. Br J Surg. 1969;56:45-59.
Angiology. 2004;23:206-12. 9. Shull KC, Nicolaides AN, Fernandes
4. Lindner DJ, Edwards JM, Phinney ES, Taylor Fernandes J, Miles C, Horner J, et al.
LM Jr, Porter JM. Long-term hemodynamic Significance of popliteal reflux in relation to
and clinical sequelae of lower extremity deep ambulatory venous pressure and ulceration.
vein thrombosis. J Vasc Surg. 1986;4:436-42. Arch Surg. 1979;114:1304-6.

10. Nicolaides AN, Hussein MK, Szendro G, 21. Karp RB, Wylie EJ. Recurrent thrombosis
Christopoulos D, Vasdekis S, Clarke H. The after iliofemoral venous thrombectomy. Surg
relation of venous ulceration with ambulatory Forum. 1966;17:147-9.
venous pressure measurements. J Vasc Surg. 22. Lwen A. Uber thrombectomie bei
1993;17:414-9. Venenthrombose und Arteriespasmus.
11. Nicolaides AN, Sumner DS, eds. Investigation Zentralbl Chir. 1937;64:961-8.
of Patients with Deep Vein Thrombosis and 23. Masuda EM, Kistner RL, Ferris EB. Long-
Chronic Venous Insufficiency. Los Angeles: term effects of superficial femoral vein
Med-Orion Publishing Company; 1991. ligation: thirteen-year follow-up. J Vasc Surg.
12. Eklf B, Rutherford RB. Surgical 1992;16:741-9.
thrombectomy for acute deep venous 24. Endrys J, Eklf B, Negln P, Zka I, Peregrin
thrombosis. In: Rutherford RB, ed. Vascular J. Percutaneous balloon occlusion of surgical
Surgery. 6th ed. Elsevier, Saunders; 2005. arteriovenous fistulae following venous
p2188-98. thrombectomy. Cardiovasc Inter Rad.
13. See-Tho K, Harris EJ Jr. Thrombosis with 1989;12:226-9.
outflow obstruction delays thrombolysis and 25. Plate G, Ohlin P, Eklf B. Pulmonary
results in chronic wall thickening of rat veins. embolism in acute iliofemoral venous
J Vasc Surg. 1998;28:115-22. thrombosis. Br J Surg. 1985;72:912.
14. Kistner RL. Valve repair and segment 26. Plate G, Einarsson E, Ohlin P, Jensen R,
transposition in primary valvular insufficiency. Qvarfordt P, Eklf B. Thrombectomy with
In: Bergan JJ, Yao JST, eds. Venous Disorders. temporary arteriovenous fistula: the treatment
Philadelphia: WB Saunders; 1991. p261-72. of choice in acute iliofemoral venous
15. Homans J. Exploration and division of the thrombosis. J Vasc Surg. 1984;1:867-76.
femoral and iliac veins in the treatment 27. Plate G, Akesson H, Einarsson E, Ohlin
of thrombophlebitits of the leg. JAMA. P, Eklf B. Long-term results of venous
1941;224:179-86. thrombectomy combined with a temporary
16. Mahorner H. New management for arterio-venous fistula. Eur J Vasc Surg.
thrombosis of deep veins of extremities. Am 1990;4:483-9.
Surg. 1954;20:487-98. 28. Plate G, Eklf B, Norgren L, Ohlin P,
17. Mahorner H, Castleberry JW, Coleman WO. Dahlstrm JA. Venous thrombectomy for
Attempts to restore function in major veins iliofemoral vein thrombosis: 10-year results
which are the site of massive thrombosis. Ann of a prospective randomized study. Eur J
Surg. 1957;146:510-22. Endovasc Surg. 1997;14:367-74.
18. Haller JAJ, Abrams BL. Use of thrombectomy 29. Schwarzbach MH, Schumacher H, Bckler
in the treatment of acute iliofemoral venous D, Frstenberger S, Thomas F, Seelos R,
thrombosis in forty-five patients. Ann Surg. et al. Surgical thrombectomy followed by
1963;158:561-9. intraoperative endovascular reconstruction for
19. Lansing AM, Davis WM. Five-year follow-up symptomatic iliofemoral venous thrombosis.
study of iliofemoral venous thrombectomy. Eur J Vasc Endovasc Surg. 2005;29:58-66.
Ann Surg. 1968;168:620-8. 30. Blttler W, Heller G, Largiadr J, Savolainen
20. Edwards WH, Sawyers JL, Foster JH. H, Gloor B, Schmidli J. Combined regional
Iliofemoral venous thrombosis: reappraisal thrombolysis and surgical thrombectomy for
of thrombectomy. Ann Surg. 1970;171: treatment of iliofemoral vein thrombosis. J
961-70. Vasc Surg. 2004;40:620-5.

31. Comerota AJ, Gale SS. Technique of 33. Nicolaides AN, Fareed J, Kakkar AK, Breddin
contemporary iliofemoral and infrainguinal HK, Goldhaber SZ, Hull R, et al. Prevention
venous thrombectomy. J Vasc Surg. and treatment of venous thromboembolism:
2006;43:185-91. international consensus statement. (guidelines
32. Bller HR, Agnelli G, Hull RD, Hyers TM, according to scientific evidence). Int Angiol.
Prins MH, Raskob GE. Antithrombotic 2006;25:101-61.
therapy for venous thromboembolic 34. Kearon C, Kahn SR, Agnelli G, Goldhaber
disease: the seventh ACCP conference on S, Raskob GE, Comerota AJ. Antithrombotic
antithrombotic and thrombolytic therapy. therapy for venous thromboembolic disease.
Chest. 2004;126:401S-28S. Chest. 2008;133:454S-545S.

CHAPTER
11
IVC Filters
Joseph P. Hart and Claudio J. Schnholz
T he inferior vena cava (IVC) filter is

implanted with the primary intent to
prevent fatal pulmonary embolism. The
the death of a trauma patient with multiple
orthopedic injuries that could not be sal-
vaged from fatal pulmonary embolism. They
most basic indications for placement of IVC borrowed from technology used in oil pipe-
filters are: contraindication to anticoagula- lines in which retained sludge and debris are
tion, a failure of anticoagulation to prevent trapped by a cone-shaped filter. Such a filter
pulmonary embolism (PE), a deep venous allows oil to continue to flow around its pe-
thrombosis (DVT) occurring while on anti- ripheral edges while allowing debris to slowly
coagulation, or an anticoagulated patient for erode from the midportion of the filter. Like-
DVT or PE developing a complication from wise, Greenfields intent was to create a filter
anticoagulation, requiring cessation of ther- that would trap thromboembolic debris in
apy. Prophylactic indications are evolving its central portion and allow continued ve-
but controversial and are somewhat variable nous flow around it. This system permits the
from author to author and from institution to resolution of the clot over time, prevention
institution. Development of the vena cava fil- of pulmonary embolism, and eventual clear-
ter is attributed to a vascular surgeon, Lazar ance of the clot from the filter in most cases,
Greenfield, and a petroleum engineer, Gar- so that another episode could be resolved in
man Kimmell. Greenfield was motivated by a similar fashion, all while maintaining caval
patency.
The device and the procedure have be-
come increasingly common due to the gener-
Davies md and Alan B. Lumsden md, eds. al escalation of enthusiasm for endovascular
Cardiotext Publishing, ISBN 978-1-935395-22-5. procedures, relatively straightforward nature
117

of placement, increased patient and practi- treatment of DVT and PE, and in this chap-
tioner awareness due to educational efforts, ter, we discuss the indications, techniques,
and medicolegal concerns about thrombo- and outcomes for placement of IVC filters.
embolism. However, relatively strict adher-
ence to conservative indications for filter
placement is advised. Evolving indications IVC Filter Indications
such as PE prophylaxis in trauma or bariatric
surgery patients are of great interest since ve- Only one trial, by Decousus and colleagues,
nous thromboembolic disease is a significant published in the New England Journal of
complication in the management of these Medicine in 1998, has been performed that
patients. However, prospective randomized established the role of inferior vena cava fil-
data in these subgroups are, in general, lack- ter use.1-3 This trial demonstrated that IVC
ing, and while interest exists in studying filters are effective in preventing pulmonary
these indications, they are not without con- embolism in patients with proximal deep
troversy within the pulmonary, hematologic, venous thrombosis. Four hundred patients
general medical, surgical, and endovascular with proximal DVT were randomly assigned
communities. Retrievable filters have further to placement of IVC filter or no placement of
lowered our threshold to utilize IVC filters, IVC filter, with 200 patients in each group.
especially in these subgroups where the indi- They found significant differences in inci-
cations are less clear. Retrievability does not dence of all PE between the 2 groups at day
equate with actual retrieval. It is thought that 12. In the filter group, 2 patients, or 1.1%,
since they can be retrieved and are tempo- had either a symptomatic or asymptomatic
rary, most complications due to the filter PE, whereas 9 patients in the no-filter group,
can be reduced or eliminated with the use of or 4.8%, had a symptomatic or asymptom-
retrievable filters. atic PE. Four types of permanent IVC filters
The retrievable filters have unique lia- were placed, and the authors concluded that
bilities and these devices do not yet have in high-risk patients with proximal DVT,
long-term follow up data. With the recogni- there was a significant early beneficial effect
tion of the fact that patients fail to follow up of vena cava filter placement for the preven-
and undergo retrieval, and the technical in- tion of PE. However, this was in part offset
ability to retrieve the filter either due to place- by an increased rate for recurrent DVT in
ment exceeding the indicated dwell time or the filter group and an absence of any dif-
to simply technical inability to remove the ference in mortality between the groups. To
filter, a relative minority of filters are ever ac- this day, this remains the only prospective
tually retrieved. Thus, although retrievabil- randomized controlled trial regarding the
ity lends a new and exciting dimension to use of IVC filters to prevent PE.
IVC filter placement and creates the oppor- Despite this, the use of IVC filters has
tunity to do so in a temporary fashion, these increased dramatically, especially in the
devices will require further study to define United States. This increase is due partially
their exact long-term role in the treatment of to greater awareness of venous thromboem-
patients with venous thromboembolism and/ bolic disease (VTE) in general. However,
or prevention of complications from venous it is clear that further work to evaluate the
thromboembolism. Previous chapters have rational role of IVC filter placement in the
discussed the pathophysiology and medical treatment of patients with VTE or of pa-

IVC Filters 119
tients in need of VTE prophylaxis is neces- discharge from the hospital within an average
sary. The clearest indications for IVC filter of 10 days. These authors concluded that the
placement are recurrent acute or chronic super-obese male patient is at a very high risk
VTE on adequate therapeutic anticoagula- of developing PE and that their relative risk
tion, contraindication to or complication significantly exceeds that of other Roux-en-Y
of anticoagulation in the presence of VTE, gastric bypass patients. It was also noteworthy
and an inability to achieve or maintain that they found that the PE risk lasted several
therapeutic anticoagulation in the presence weeks after discharge, therefore extending
of documented VTE. Relative indications PE prophylaxis for several weeks after surgery
include the following: chronic PE treated may be warranted. In the group that had a
with thromboendarterectomy; massive PE BMI >55 kg/m2, patients were assigned to ei-
treated with thrombolysis and/or thrombec- ther receive or not receive an IVC filter. In
tomy, filter placement for thrombolysis for the group with the BMI >55 kg/m2 that had
iliocaval DVT (controversial); large, free- no IVC filter placed, the PE rate was 28%
floating proximal or caval DVT; VTE with and the mortality was 11%. In the BMI >55
limited cardiopulmonary reserve; recurrent kg/m2 group who received filters, there was
PE with a filter in place; difficulty establish- a 0% incidence of PE and a 0% PE mortal-
ing therapeutic anticoagulation; and poor ity rate. Their conclusion was that there was
compliance with anticoagulation or high a significant reduction in the perioperative
risk for anticoagulation such as unsteady gait PE rate when the patient with a BMI >55 kg/
or frequent falls. Prophylactic indications for m2 had an IVC filter placed. IVC filters were
IVC filters are usually cited when primary used in addition to subcutaneous heparin and
prophylaxis is not feasible but PE prevention sequential compression devices (SCD). In
is deemed necessary. Examples of this are a the morbidly obese, technical issues for IVC
trauma patient with a high risk for VTE, a filter placement can arise related to imaging
patient with a high risk for VTE undergoing table weight restrictions, inadequate imag-
a surgical procedure, or a medical condition ing due to body mass, and x-ray penetration
in a patient with a high risk for VTE.4 issues. Alternative imaging techniques such
Surgery for morbid obesity is an area as intravascular ultrasonography may be use-
in which prophylactic IVC filters are often ful in this setting. Reports exist of the use of
entertained and placed. Recent data indi- carbon dioxide cavography in these patients.
cate that the fatal PE rate with morbid obe- However, published evidence documenting
sity surgery within 60 days may be as high this practice so far is limited.
as 0.85%. Risk factors that contribute to this IVC filter placement in the hands of
are underlying venous insufficiency, sleep an experienced practitioner is an extremely
apnea, and the patients baseline body mass safe and low-risk procedure; however, there
index (BMI).5 Abou-Nukta and colleagues are contraindications to placement that in-
compared age and BMI of morbidly obese clude the following: an IVC that is either
patients in a randomly selected Roux-en-Y occluded or completely filled with thrombus
gastric bypass patient control group. Patients with no safe region to deploy a filter below
were broken down into BMI >55 kg/m2 and the right atrium; active septicemia or bacte-
BMI <55 kg/m2. In super-obese men with remia, which may provide an opportunity to
BMI >55 kg/m2 the incidence of PE was 4%. colonize the filter with active infection; and
Nine of the 11 patients developed PE after a thrombus presenting between the proposed

access site and site of deployment of the fil- record for safety and durability than other,
ter. Sepsis or bacteremia is often considered better known familiar permanent filter de-
a relative contraindication only, and it is con- signs, can and often do at this point remain
troversial as to whether infection should in- as permanent filters. Consequently, addi-
terfere with IVC filter placement. tional studies to document their long-term
In patients who have a retrievable filter safety, durability, and functionality still need
in place, and if during the course of treat- to be done.
ment it is concluded that the filter will no Suprarenal IVC filter placements may
longer be required or is no longer effective, need to be considered in some cases. Situa-
there are several considerations for removal tions in which this might be considered are
of retrievable filters. If the patients baseline renal vein thrombus, pregnancy, IVC clots
low risk for PE has returned after resolu- either above the renal veins or above an IVC
tion of other conditions or if the patient filter, duplicated IVC, PE that has been doc-
can return to routine anticoagulation, con- umented to occur from a gonadal vein, or an
sideration should be given to filter removal. excessively short infrarenal IVC.6
Retrievable filters allow the option for re- Occasionally, superior vena cava (SVC)
moval if it has become ineffective from mi- filters are entertained. In our experience,
gration, excessive angulation, or fracture. If these are rarely, if ever, necessary, and even
a continued need for the filter exists, these proponents of this treatment acknowledge
filters could be replaced in a more optimal the controversy surrounding it.7 However,
position or location. Active bacteremia or some literature does exist that documents
septicemia is cited by some as a reason not their use in certain cases and the interven-
to place a filter; however, if a patient is ac- tionalist or surgeon should be aware of these
tively bacteremic at the time filter placement reports, if only to address the concerns of
becomes necessary, a retrievable filter may consulting physicians who request their
be a good choice in such a patient. Another placement. Upper extremity DVT is in the
contraindication to filter retrieval is a filter vast minority, comprising less than 10% of
that is largely or fully laden with clot that all DVTs. However, upper extremity and
cannot be remedied by endovascular means. major central thoracic vein thromboses are
This filter cannot be removed since it pres- becoming more common with the increased
ents an undue risk of PE during the retrieval use of vascular access devices. Although its
procedure. For this and various other rea- incidence is uncommon, upper limb DVT
sons, many retrievable filters that are initially is more likely to lead to PE, but these emboli
placed with the intent of future removal will are smaller and less likely to be symptomatic
become permanent. An additional contra- or fatal. Considerable controversy contin-
indication to filter removal is persistence of ues to surround these questions with regard
clot in the pelvic or lower extremity veins to management of upper extremity DVT.
unless the patient becomes a candidate for Placement of a vena cava filter in the reverse
anticoagulation therapy. Until the patient orientation in the SVC is conceptually a po-
with the retrievable filter has their risk for PE tential method of managing these thrombo-
returned to normal baseline, or if the patient ses in patients who have a contraindication
becomes a candidate for anticoagulation, fil- to anticoagulation or have higher fall risk,
ters likely will not be removed. Retrievable which is not uncommon in these popula-
filters, although they have a shorter track tions. Filter placement in this location and

IVC Filters 121
in reverse orientation is technically chal- from the left iliac vein excludes a duplicated
lenging. There is a very short landing zone cava. The width of the vena cava should be
for placement of these filters in the SVC documented. After documenting caval anat-
below the confluence of the major thoracic omy and excluding caval thrombosis, du-
veins and above the right atrium. Any complication, megacava, renal vein anomalies,
plications, such as inadvertent deployment and/or the presence of a left-sided vena cava,
even slightly beyond the intended landing filter placement can proceed. While the in-
area for the filter, could result in emboliza- dividual types of IVC anomalies and condi-
tion to the right atrium, and any perforations tions are rare, the incidence of an anatomic
that occur would likely be more serious. A variation of the IVC, taken all together, ap-
femoral insertion kit is used if the filter is to proaches 15%. The filter is typically placed
be placed from the jugular position, and a with its apex at or just below the level of the
jugular insertion kit is used if the filter is to lowest renal vein. Some manufacturer in-
be placed from the femoral approach. structions for use (IFU) indicate placement
of the upper portion of the cone of the filter
within the outwash of the renal veins to aid
IVC Filter Techniques the resolution of thrombi captured by the fil-
ter. Placement above the renal veins should
IVC filter implantation is a relatively straight- be reserved for the indications stated earlier.
forward procedure that ought to be well Deployment protocols for specific filters are
studied and mastered by the vascular inter- per the IFU for the individual filters. Table
ventionalist or endovascular surgeon. Patient 11.1 contains filter device details, and fluoro-
preparation, intraprocedural imaging, and scopic images of several types appear in Fig-
meticulous technique are required. Most ure 11.1. The practitioner should be familiar
implants are performed either via the com- with the manufacturers IFU of the various
mon femoral or the jugular vein approaches. filters available at their institution that they
Additional approaches can be via the antecu- routinely use. Successful deployment is con-
bital or the subclavian veins, as well as other sidered routine once the above conditions
veins in unique situations. Extremely careful have been assessed and should occur 95%
technique should be utilized when obtain- to 99% of the time in appropriately selected
ing percutaneous access since many patients patients.
are either on anticoagulation, have had one In the event that any of the abovemen-
venous thrombosis, or have had a bleeding tioned anomalies of the vena cava are pres-
complication. The location and side of any ent, significant alterations of filter placement
DVT should be determined before access is are required. In duplicated cava, a filter in
obtained. Ultrasound-guided access to the each cava should be considered. Alterna-
vein is recommended. An initial cavogram tively, a filter in the suprarenal IVC is an
is obtained using a pigtail catheter, another option. If megacava is encountered, options
multiside hole catheter, or the kits delivery include the use of a Birds Nest Filter (Cook)
sheath, which is optimally placed in the or placing 2 filters with 1 filter in each com-
confluence of the iliac veins either from mon iliac vein. With a left-sided vena cava,
the femoral or jugular vein approach; or al- the filter can be placed in the solitary left-
ternatively, with a catheter tip placed in the sided vena cava. Careful documentation of
left iliac vein since opacification of the cava the presence of the aorta to the right of the

Table 11.1 AvailableIVCFilterDeviceDetails 122
SheathOuter
Retrievable Diameter(Fr) MaximumIVC Jugular
Filter Manufacturer (Route) Material [Jugular] Diameter(mm) Placement
LD2 VTD Nearly Final Pages.indd 122

Greenfield (24 Fr) Boston Scientific No Stainless-steel wire 28.0 28 Yes
Titanium Greenfield Boston Scientific No Titanium wire 14.3 28 Yes
Greenfield (12 Fr) Boston Scientific No Stainless-steel wire 14.0 28 Yes
Birds Nest Cook No Stainless-steel wire 13.8 40 Yes
Gnther Tulip Cook Yes (Jugular) Elgiloy wire 10.0 30 Yes
Celect Cook Yes (Jugular) Conichrome 7.0 [8.5] 30 Yes
Vena Tech LP B. Braun No Phynox wire 9.0 35 Yes
Simon Nitinol CR Bard No Nitinol wire 9.0 28 Yes
G2 X CR Bard Yes (Jugular) Nitinol wire 7.0 [10.0] 28 Yes
Trap Ease Cordis Endovascular No Nitinol hypotube 8.5 30 Yes
Opt Ease Cordis Endovascular Yes (4 weeks) Nitinol hypotube 8.5 30 Yes
(Femoral)
Option Angiotech Yes (Jugular) Nitinol 6.0 30 Yes
5/13/11 10:27 AM
IVC Filters 123
Figure 11.1 Fluoroscopic imaging of 6 common IVC filter designs in use today. 1. Greenfield Titanium (Boston
Scientific). 2. Trap Ease (Cordis). 3. VenaTech LP (VenaTech). 4. Gnther Tulip (Cook). 5. Opt Ease (Cordis). 6. G2 X
(CR Bard). The first 3 are examples of permanent filter designs. The latter 3 are retrievable and have hooks at the top
or bottom (for either transjugular or transfemoral access for retrieval, respectively) via dedicated retrieval systems or
interventional sheath/loop snare combinations.
left-sided vena cava is needed before deploy- injection of contrast at 20 mL/s for a total of
ment of the filter. For renal vein anomalies, 40 mL. Another less established technique is
such as a circumaortic left renal vein or the to document the major vena caval branches
presence of multiple renal veins on one side, with selective catheterization without con-
the filter should be placed below the level of trast administration.
the lowest renal vein. Many practitioners have chosen to uti-
A marker pigtail catheter or sizing pig- lize retrievable filters on a fairly regular basis
tail specifically designed to measure the size given their apparent reliability as long-term
of the vena cava can be used if it was not filters from early experience.8,9 Retrieval pro-
clearly documented prior to the procedure. cedures are reviewed in Figures 11.2 and 11.3.
Typically, a cavogram is performed with an The Bard recovery filter has been updated,

Figure 11.2 Recovery sequence of G2 X (CR Bard) IVC filter using a recovery cone (Bard): 1. Initial cavagram
and positioning of recovery cone at apex of filter. The filter is seen to be free of embolic debris or thrombus. 2. After
positioning of the cone over the filter, the sheath is advanced to close the cone. 3. Retraction of the filter into the cone.
4. After withdrawal of the
filter and sheath removal,
a final cavogram
demonstrates complete
removal of the filter.
Figure 11.3 Recovery sequence of Gnther Tulip (Cook) IVC filter using a loop snare and sheath: 1. Initial
cavagram is performed. The filter is free of embolic debris or thrombus. 2. Positioning of snare at apex of filter. 3.
After positioning of the snare over the filter and capture of the hook, the sheath is advanced to close the filter and
begin recovery. 4. Retraction of the filter into the sheath by the snare. 5. After withdrawal of the filter, final cavagram
is performed to demonstrate absence of spasm, thrombus, or extravasation. 6. Recovered filter with mix of chronic
fibrinous material at the apex and acute thrombotic material that is typically seen at retrieval and probably forms while
filter is in sheath with static blood.

IVC Filters 125
which is now the Bard G2 X, and it can be re- These retrieval procedures are performed
covered with either a recovery cone or with a with the patient under anticoagulation in
snare from a jugular approach. Likewise, the many cases.10 The patient is ideally accessed
Gnther Tulip and the Cook Celect filters under sedation via the right internal jugu-
can be recovered with a snare from a jugular vein with ultrasound guidance starting
lar approach. The Cordis Opt Ease is recov- with a micropuncture technique and then
ered from a femoral approach also using a upsizing to a short, interventional sheath.
snare. Anecdotal evidence indicates that the Using standard techniques, the inferior vena
Cordis Opt Ease has a shorter recovery win- cava is accessed, and a catheter or sheath is
dow (Figure 11.4). For this reason, we would placed at or below the level of the filter. A
typically prefer any of the other retrievable cavogram is obtained, and documentation
filters, as there is accumulating evidence of patency of the filter or determination of
that they can often be safely retrieved well clot burden is performed. In cases of small
beyond the manufacturers IFU. This may clots within the filter cone (25% or less cone
extend the opportunity for filter retrieval in volume), retrieval may be performed safely.
a number of patients whose risk for PE does However, when there is large clot burden
not return to baseline early, who are initially (>25%), filters are usually not removed, and
lost to follow-up, or whose ability to tolerate patients are typically anticoagulated and
anticoagulation does not return for a longer brought back at a later time for interval reas-
period of time. However, this needs to be sessment for possible later retrieval. If a pat-
done on an individualized basis, with great ent filter is documented, a recovery cone in
care and attention to signs that retrievability the case of the Bard G2, a loop snare in the
may not be possible at the time of removal. case of the Bard G2 X, or a loop snare or
Figure 11.4 Opt Ease (Cordis Endovascular) IVC filter retrieval attempted with resulting caval stenosis and inability
to remove filter. The patient presented late for retrieval but strongly wished to undergo attempted retrieval despite
counseling about decreased potential for successful filter removal and increased risk of stenosis or injury. A. Caval
stenosis and retained IVC filter. The patient was maintained on warfarin, in accordance with his history of recent PE
and DVT. B. Three-week CT venogram confirming resulting stenosis (rather than spasm) and continued filter patency.
C. Four-month CT venogram again documenting continued filter patency and some favorable remodeling of the IVC. It
was planned to continue warfarin in this patient (who suffered DVT and PE perioperatively after oncologic surgery) for
at least a full 12 months and follow the IVC morphology on subsequent CT scans during oncologic followup.

the respective recovery systems for the Cook filter placement under fluoroscopy alone.
Celect or Tulip filters, is brought into place. Follow-up IVUS imaging can also provide
A snare is brought around the hook and the an additional confirmation of correct filter
sheath is advanced over the recovery cone or deployment and the relative position to renal
the snare to collapse the filter within the de- veins following deployment. This combined
vice. The filter is then withdrawn carefully IVUS and fluoroscopic approach can be a
under fluoroscopy to document that it is not single-access site technique. Ashley and col-
dislodged from the sheath. The catheter or leagues have argued that IVUS more accu-
sheath is then reinserted into the abdominal rately localizes the renal veins and measures
vena cava and a cavogram is obtained to rule the diameter of the vena cava than contrast
out perforation, stenosis, or thrombus forma- venography.15 Transabdominal ultrasonog-
tion in the cava following the procedure. raphy can fail to identify proper anatomical
Careful attention to both meticulous access landmarks in up to 10% of patients, but in
technique and hemostasis of the internal centers that have made a committed effort to
jugular vein, particularly if the patient is an- this approach, transabdominal ultrasonogra-
ticoagulated, is of course critical. phy can be utilized to document anatomy
Rosenthal and others have identified and adequate filter positioning.16 If technical
techniques to place inferior vena cava filters support and expertise exist to obtain quality
using intravascular ultrasonography (IVUS) transabdominal venous images, and body
at the bedside of patients in the ICU, pro- habitus, bowel gas, or ileus does not preclude
viding a way to insert an IVC filter in pa- adequate visualization, this is a valuable al-
tients who cannot be easily transported. The ternative technique.
IVUS bedside ICU technique has been ad-
opted at other institutions as well.11-14 Typi-
cally, this requires a 2-vein accessvia both IVC Filter Outcomes
the right and left common femoral veins
and specialized standing protocols to bring Complications of IVC filter placement along
IVUS and filter deployment equipment to with their incidence are important data to
the ICU.11 More recently, beyond the ini- incorporate into the decision to place, the
tial IVUS studies done by Rosenthal, other placement procedure, and follow-up of these
workers have documented a single-vein ac- devices. Migration and access site throm-
cess technique using markings and dimen- bosis are 2 frequent and potentially highly
sions obtained by using IVUS and using significant adverse events. Conservative esti-
this length data to guide deployment. An- mates place access site thrombosis at a rate
other use for IVUS is in a procedure suite of 0% to 6% of placements. Other reports
with fluoroscopy in cases where avoidance cite access site thrombosis occurring at 2%
of contrast administration, such as in cases to 35%. Access site thrombosis appears to be
of contrast allergy or renal insufficiency, is greater with the femoral route. Ultrasound
required. Vascular landmarks are assessed guidance, use of careful micropuncture
using IVUShepatic veins, left renal artery, needle technique, and the smallest-diameter
renal veins, and iliac veins. Then, these land- delivery system appropriate to a given pa-
marks, fluoroscopy of the IVUS catheter in tient may limit such thromboses. Access site
the planned filter landing zone can identify thrombosis is a particularly worrisome out-
a bony landmark, which then serves to guide come when placement was for a purely pro-

IVC Filters 127
phylactic indication. Filter migration is also into the cardiac chambers, although rare, is
a worrisome problem with an uncertain in- a dreaded complication of filter placement.
cidence. Estimates of migration occurrence Injury to adjacent retroperitoneal structures,
vary from 0%18% to 3%69%. Migration which fortunately is extremely rare, is also
may be limited to some extent by careful compelling evidence of a perforation relat-
IVC sizing and meticulous filter deployment ed to filter placement. Occasionally, a filter
procedure in accordance with each devices strut has been seen to migrate through the
instructions for use. Reports aiming to quan- caval wall and into adjacent vessels, verte-
tify both access site thrombosis and filter brae, or viscera. In the absence of bleeding
migration are heavily influenced by follow- or gastrointestinal complications, this prob-
up protocol and imaging modality as well ably can best be managed by serial moni-
as whether or not asymptomatic patients are toring. Gastrointestinal complications of
screened for these adverse outcomes. filter placement may include bleeding or
Numerous other complications occur perforation of the adjacent viscera, particu-
related to these devices: death may occur, larly the duodenum. These rare cases may
but well under 1% of cases; recurrent clinical present either as abnormalities on CT scan-
PE may occur in 2% to 5%; filter emboliza- ning for related or other reasons or upper gas-
tion may occur in 2% to 5%; IVC occlusion trointestinal bleeding when visceral ulcers
may occur in 2% to 30%; depending on the are created by local strut trauma. Central
series examined, IVC penetration can occur venous guidewire manipulation or central
in 0% to 41%; and filter fracture can occur in line insertions have also contributed to the
2% to 10% (Table 11.2). IVC penetration is a dislodgement of previously placed IVC fil-
controversial issue. A computed tomography ters. Guidewires may become entangled
(CT) scan may poorly identify a collapsed in the vena cava filter during other venous
vena cava around the limbs of an IVC fil- interventional procedures with a J or other
ter and present the image of caval penetra- wire advanced well into the vena cava and/or
tion, when in fact, this is an underfilled cava during central line placement. It is possible
collapsed around an IVC filter. It may be that many of these may be unrecognized
more reasonable to be concerned with caval by the operator and manifest only as mild
penetration when clear evidence of this ex- to moderate difficulty removing the wire at
ists and/or a complication of such penetra- some point during the procedure. If such a
tion occurs, such as a visceral perforation or wire becomes impossible to remove or filter
bleeding complication. Filter misplacement entanglement is suspected, intervention to
or migration may occur despite meticulous remove it using catheter techniques may be
technique and careful deployment. Rapid necessary. Such conditions are likely under-
and precise deployment of the filter reduces recognized and probably underreported. PE
the incidence of strut asymmetry, incom- or recurrent PE despite the presence of a
plete opening (Figure 11.5), and tilting. Mis- filter can occur. Anticoagulation should be
placement of the filter at insertion may occur maintained whenever possible to minimize
through a number of mechanisms, and these recurrent PE. Renal complications include
filters are at an increased risk for migration. renal vein thrombosis related to suprarenal
Migration of either fractured components or filter placement. Rarely, obstructive uropa-
the entire device may occur. Central migra- thy due to a pelvic obstruction of the ureter
tion of a filter into the right atrium or further related to vena cava wall strut perforation

Table 11.2 IVCFilterPlacementComplicationsfromSelectedAuthors
AccessSite
Major Postfilter Thrombosis
Procedural RecurrentPE (Symptomatic) CavalThrombosis
Becker et al.,17 1992 0.0%6.8% 2% 0.0%15.1%
Mohan et al.,18 1995 0.0%6.3%b 2.0%4.4% 0.0%14.6%
Athanasoulis et al.,19 2000 <1% 5% 2% 5%
Sharafuddin et al.,20 2001 1.9%3.4% 2% 0%19%
Patel,21 2004 2%5% 0%6% 2%30%
Hann et al.,22 2005 0%9% 0.0%13.1% 0.0%9.5%
Hoppe et al., 23
2007 5% 2% 5%
Georgiades et al.,6 2008 1% g 2% 19%
Abbreviations: PE = pulmonary embolism; DVT = deep venous thrombosis.
Figure 11.5 Fluoroscopic (A) and CT (B) imaging of partial incomplete opening seen with a Greenfield IVC filter
(Boston Scientific). Asking the patient to cough or individual strut manipulation with an angled tipped catheter are
options to try to correct this during the initial deployment procedure, but strut tips are usually firmly placed in the IVC
wall immediately after release. This deployment configuration will afford some protection from large PE. However, if PE
occurs in this situation or more optimal filter protection from PE is needed, a second IVC filter below (if cava and filter
are patent) or a suprarenal IVC filter (if the filter itself is shown to be a possible source or caval thrombus is present)
may be considered.

IVC Filters 129
Table 11.2 (cont.) IVCFilterPlacementComplicationsfromSelectedAuthors
Filter 30day
Filter Migration Filter 30dayMortality Mortality Recurrent
Fracture (Major) Infection (duetofilter) (overall) DVT
0.0%59.3%a <<1%
0.0%10.9%c
<1% <1% <<1% <1% 17%
6%8%d (Rare)
2%10% 0%18%e <1%
0%32%
<1% f
<1% 20%
2% 6% h
a
All migration, includes single-device series. bFilter misplacement. cNot limited to 30 days. dSingle, early device data
cited here. eAll migration. fWith clinical consequences. gIatrogenic PE. hCombined migration and malposition.
may occur. This may manifest as either he- experience has been gained through multiple
maturia or hydronephrosis. Careful imaging models of the Greenfield filter. The Green-
and active questioning of position during field filter is widely regarded as a filter with
deployment will avoid other such highly an excellent long-term safety and efficacy re-
unusual mishaps such as placement in the cord. Greenfield and colleagues have main-
mesenteric vein via the portal system and/ tained large registries of filters that they have
or aortic placement, both of which have placed, and their 20-year experience with
been documented to occur in extremely rare these filters documented an incidence rate of
cases.24 Frank phlegmasia complicating pro- recurrent PE at 4% and a caval patency rate
phylactic filter placement is a real, though of 96%, and caval movement of little or no
uncommon, entity.25 Concern regarding this clinical significance was seen in 8% of their
is one of the factors driving increased consid- cases. They reported minimal procedural
eration for retrievable filters. Fortunately, in morbidity and mortality, and they concluded
patients who can undergo either mechanical in their analysis that insertion of the stainless
or pharmacomechanical thrombolysis with steel Greenfield vena cava filter provides pro-
the filter in place, phlegmasia from a filter tection from PE while maintaining patency
occlusion can be treated with a minimally of the IVC over long-term follow-up.
invasive approach to attempt to resolve the Reports about experience with retriev-
thrombus and then anticoagulation can be able filters that have been removed after ex-
resumed (Figure 11.6). tended dwell times past the manufacturers
The IVC filter with the longest clinical IFU is beginning to become available.27
experience is the Greenfield filter.26 Extensive When retrieval is attempted and the patient

Figure 11.6 A patient presenting with bilateral sudden onset limb edema and discomfort occurring late after a
filter placed for VTE immediately after spine surgery. There was no frank phlegmasia or compromised lower extremity
viability. A CT venogram was the initial study obtained (A)and was to some extent equivocal as is often the case due
to venous contrast opacification patterns on CT. Correlation with catheter venography did show that the filling defect
below the outflow of the right renal vein was in fact filter-associated thrombus (marked by arrow in A ). Initial catheter
venogram (B) confirms massive illiocaval thrombosis. Extensive pharmacomechanical debulking of clot was performed
with iliofemoral and lower IVC improvement (C) in an effort to avoid long-term complications related to lower extremity
venous insufficiency. Due to contrast volume and sedation time, it was elected to bring the patient back for a staged
second procedure to complete thrombus removal. He was maintained on anticoagulation and had symptomatic
improvement but required coronary angiography and intervention. Due to this and transient renal impairment, it was
elected not to reintervene on his cava and IVC filter. He was discharged on warfarin anticoagulation with continued
symptomatic improvement.
was not lost to follow-up or had secondarily patients are either lost to follow-up, the deci-
requested to keep the filter in place, techni- sion to leave the filter in place is made, or the
cal success approaches 85% or greater. Fac- patients themselves decline to undergo the
tors correlating with failure of filter retrieval removal procedure.
were due to device angulation, tilt, or cant- Filter placement and its use in trauma
ing within the IVC. Most clinical series of patients have been assessed by Quirke and
retrievable filters are reported to be free or colleagues by questionnaire.28 They que-
relatively free of symptomatic PE, bleeding, ried respondents that were largely at level I
thrombosis, or caval stenosis. These proce- trauma centers that have over 1000 trauma
dure times should be short when retrieval is admissions per year. These centers reported
straightforward. It is reasonable to conclude a low complication rate, with most of these
that when filter retrieval is attempted, it is centers reporting one or fewer complications
performed successfully in a large proportion per year despite their relatively high volume.
of cases, despite growing evidence that many Interesting trends were observed in these

IVC Filters 131
respondents. There seemed to be relative Becker and colleagues reviewed mul-

agreement on absolute indications such as tiple large series of IVC filters.17 Numerous
PE, while on therapeutic anticoagulation, methodological inconsistencies were iden-
the presence of a DVT with contraindica- tified within many of these studies. These
tion to anticoagulation, and free-floating authors concluded in their review that recur-
iliofemoral thrombosis by venogram or du- rent clinical PE was rare after filter place-
plex imaging. Relative indications were met ment and that only 8 deaths were reported in
with somewhat less agreement such as any the over 2500 cases that they reviewed. Filter
DVT documentation, spinal cord injury, complications were common but rarely life
pelvic fracture, multiple lower extremity threatening, and an extremely low rate of
fractures, concurrent cancer, prolonged bed death from filter complications was noted
rest, or obesity. Having the potential for re- among the reviewed studies. These authors
movability through use of a retrievable filter, concluded that IVC filters appeared to be ef-
many centers reported that they increased fective in preventing PE but not eliminating
the use of IVC filters for prophylactic use, it completely. Despite the large published
and this was found to be highly significant in experience with IVC filters, many questions
their study. Although the study documented remain about their indications, safety, and
few complications, it was a retrospective effectiveness. Anticoagulation therapy, if no
questionnaire-based analysis, which limits contraindication exists, ought to be contin-
its reliability. ued in conjunction with IVC filters when-
PE is one of the most feared complica- ever possible.
tions in trauma and neurosurgical trauma Mohan and colleagues conducted a
practice. Such patients with multisystem comparative analysis of multiple filter types
injuries, extremity or pelvic fractures, and (including various Greenfield type filters
head or spinal cord injuries often pose a such as the titanium and stainless steel,
significant dilemma for the surgeon or in- Venatech, Birds Nest, and Simon Nitinol
tensivist due to potential inability to use filters) and their complications.18 They con-
conventional anticoagulation therapy or at cluded that there is a higher rate of IVC
times even sequential compression devic- thrombosis with the Birds Nest filter vs. the
es. The incidence of DVT is high among various other filter types studied in their
trauma and neurosurgical trauma patients, analysis. They also found a higher mortal-
and the attendant risk of PE is an impor- ity rate with the use of the Birds Nest filter
tant cause of morbidity and mortality. IVC when compared to other types in this study.
filter placement has evolved and escalated However, this is a limited finding given the
in these groups over the past 3 decades. size of their study.
However, as discussed, significant rates Guidelines for the use of retrievable IVC
of complications are reported from their is an area of considerable attention at this
use in all groups.29 The available data on point in time.8,9,11,30 The practices of many
IVC filter placement in trauma and neu- interventionalists have evolved toward place-
rosurgery patients remain incomplete, and ment of retrievable filters as their first choice
the potential complications of IVC filters when there was any possibility or need for
must be carefully considered to optimally future retrieval. Kaufman and colleagues31
select patients who will benefit from their have endeavored to establish guidelines for
use. the use of retrievable filters through the

Society of Interventional Radiology (SIR). use, especially in prophylactic scenarios, will

Their consensus document emphasized continue to undergo further study and evo-
that the primary means for treatment and lution. In particular, these devices need to
prophylaxis of VTE are and should remain undergo cost effectiveness evaluation and it
pharmacologic. A few unique indications for is likely that practices surrounding their use
primary placement of retrievable vena cava will evolve significantly in the short and me-
filters exist and they are not distinct from the dium term.
indications for permanent vena cava filters. More aggressive percutaneous cathe-
Some patients with indications for place- ter-directed thrombolysis (CDT) of venous
ment of a retrievable vena cava filter have thrombosis is on the increase. Many authori-
limited periods of risk for clinically signifi- ties report a very low threshold to place, or
cant PE or contraindication to anticoagula- even routine placement of, a temporary IVC
tion and thus may not require permanent filter to protect against periprocedural PE
protection from PE with a vena cava filter. during CDT; however, such a strategy is
Patients with filters in place should be man- controversial. There remains a question as
aged by pharmacologic methods according to whether IVC filter deployment conveys
to their VTE status and risk of anticoagu- benefit to CDT patients. Protack and col-
lation as soon as anticoagulation is feasible leagues have sought to define the outcomes
and safe in their case. They concluded that of CDT with and without prophylactic IVC
there is no absolute indication for discontin- filter placement for lower extremity DVT.32
uation of filtration unless the filter itself is Average follow-up was 2 years. No patients
documented as a source of major morbidity developed PE during therapy, even though
that would be relieved by retrieval or conver- only 20% had IVC filter placement prior to
sion. They also noted that discontinuation of or during CDT. They concluded that CDT
filtration should occur only when the risk of without protective IVC filter placement is
clinically significant PE is reduced to an ac- safe and effective in treating acute DVT and
ceptable or baseline level and is estimated to that selective rather than routine IVC filter
be less than the risk of leaving the filter in placement is a safe and appropriate approach
place. Finally, they further concluded that in such patients.
there is not sufficient evidence in the litera-
ture to support true evidence-based recom-
mendations for the use of retrievable filters in references
patients with VTE or for prophylaxis of VTE 1. Decousus H, Leizorovicz A, Parent F, Page Y,
at this time. Nonetheless, many practitioners Tardy B, Girard P, et al. A clinical trial of vena
are choosing to use temporary filters as their caval filters in the prevention of pulmonary
filter of choice even in patients who are likely embolism in patients with proximal deep-vein
to require permanent filter placement. The thrombosis. Prvention du Risque dEmbolie
choice of retrievability, timing, and eventual Pulmonaire par Interruption Cave Study
final decision for removal is individualized Group. N Engl J Med. 1998;338(7):409-15.
in nearly all cases. Careful communication 2. Geerts WH, Bergqvist D, Pineo GF, Heit JA,
with either the patients primary physician Samama CM, Lassen MR, et al. Prevention
or surgeon may lead to more IVC filters re- of venous thromboembolism. Chest.
moved within the window of retrievability. 2008;133:381S-453S.
Retrievable filters and the evidence for their 3. Young T, Tang H, Aukes J, Hughes R. Vena

IVC Filters 133
caval filters for the prevention of pulmonary placement in the military multitrauma
embolism. Cochrane Database Syst Rev. patients: a single center experience. Vasc
2007;17(4):CD006212. EndovascSurg. 2009;43(5):497-501.
4. Kaufman JA. Development of a research 13. Wellons ED, Rosenthal D, Shuler FW, Levitt
agenda for IVC Filters. Endovascular Today. AB, Matsuura J, Henderson VJ. Real-time
2007;6(10):67-70. intravascular ultrasound-guided placement of
5. Abou-Nukta F, Alkhoury F, Arroyo K, Bakhos a removable inferior vena cava filter. J Trauma.
C, Gutweiler J, Reinhold R, et al. Clinical 2004;57(1):20-5.
pulmonary embolus after gastric bypass 14. Ebaugh JL, Chiou AC, Morasch MD,
surgery. Surg Obes Relat Dis. 2006;2(1): Matsumura JS, Pearce WH. Bedside vena cava
24-8. filter placement guided with intravascular
6. Georgiades CS, Hong K. Inferior vena cava ultrasound. J Vasc Surg. 2001;34(1):21-6.
filters. In: Cameron JL, ed. Current Surgical 15. Ashley DW, Gamblin TC, Burch ST, Solis
Therapy. 9th ed. Philadelphia: Mosby; 2008. MM. Accurate deployment of vena cava filters:
p908-13. comparison of intravascular ultrasound and
7. Usoh F, Hingorani A, Ascher E, Shiferson contrast venography. J Trauma. 2001;50(6):
A, Tran V, Patel N, et al. Superior vena cava 975-81.
perforation following the placement of a 16. Amankwah KS, Seymour K, Costanza M,
superior vena cava filter in males less than 60 Berger J, Gahtan V. Transabdominal duplex
years of age. Vascular. 2009;17(1):44-50. ultrasonography for bedside inferior vena
8. Berczi V, Bottomley JR, Thomas SM, Taneja cava filter placement: examples, technique,
S, Gaines PA, Cleveland TJ. Long-term and review. Vasc Endovascular Surg.
retrievability of IVC filters: should we abandon 2009;43(4):379-84.
permanent devices? Cardiovasc Intervent 17. Becker DM, Philbrick JT, Selby JB.
Radiol. 2007;30(5):820-7. Inferior vena cava filters: indications,
9. Comerota AJ. Retrievable IVC filters: a safety, and effectiveness. Arch Intern Med.
decision matrix for appropriate utilization. 1992;152(10):1985-94.
Perspect Vasc Surg Endovasc Ther. 18. Mohan CR, Hoballah JJ, Sharp WJ, Kresowik
2006;18(1):11-7. TF, Lu CT, Corson JD. Comparative efficacy
10. Hoppe H, Kaufman JA, Barton RE, Petersen and complications of vena caval filters. J Vasc
BD, Lakin PC, Deloughery TG, et al. Surg. 1995;21(2):235-45.
Safety of inferior vena cava filter retrieval in 19. Athanasoulis CA, Kaufman JA, Halpern
anticoagulated patients. Chest. 2007;132(1): EF, Waltman AC, Geller SC, Fan CM.
31-6. Inferior vena caval filters: review of a 26-year
11. Rosenthal D, Wellons ED, Levitt AB, single-center clinical experience. Radiology.
Shuler FW, OConner RE, Henderson VJ. 2000;216(1):54-66.
Role of prophylactic temporary inferior 20. Sharafuddin MJ, Corson JD. Percutaneous
vena cava filters placed at the ICU bedside devices for vena cava filtration. In: Ernst CB,
under intravascular ultrasound guidance in Stanley JC, eds. Current Therapy in Vascular
patients with multiple trauma. J Vasc Surg. Surgery. 4th ed. St. Louis: Mosby; 2001. p884-
2004;40(5):958-64. 91.
12. Aidinian G, Fox CJ, White PW, Cox MW, 21. Patel NC. Inferior vena cava interruption.
Adams ED, Gillespie DL. Intravascular In: Waldman DL, Saad WEA, Patel NC,
ultrasound-guided inferior vena cava filter eds. Interventional Radiology Secrets.

Philadelphia: Hanley & Belfus; 2004. inferior vena cava filters. J Vasc Interv Radiol.
p206-15. 2009;20(9):1193-9.
22. Hann CL, Streiff MB. The role of vena 28. Quirke TE, Ritota PC, Swan KG. Inferior
caval filters in the management of venous vena caval filter use in U.S. trauma centers:
thromboembolism. Blood Reviews. a practitioner survey. J Trauma. 1997;43(2):
2005;19(4):179-202. 333-7.
23. Hoppe H, Kaufman JA. Inferior vena cava 29. Giannoudis PV, Pountos I, Pape HC, Patel
filters. In: Kandarpa K, ed. Peripheral Vascular JV. Safety and efficacy of vena cava filters in
Interventions. Philadelphia: Lippincott trauma patients. Injury. 2007;38(1):7-18.
Williams & Wilkins; 2007. p401-15. 30. Girard P, Stern J-B, Parent F. Medical
24. Ascher E, Hingorani A, Yorkovich WR. literature and vena cava filters: so far so weak.
Complications of vena cava filters. In: Towne Chest. 2002;122:963-7.
JB, Hollier L, eds. Complications in Vascular 31. Kaufman JA, Kinney TB, Streiff MB, Sing RF,
Surgery. 2nd ed. New York: Marcel Dekker; Proctor MC, Becker D, Cipolle M, Comerota
2004. p. 569-79. AJ, Millward SF, Rogers FB, Sacks D, Venbrux
25. Harris EJ Jr, Kinney EV, Harris EJ Sr, Olcott AC. Guidelines for the use of retrievable
C 4th, Zarins CK. Phlegmasia complicating and convertible vena cava filters: report
prophylactic percutaneous inferior vena caval from the Society of Interventional Radiology
interruption: a word of caution. J Vasc Surg. multidisciplinary consensus conference. J Vasc
1995;22(5):606-11. Interv Radiol. 2006;17(3):449-59.
26. Greenfield LJ, Proctor MC. Twenty-year 32. Protack CD, Bakken AM, Patel N, Saad
clinical experience with the Greenfield filter. WE, Waldman DL, Davies MG. Long-term
Cardiovasc Surg. 1995;3(2):199-205. outcomes of catheter directed thrombolysis
27. Cantwell CP, Pennypacker J, Singh H, Scorza for lower extremity deep venous thrombosis
LB, Waybill PN, Lynch FC. Comparison without prophylactic inferior vena cava filter
of the recovery and G2 filter as retrievable placement. J Vasc Surg. 2007;45(5):992-7.

CHAPTER
12
Percutaneous
Pulmonary embolectomy
Saher Sabri, Wael e. A. Saad, and Alan H. Matsumoto
V enous thromboembolic disease re-

mains the third most common car-
diovascular disease and one of the leading
as possible. The therapeutic options that are
available should be tailored to each patient
and clinical scenario. In this chapter, we dis-
causes of sudden death in the United States. cuss the management of patients with acute
The true incidence of pulmonary embo- PE using catheter-directed thrombolysis,
lism (PE) is unknown, but based on his- percutaneous embolectomy and embolus
toric projections, it is estimated that more fragmentation techniques, and/or surgical
than 600,000 cases of PE occur every year embolectomy. The role of anticoagulation is
in the United States.1 Approximately 10% of not addressed in this chapter.
patients with PE do not survive their initial
event. Of those who do survive, approxi-
mately 70% fail to have the diagnosis made Endovascular and
and experience a mortality rate of 30%. If
the diagnosis of PE is made promptly and Surgical Interventions
appropriate therapy initiated, the mortality
rate can be reduced to less than 10%.1,2,3
Once the diagnosis of acute PE is Indications
made, treatment should be initiated as soon When obstruction of 70% of the pulmonary
arterial circulation occurs, the right ventricle
needs to be able to generate a systolic pres-
sure in excess of 50 mm Hg and a mean
Davies md and Alan B. Lumsden md, eds. pulmonary artery pressure greater than 40
Cardiotext Publishing, ISBN 978-1-935395-22-5. mm Hg to maintain pulmonary perfusion.
135

A previously normal right ventricle is incapa- Echocardiographic findings indicating

ble of generating a systolic pressure exceed- right ventricular dysfunction as
ing 50 mm Hg, so any incremental embolic manifested by RV dilation and
obstruction to the vasculature beyond this afterload stress with pulmonary
point results in right ventricular failure.4 The hypertension
degree of obstruction of the pulmonary arte- Failure of or contraindication to
rial circulation required to cause a change in anticoagulation and thrombolysis
the pulmonary arterial hemodynamics also in a patient with compromised
depends upon the amount of underlying car- hemodynamics or persistent
diopulmonary disease prior to an embolic hypoxemia
event. Therefore, a patient with massive PE
obstructing the majority of the pulmonary Surgical embolectomy rather than catheter
circulation or a submassive PE superposed thrombectomy should be considered in the
on underlying cardiopulmonary disease may presence of free-floating cardiac thrombi or
present with right ventricular dysfunction or in patients with paradoxical embolism from
compromised hemodynamics. In this subset a large atrial septal defect.7,8
of patients, anticoagulation therapy alone
may not be adequate and more aggressive in-
tervention with thrombolysis and/or pulmo- Intravenous Thrombolytic Therapy
nary embolectomy and clot fragmentation The Consensus Development Conference
techniques should be considered. recommended that 2-hour bolus intravenous
The most validated risk-assessment (IV) thrombolytic therapy be considered in
tool is echocardiography. Right ventricular any patient who has a perfusion defect in-
hypokinesis on echocardiography predicts volving equivalent to one or more lobes and
a doubling of mortality within the next 30 hemodynamic compromise.9,10 Despite the
days, even among initially normotensive numerous randomized trials demonstrating
patients.5 Right ventricular enlargement on faster improvement in pulmonary perfusion
chest computed tomography (CT) also por- and hemodynamics and better lung diffus-
tends a greater likelihood of death or major ing capacities and pulmonary capillary blood
in-hospital complication.6 Accepted indica- volumes in patients receiving thrombolytic
tions for catheter-directed therapy are: therapy, when symptoms and mortality rates
at 6 and 12 months are analyzed, there is
Hemodynamic instability, defined as a no statistical benefit in outcomes between
SBP of <90 mm Hg, a drop in SBP of patients who received IV thrombolytic vs.
>40 mm Hg, or ongoing administration heparin therapy.11,12 In addition, bleeding
of catecholamines for systemic arterial complications were more frequent in pa-
hypotension or persistent hypoxemia tients undergoing 2-hour IV bolus throm-
despite appropriate anticoagulation and bolysis. Therefore, screening patients with a
oxygen therapy careful history and physical examination and
Subtotal or total occlusion by a review of old medical records is extremely
embolus of the left and/or right main important to exclude contraindications for
pulmonary artery by chest computed thrombolysis (which were discussed in detail
tomography (CT) or by conventional in earlier chapters). The argument that most
pulmonary angiography patients with acute PE will have a contrain-

Percutaneous Pulmonary Embolectomy 137
dication for thrombolytic therapy is not sup- teplase, Genetech, South San Francisco,
ported by a large patient survey that revealed CA) for the treatment of acute massive PE.
that 50% of patients with high-probability The findings of this study suggested that the
lung scans or pulmonary angiographic evi- intrapulmonary infusion of rtPA does not
dence for PE are acceptable candidates for offer significant benefit compared to intra-
treatment with IV thrombolysis.13 venous administration. However, this study
did not use the standard catheter-directed
technique currently used by most interven-
Catheter-Directed Thrombolysis tionalists, which includes embedding the
Catheter-directed thrombolytic therapy with infusion catheter directly into the thrombus,
intrapulmonary administration of a throm- while attempting to fragment the clot. Since
bolytic agent has been described by several rtPA must cleave clot-bound plasminogen
investigators with encouraging results.14,15 to create the active enzyme plasmin, infu-
Catheter-directed techniques aim to acceler- sion of rtPA in the main pulmonary artery
ate clot lysis and achieve rapid reperfusion confers minimum benefit to simple IV infu-
of the pulmonary arteries and lung paren- sion as most of the drug does not penetrate
chyma. In one study, 13 patients were treated the clot, but rather, flows in the path of
with urokinase (Abbott Labs, Abbott Park, least resistance (the patent pulmonary arte-
IL) for angiographically proven PE within rial segments). In addition, experience with
14 days of major surgery.14 The catheter peripheral arterial bypass grafts and throm-
was positioned in the pulmonary artery clot bosed dialysis fistulas has demonstrated that
and 2200 IU/kg of urokinase were injected catheter-directed thrombolysis is associated
directly into the clot. Continuous infusion with better rates of lysis, more rapid lysis, the
of urokinase at 2200 IU/kg/h until the clot need for lower doses of the lytic agent, and
lysed or up to a maximum of 24 hours was fewer complications.17 Ultrasound-enhanced
then performed. Follow-up pulmonary angi- catheter-directed thrombolysis using the
ography at 24 hours revealed that 98% of the EndoWave system (EKOS, Bothell, WA) has
clots had completely disappeared from the been used by some investigators to acceler-
pulmonary vasculature. No deaths or bleed- ate the thrombolysis. In one study the me-
ing complications occurred. In another se- dian infusion time was 24 hours with 76%
ries, 16 patients with massive PE were given complete resolution of clot burden.18
a bolus of 50,000 IU of urokinase directly
into the clot.15 An infusion of 1,000,000 IU
of urokinase was then given into the right Surgical Embolectomy
atrium over a 12-hour period. Cardiac out- For patients with massive central PE with
put, total pulmonary vascular resistance, marked compromised hemodynamics who
and mean pulmonary artery pressures all are too unstable to undergo or have a contra-
improved following the thrombolytic thera- indication to thrombolytic therapy, the only
py. One patient did suffer a severe bleeding remaining therapeutic options are percuta-
complication. neous techniques with clot fragmentation or
In 1988, Verstraete et al16 published the surgical embolectomy. Traditionally, surgi-
results of a multicenter comparative study cal embolectomy has been associated with
of intravenous vs. intrapulmonary infusion a high perioperative mortality rate. With
of rtPA (100 mg over a 7-hour period) (Al- improvements in anesthesia and cardiopul-

monary bypass technology, 30-day survival ventricular function. In the absence of one
rates after surgical pulmonary artery embo- of these noninvasive imaging studies (ie, a
lectomy have been reported to be as high as ventilation/perfusion nuclear medicine lung
75% at centers specializing in the treatment scan), a pulmonary angiogram is performed
of thromboembolic disease. In one study, 20 using a minimum amount of contrast to
patients who presented with acute PE and document the presence and distribution of
cardiogenic shock had a 5-year survival of PE and to measure pulmonary arterial and
100% following surgical embolectomy. In right heart pressures. However, pulmonary
this study, patients undergoing surgical em- arterial pressures (PAP) should be measured
bolectomy had a more favorable New York prior to the injection of any contrast and
Heart Association classification level and a prior to any intervention. Most operators
lower incidence of chronic pulmonary ar- agree that the injection of 10 mL of contrast
terial hypertension at follow-up when com- is sufficient to obtain a baseline pulmonary
pared to a similar cohort of patients receiving angiogram, especially since the diagnosis of
medical therapy.19 In a more recent study, 47 a massive PE has often already been made
patients underwent surgical thrombectomy with another imaging modality. The use of
at a single center with 6% perioperative mor- a large amount of contrast (ie, >20 mL with
tality and 83% 3-year survival.20 one injection) may be dangerous due to the
Yet, very few centers in the United risk of worsening right ventricular failure
States have widespread experience with this and cardiogenic shock from acute volume
type of surgery. Therefore, various percuta- overload.7
neous transvenous devices designed to re- If the patient is unable to lay supine,
move or fragment centrally obstructing PE the pulmonary angiogram and catheter-di-
have been developed. rected intervention can be performed from
a basilic or brachial vein approach. Other-
wise, the study is usually performed from
Percutaneous Embolectomy the femoral vein or less commonly, the in-
ternal jugular vein approach. The diagnostic
Technique pulmonary arterial catheter is removed over
We advocate that all patients undergoing a guidewire, maintaining access in the pul-
percutaneous embolectomy be intubated monary artery. An appropriate-sized 70- or
and on mechanical ventilation, preferably 90-cm introducer sheath (usually 6F11F
under heavy sedation or general anesthesia. depending on the device used) is inserted
Meticulous attention is also required for on- into access vein and positioned with its tip
going vasopressor, fluid, electrolyte, and ven- in the main pulmonary artery. The sheath
tilator management. Patient management can be used to perform suction embolecto-
includes continuous assessment of volume my, but more importantly, it is used to mini-
status, urine output, end-organ perfusion, mize trauma to the heart and stabilize the
and ECG monitoring during the procedure. various mechanical thrombectomy devices
Most often, a computed tomographic (CT) during their use in the pulmonary arterial
or magnetic resonance (MR) pulmonary segments. The sheath is connected to a hep-
arterial study or an echocardiogram dem- arinized saline flush (4000 IU heparin/1000
onstrates the clot distribution and burden cc normal saline) and infused at 30 cc/hr.
and the presence of compromised right The sheath side port can also be connected

to a pressure transducer to allow real-time Falls, NY or EV3, Minneapolis, MN) can be

monitoring of the main pulmonary artery used to pulse-spray a bolus of the lytic agent
during the intervention. into the clot to effect better drug distribu-
Performing catheter-directed intervention; however, use of an end-hole catheter
tions in the segmental arteries is discouraged that could erode through a pulmonary ar-
due to the risk of arterial perforation. Unless tery branch during cardiac pulsations obvi-
strongly contraindicated, the use of intraclot ates against its use for infusion therapy. For
infusion of thrombolytic agents with the intraclot infusion of a lytic agent, a pigtail
appropriate embolectomy device is recom- catheter that has been embedded into the
mended. A 5 or 6 F pigtail catheter can be clot is recommended.
embedded directly into the clot, manually The available literature on the use
rotating the catheter with hopes of fragment- of the various percutaneous embolectomy
ing the clot and creating more surface area devices is summarized in Table 12.1. The
for the lytic agent. A pigtail catheter is used published experience with the use of these
for this application since it has multiple side devices is limited to small case series and
holes, and its rounded configuration is less case reports from single institutions. In ad-
likely to traumatize the pulmonary artery. dition, many of the catheter-directed stud-
Typically, 4 to 10 mg of rtPA is vigorous- ies for the treatment of acute PE described
ly injected in small aliquots (pulse-sprayed) the use of a combination of techniques
using the pigtail catheter or power-pulsed in which local or systemic thrombolytics
using a dedicated mechanical device such as were applied in a very heterogeneous pa-
the AngioJet Xpeedior catheter (MEDRAD, tient population, which makes it difficult
Inc., Warrendale, PA) directly into the clot to compare the efficacy and outcomes re-
(Figure 12.1). Alternatively, a multi-side hole lated to the use of these devices. Given
infusion catheter (AngioDynamics, Glen these limitations, the summary of the re-
Figure 12.1. 71-year-old male with acute shortness of breath. A. Oblique coronal MPR CT. B. Left pulmonary
angiogram demonstrates left PA embolus. Mean PAP was 42 mm Hg. C. Angiogram post AngioJet and catheter-
directed thrombolysis with tPA over 12 hours. Mean PAP decreased to 30 mm Hg.

Table 12.1.SummaryofPublishedArticlesonCatheter-Directed 140
InterventionsforAcutePulmonaryEmbolism
MeanBP MeanBP MeanPAP MeanPAP Clinical

Patient pre(mm post(mm pre(mm post(mm Success Mortality
Technique no. Hg) Hg) Hg) Hg) (%) (%)
LD2 VTD Nearly Final Pages.indd 140

Aspiration (Greenfield embolectomy
device or guiding catheters)
No lytics 89 60 81 33 21 81 25
Systemic lytics 9 50 87 31 20 100 10
Local lytics 9 31 24 100 0
Fragmentation (balloons, rotating
pigtail, or standard catheters)
No lytics 3 28 63 38 29 67 0
Systemic lytics 21 70 93 25 21 71 5
Local lytics 121 67 81 33 22 95 4
Amplatzerthrombectomydevice
(ATD)
No lytics 8 86 106 49 53 88 12
Local lytics 6 85 93 100 0
Hydrodynamicthrombectomy
devices
Hydrolizer
Local lytics 12 47 97 46 30 92 8
Systemic plus local 8 43 36 100 12
AngioJet
No lytics 25 42 30 75 0
Local lytics 21 87 13
Combinationoftechniques
FragmentationplusApirex 18 74 88 37 31 89 11
Fragmentation,aspiration, 12 83 17
hydrodynamicdevicespluslytics
Adapted from Kucher N. Catheter embolectomy for acute pulmonary embolism. Chest. 2007;132:657-63. With permission.
5/13/11 10:27 AM
sults from these series reveals an overall ported on their experience using this device
clinical success rate with catheter-directed in 46 patients over a 22-year period.21 Em-
therapy for acute PE of >80% (Table 12.1), boli were extracted in 35 (76%) of 46 pa-
with clinical success being defined as im- tients. There was an average reduction in
mediate hemodynamic improvement. The mean pulmonary artery pressure of 8 mm
reported mortality rates range from 0% to Hg and a significant increase in mean car-
25%, again reflecting the wide variations in diac output after embolectomy. The 30-day
the patients being treated. mortality rate was 30%. When subgroups of
The ideal thrombectomy device patients were analyzed, embolectomy was
should: (1) be easy to use and position with- most successful for major and submassive
in the pulmonary artery clots, (2) be highly PE, and least likely to be helpful in patients
maneuverable to allow rapid right heart with chronic, recurrent PE. Experience
passage and advancement into major pul- with the Greenfield suction embolectomy
monary arteries, (3) be able to promote com- catheter has been relatively small. The de-
plete removal of clots or fragmentation of vice is somewhat bulky, requires familiarity
clots into very small particles, and (4) have with the control handle, and is designed for
a low profile and be safe to use in the pul- insertion via a surgical venotomy, since it
monary circulation.7,8 A review of the most requires insertion through a 22F sheath.
commonly used devices is detailed below. The device was not widely used, so it is no
None of the currently available devices de- longer manufactured.
scribed below are FDA-approved for applica-
tion in the pulmonary arterial system, and
their use for acute PE represents an off-label
use of these devices.
Greenfield Embolectomy Device

The Greenfield device (Boston Scientific,
Natick, MA) (Figure 12.2) is a 10F braid-
ed catheter designed for pulmonary artery
embolectomy. The catheter is maneuvered
by using a large control handle and is de-
signed for insertion via a femoral or jugu-
lar venotomy. The tip of the catheter has
threads on it that allow the use of either a
5-mm or 7-mm diameter plastic cup. Once
the cup on the catheter tip comes into con-
tact with the embolic material, manual Figure 12.2. The Greenfield suction embolectomy
suction is generated via a side port on the device consists of a large handle with a joy stick
control handle. The catheter and the clot (straight arrow) that is used to control the steerable
are then removed as a unit through the ve- catheter. A 5-mm cup (curved arrow) is seen on the
end of the catheter. From Uflacker R. Interventional
notomy site or a vascular sheath. Multiple
therapy for pulmonary embolism. J Vasc Interv Radiol.
passes with the catheter may be required.
2001;12:147-64. Used with permission from Elsevier.
Dr. Greenfield and his colleagues have re-

(Figure 12.3). An air turbine can generate

Amplatz Thrombectomy Device up to 150 rpm at 50 psi. The high speed of
The Amplatz Thrombectomy Device (ATD) the impeller creates a vortex that recirculates
(EV3) consists of a 120-cm-long, 8F, poly- and pulverizes acute clot, creating a fluid
urethane catheter with an impeller mounted with very small, suspended particles. There
on a drive shaft inside a metal cap 5 mm in are 2 side ports in the distal catheter that
length. The metal cap has 3 side ports that allow for high-pressure infusion of saline to
are used for recirculation of clot particles reduce friction and cool the system or to in-
ject contrast medium to facilitate visualiza-
tion during fluoroscopic manipulation.22
The device is advanced through a mul-
tipurpose 10F guiding catheter positioned
into the clot. To avoid bends and kinks, the
device should be introduced and advanced
very carefully through the guiding catheter.
The device is advanced in a slow, back-and-
forth motion. The multipurpose configura-
tion of the guiding catheter allows for some
degree of steerability of the device inside the
Figure 12.3. A drawing of the distal tip of the ATD pulmonary artery. After thrombectomy with
shows a recessed impeller within a 5-mm metal capsule. the ATD, some of the resulting fluid may
Thrombus is aspirated, liquefied, and then expelled be aspirated through the guiding catheter.
through the side ports (curved arrows). The authors
Contrast medium can be injected through
would like to acknowledge Leanne Lessley for her expert
the guiding catheter to visualize the amount
help in image illustration.
of residual clot (Figure 12.4).
Figure 12.4. A. Left pulmonary angiogram demonstrates left main pulmonary embolus.B. Amplatz
Thrombectomy Device (ATD) was used for 7 minutes. C. Post-angiogram demonstrates decrease in clot
burden.

Initial experience with the ATD device with local thrombolytic therapy and
showed clinical improvement in a limited achieved clinical success in one patient with
group of patients, with reduction of the re- no complications. However, we encountered
spiratory symptoms and improvement of an intraprocedural death in one patient with
hypotension.22,23 Transient hemoptysis and massive main pulmonary artery clot who was
arrhythmias have been described as compli- hypotensive and moderately hypoxemic prior
cations associated with the use of the ATD. to initiation of the catheter-directed treat-
Hemolysis also occurs commonly with the ment. Upon activation of the over-the-wire
use of the ATD, but there are no reported Trerotola device in the main pulmonary, the
cases of associated renal failure. The use of patient developed electrical-mechanical dis-
the ATD has been limited recently due to its sociation on our procedure table and could
bulkiness and lack of steerability. not be resuscitated. The experience with this
device remains very limited and its safety in
native pulmonary arteries is unproven.
Arrow-Trerotola Percutaneous
Thrombolytic Device
Thrombus Fragmentation Catheters
The Arrow-Trerotola device (Arrow, Read-
ing, PA) is a low-speed (3000 rpm) rotational and Balloons
basket designed for thrombectomy in dialy- The theoretical advantage of the fragmenta-
sis grafts. It scrapes the walls of the vessel tion technique is that the central pulmonary
and fragments the thrombus. The device artery volume is roughly 50% of the volume
was modified for the treatment of PE in an of the branch pulmonary arterial segments.
animal model and was shown to be effective Therefore, by achieving immediate redistri-
in fragmenting the clots and relatively safe bution of the occlusive thrombus from the
for treating large acute central pulmonary central main pulmonary artery to the more
emboli. A modified 8F, 120-cm-long device peripheral pulmonary artery branches, the
was redesigned with a nitinol wire basket afterload on the right ventricle is immedi-
that measures 9 to 15-mm in diameter when ately reduced. In addition, following clot
expanded. When this device was applied in fragmentation, a greater surface area of the
a porcine pulmonary artery, histologic speci- thrombus is exposed to allow greater activa-
mens showed that there was moderate acute tion of clot-bound plasminogen to plasmin
intimal injury, but no evidence of pulmonary by the infused lytic agents, if thrombolysis is
artery disruption. In a case report on the use used in combination with the fragmentation
of the Trerotola device (7F 80-cm-long over- technique.
the-wire device) for mechanical thrombec-
tomy of massive pulmonary embolism,24 the Balloon Catheters
device was found to be difficult to direct into Balloon angioplasty (616 mm in diameter)
some of the vessels being treated. In this for fragmentation of large central pulmonary
case report, there was no improvement in emboli has been used in association with
pulmonary pressures and large portions of local thrombolytic infusion with encourag-
clot remained untreated, although clinical ing results. Recovery rates of 87.5%, as mea-
improvement was observed. We have per- sured by pulmonary artery pressures, blood
formed 2 procedures at our institution com- O2 values, and clinical outcomes, have been
bining the use of the over-the-wire Trerotola reported with use of this technique.25

Angiographic and Pigtail Catheters but usually in combination with catheter-di-

Various angiographic or pigtail catheter de- rected thrombolysis or suction embolectomy
vices have been used to fragment centrally via large guiding catheters (10F). Distal mi-
located emboli by direct mechanical action. gration of large clot fragments does remain a
The majority of the reported patients were concern with clot fragmentation techniques
also treated with local or systemic throm- without adjuvant thrombolysis, as it may lead
bolysis26,27 (Table 21.1); therefore, it is un- to acute worsening in the hemodynamic sta-
clear whether thrombus fragmentation with tus of the patient depending upon how the
a catheter without thrombolysis is effective. clot redistributes.
The rotatable pigtail catheter is a cus-
tom-made 5F pigtail catheter. It is 110 cm
in length and has 10 side holes for contrast
injection (Figure 12.5). The catheter is intro-
duced via a flexible 5.5F sheath. The cath-
eter is designed to be used over an 0.035-in
movable-core J wire. An electrical motor is
connected to a luer lock adapter on the proxi-
mal hub end of the catheter and can generate
rotational speeds up to 500 rpm. The pigtail
catheter is designed to rotate within the
sheath with a guidewire exiting through the
distal side hole. The sheath prevents precess-
ing of the catheter shaft, thereby minimizing
damage along the venous access route. The
wire serves as a rigid central axis for pigtail
rotation. During activation of the electrical
motor, the catheter can be advanced and
pulled back over the wire. The catheter can
also be used for follow-up angiograms after
clot fragmentation.8 In 20 patients with
massive PE, catheter intervention with the
rotatable pigtail catheter showed a 33% re-
canalization rate by fragmentation, but the
catheter was more effective with adjuvant
thrombolytic therapy. The mortality rate in
this series was 20%.28 A recent study reported
Figure 12.5. The drawing depicts the rotation of the
on the use of a modified rotatable pigtail pigtail catheter about the axis of a stationary guidewire.
catheter with or without thrombolytics and Emboli are fragmented by the mechanical action of
showed shorter hospital stay and higher sur- the rotating catheter loop. During rotation, the pigtail
vival rate when utilizing catheter fragmen- catheter can be advanced or withdrawn over the
tation and aspiration without thrombolytics guidewire. Side holes are present within the catheter
to allow contrast injection following the fragmentation
than when thrombolytics are added.29
procedure. The authors would like to acknowledge
In summary, clot fragmentation tech-
Leanne Lessley for her expert help in image illustration.
niques have been employed with success,

tion channel with metallic tubing looped

Suction Embolectomy with Guide 180 to enable retrograde injection of fluid
Catheters/Sheaths at high velocities. High-velocity injection
Manual suction embolectomy had been through the small lumen of the metallic tub-
used alone or as an adjunct to other tech- ing creates lower pressure dynamics in the
niques. An 8F16F guiding catheter/sheath larger lumen. A Bernoulli effect is created,
is advanced into the thrombus in the main resulting in a vortex, causing fragmentation
right or left pulmonary artery (PA). A 10- to of the clots by the pressure gradient.8 The
30-mL syringe with a luer lock connector is fragmented clots can then be aspirated via
then used to apply suction while the cath- the larger lumen. The Hydrolyzer has been
eter is moved slowly to and fro over several reported to be effective in removing an acute
centimeters within the PA. During advance- clot from vessels up to 9 mm in diameter.
ment, it is important to be aware of any re- The device is no longer available in the Unit-
sistance, which may indicate subintimal ed States.
passage. When blood readily enters the sy- The AngioJet Xpeedior (MEDRAD,
ringe, the material has cleared the catheter. Inc.) is a 6F over-the-wire mechanical throm-
The syringe is then removed and its contents bectomy device and is probably the most
are expressed over a gauze-draped basin. It efficacious catheter among the currently
is necessary to readvance the catheter into available hydrodynamic devices (see Figure
the PAs for each successive aspiration. This 8.1 on page 91). Power saline jets at speeds
technique has been described with or with- up to 300 miles per hour are injected in a ret-
out local thrombolytics.30,31 In a recent study, rograde fashion to create a low-pressure zone
clinical success was 100% in 15 patients uti- around the catheter tip, causing a Bernoulli
lizing an 8Fr guiding catheter. In another effect and a recirculating vortex. The in-
study, the survival rate utilizing this tech- jected saline is removed in a euvolemic fash-
nique was 72%. The study also showed that ion via a suction port on the catheter. The
this technique is more efficacious with acute thrombus is withdrawn toward the catheter
PE when the procedure is performed < 48 into the vortex and fragmented into small
hours from the onset of symptoms.32 particles that are, in theory, removed along
with the injected fluid through the suction
port of the device. Since the AngioJet Xpeed-
Hydrodynamic Thrombectomy
ior was not designed to treat vessels larger
Devices than 12 mm in diameter, it is also of limited
Although none of the currently available effectiveness in the treatment of central PE
hydrodynamic or rheolytic catheter devices located in larger arteries. However, the re-
were designed for the treatment of large sultant disruption of the clot with the device
arteries, they have been successfully used and enhancement in pulmonary perfusion
in an off-label fashion for the treatment of often is sufficient to improve hemodynamics
patients with massive PE. The Hydrolyzer and clinical outcomes in these patients.7,33-36
(Cordis; Warren, NJ) is a 7F, 80-cm over- In one study, 14 patients were treated with
the-wire catheter, with a large side hole near this device. Adjunctive local thrombolysis
the distal tip. The larger catheter lumen is was performed in 5 patients. Clinical suc-
used for aspiration of the fragmented clots, cess was obtained in 86% of patients. Proce-
and the smaller lumen serves as the injec- dural mortality occurred in one patient who

presented in cardiogenic shock and nonfatal pressure jets. The effect of adenosine released
hemoptysis occurred in one patient.35 In an- from lysed cells on the atrioventricular node
other study, 51 patients were treated with the has also been suggested as a possible mecha-
AngioJet device, 21% of whom also received nism. Most typically, the bradyarrhythmias
local thrombolytic therapy. The in-hospital will stop within 10 seconds of deactivating
mortality rate was 15%.36 the device. In addition, activation of the de-
The use of this device close to the heart vice for short bursts (ie, less than 10 seconds)
is commonly associated with bradyarrhyth- with 15- to 20-second pauses between device
mias, including transient asystole, which activation minimizes the occurrence of the
can cause significant symptoms and hinder bradyarrhythmias. A few procedural-related
its use in some patients. The incidence of deaths with use of the AngioJet for acute
the bradyarrhythmias appears to increase PE have been described in the literature,
with the proximity of the device to the heart which leads many operators to recommend
and the duration of device activation. The against using this device in the pulmonary
incidence of bradyarrhythmias with the use circulation. It is worth mentioning, how-
of this device is 20% to 79% in patients un- ever, that we have used the AngioJet device
dergoing coronary artery thrombectomy.37 In in combination with local thrombolytics
one series, 2 of 17 patients (12%) who un- for pulmonary embolectomy in 16 patients
derwent pulmonary thrombectomy with the and have had no procedural-related mortali-
AngioJet had bradyarrhythmias.34 In another ties. Bradyarrhythmias were common, but
series 8% of patients required tranvenous transient (Figure 12.6). We emphasize that
pacing for significant bradycardia.36 The careful monitoring of the cardiac rhythm
cause of AngioJet-induced bradyarrhythmia and hemodynamic parameters is paramount
remains unknown. Some authors suggest during the use of this device. As mentioned
that the arrhythmias could be related to earlier, mechanical ventilation is encour-
activation of stretch receptors by the high- aged as the activation of this device in the
Figure 12.6. 72-year-old with acute chest pain. A. Coronal MPR CT demonstrates a saddle embolus extending into
right main PA. Echocardiogram showed moderate right atrium dilatation and right ventricular dysfunction.B.Right
pulmonary angiogram demonstrates right PA embolus. Mean PAP on initial study was 34 mm Hg. C.Post AngioJet and
12 hours of thrombolysis angiogram demonstrates significant decrease in clot burden. Mean PAP was 15 mm Hg.

pulmonary circulation is commonly associ- enhanced flexibility, which facilitates its

ated with the patient complaining of a sense passage through the right side of the heart
of difficulty breathing and, on occasion, into the pulmonary arteries. A recent study
becoming apneic. Again, a judicious proto- on the use of this device in 11 patients re-
col of a short duration of device activation, ported a clinical success rate of 88%.38 A
followed by a discrete pause, seems to mini- cohort study is currently being performed in
mize the occurrence of bradyarrhythmias. Europe to investigate the effectiveness and
Other complications related to this de- safety of the Aspirex device in patients with
vice include hemoglobinuria and renal in- acute, massive PE and a contraindication for
sufficiency. Therefore, we will often employ thrombolysis.7
a sodium bicarbonate drip to alkalinize the
urine and facilitate excretion of the free he-
moglobin when we use the AngioJet device.
Complications of Catheter-Directed
Interventions
Complications of catheter-directed inter-
Aspirex PE Catheter ventions for acute PE include perforation or
The 11F Aspirex catheter thrombectomy de- dissection of cardiovascular structures, peri-
vice (Straub Medical; Wangs, Switzerland) cardial tamponade, pulmonary hemorrhage,
was specifically designed and developed distal thrombus embolization, and death.7,39
for percutaneous interventional treatment Other potential complications include chest
of acute PE in arteries ranging from 6 to pain, hemoptysis, blood loss, arrhythmias,
14 mm in caliber. The central part of the contrast-induced nephropathy, anaphylactic
catheter system is a high-speed rotational reaction to iodine contrast, hemolysis, and
coil (40,000 revolutions per minute) within vascular access complications such as hema-
the catheter body that creates negative pres- toma, access site thrombosis, arterial punc-
sure through an L-shaped aspiration port at ture, pseudoaneurysm, or arteriovenous
the catheter tip. The rotating coil macerates fistula. To minimize the risk of vascular
aspirated thrombus, and removes thrombus perforation or dissection, mechanical throm-
fragments via an augerlike action (Figure bectomy should be performed only in the
12.7). The distal part of the catheter has main and lobar pulmonary arteries, not in
the segmental pulmonary arteries. The pro-
cedure should be performed with a guiding
sheath positioned in the main pulmonary to
minimize trauma to the heart and the induc-
tion of arrhythmias. The procedure should
be terminated as soon as hemodynamic im-
provement is achieved, regardless of the an-
giographic result.7,8
Fewer hemorrhagic complications
should be seen with catheter-directed throm-
bolysis than with bolus intravenous infusion,
Figure 12.7. The tip of Aspirex catheter but no randomized control data have dem-
thrombectomy device. Used with permission from Straub
onstrated this difference. However, the rate
Medical, Wangs, Switzerland.
of major bleeding was reported at 6% with

catheter-directed infusion of rtPA compared persistently hypoxemic despite appropriate

with 27% and 12%, respectively, for periph- anticoagulation and supplemental oxygen
erally infused urokinase in phases I and II of administration, should be considered for
the UPET study.8 more aggressive intervention.
Thrombolytic therapy appears to be
useful in rapidly restoring more normal he-
IVC Filter Placement modynamics and may prove to be useful in
In patients with a contraindication to anti- reducing the sequela of chronic pulmonary
coagulation, an IVC filter is placed before hypertension. With a greater understand-
initiation of catheter-directed therapy. In ing of the risks associated with fibrinolysis,
patients with compromised hemodynamics, bleeding complications can be reduced.
even if they can be anticoagulated, an op- The use of the intravenous route is favored
tional IVC filter is often placed prior to ini- when there is diffuse, bilateral acute PE, the
tiation of catheter-directed interventions for patient is young, and the patient has abso-
massive PE. We then work through the filter lutely no contraindication to thrombolysis.
to perform our catheter-directed therapy if In situations in which there is markedly
the access is from the femoral vein. Once we asymmetric clot burden or the patient is at
have completed treatment of the PE, if the some risk for bleeding, or the patient is very
patient can be fully anticoagulated and has hemodynamically compromised and there
no further need for the IVC filter, the filter is is asymmetric clot burden, catheter-directed
removed. If the clinical status of the patient techniques rather than IV therapy are rec-
is tenuous, the IVC filter remains in place ommended. Direct intrathrombus infusion
until all consultants agree that there is no of lower doses of the thrombolytic agent may
further need for the filter. also reduce bleeding complications without
decreasing the beneficial fibrinolytic affect.
Although percutaneous embolectomy
Conclusion and fragmentation techniques are intriguing,
most of the devices are not widely available.
Once the diagnosis of acute PE has been es- Despite the lack of availability of these percu-
tablished, therapy should be initiated as soon taneous devices, several studies have shown
as possible to minimize the morbidity and that simple catheter techniques aiming at
mortality associated with the embolic event mechanical fragmentation of the clot using
and to enhance the chances for pulmo- angioplasty balloons and pigtail catheters,
nary artery recanalization. Patients who do coupled with either suction embolectomy
not have any underlying cardiopulmonary and/or local thrombolytic administration,
disease and are not compromised by their have resulted in dramatic improvements in
acute embolic event can usually be treated patients with massive, acute PE.27,40 Howev-
with a 3- to 6-month course of anticoagula- er, older, more organized thrombus may be
tion. Patients who have a moderate volume refractory to thrombolysis and mechanical
of clot (>1 lobe) and are hemodynamically fragmentation and require surgical interven-
compromised (tachycardia, systemic hy- tion. Use of optional IVC filters will likely
potension, right ventricle dysfunction by help in the acute period, since the patient is
echocardiogram or other imaging modality, likely still at risk for recurrence of additional
and/or marked pulmonary hypertension), or emboli.

Acknowledgments 10. Goldhaber SZ. Thrombolysis for pulmonary

The authors would like to acknowledge Ms. embolism. Prog Cardiovas Dis. 1991;34:113-
Leanne Lessley for her expert help in image 34.
illustration. 11. Anderson DR, Levine MN. Thrombolytic
therapy for the treatment of acute pulmonary
embolism. Can Med Assoc J. 1992;146:1317-
references 24.
1. Dalen JE, Alpert JS. Natural history of 12. Konstantinides S, Geibel A, Heusel G,
pulmonary embolism. Prog Cardiovas Dis. Heinrich F, Kasper W. Management Strategies
1975;17:259-70. and Prognosis of Pulmonary Embolism-3
2. Hermann RE, Davis JH, Holden WD. Trial Investigators. Heparin plus alteplase
Pulmonary embolism: a clinical and compared with heparin alone in patients with
pathologic study with emphasis on the effect submassive pulmonary embolism. N Engl J
of prophylactic therapy with anticoagulants. Med. 2002;347:1143-50.
Am J Surg. 1961;102:19-28. 13. Terrin M, Goldhaber SZ, Thompson B. The
3. Barritt DW, Jordan SE. Anticoagulant drugs TIPE Investigators. Selection of patients with
in the treatment of pulmonary embolism: a acute pulmonary embolism for thrombolytic
controlled trial. Lancet. 1960;1:1309-12. therapy: the thrombolysis in pulmonary
4. Benotti JR, Dalen JE. The natural history embolism (TIPE) patient survey. Chest.
of pulmonary embolism. Clin Chest Med. 1989;95:279S-81S.
1984;5:403-10. 14. Molina JE, Hunter DW, Yedlicka JW,
5. Kucher N, Rossi E, De Rosa M, Goldhaber Cerra FB. Thrombolytic therapy for post
SZ. Prognostic role of echocardiography operative pulmonary embolism. Am J Surg.
among patients with acute pulmonary 1992;163:375-81.
embolism and a systolic arterial pressure 15. Gonzlez-Juanatey JR, Valds L, Amaro
of 90 mm Hg or higher. Arch Intern Med. A, Iglesias C, Alvarez D, Garca Acua
2005;165:1777-81. JM, et al. Treatment of massive pulmonary
6. Schoepf UJ, Kucher N, Kipfmueller F, thromboembolism with low intrapulmonary
Quiroz R, Costello P, Goldhaber SZ. Right dosages of urokinase: short-term angiographic
ventricular enlargement on chest computed and hemodynamic evolution. Chest.
tomography: a predictor of early death in 1992;102:341-6.
acute pulmonary embolism. Circulation. 16. Verstraete M, Miller GAH, Bounameaux
2004;110:3276-80. H, Charbonnier B, Colle JP, Lecorf G, et al.
7. Kucher N. Catheter embolectomy for acute Intravenous and intrapulmonary recombinant
pulmonary embolism. Chest. 2007;132:657- tissue-type plasminogen activator in the
63. treatment of acute massive pulmonary
8. Uflacker R. Interventional therapy for embolism. Circulation. 1988;77:353-60.
pulmonary embolism. J Vasc Interv Radiol. 17. Kandarpa K. Technical determinants of
2001;12:147-64. success in catheter-directed thrombolysis for
9. National Institutes of Health Consensus peripheral arterial occlusions. J Vasc Interv
Panel. Thrombolytic therapy and thrombosis: Radiol. 1995;6:55S-61S.
a National Institutes of Health consensus 18. Chamsuddin A, Nazzal L, Kang B, Best I,
development conference. Ann Intern Med. Peters G, Panah S, et al. Catheter-directed
1980;93:141-4. thrombolysis with the EndoWave system in

the treatment of acute massive pulmonary massive pulmonary embolism. J Vasc Interv
embolism: a retrospective multicenter case Radiol. 2002;13:163-9.
series. J Vasc Interv Radiol. 2008;19( 3):372-6. 28. Schmitz-Rode T, Janssens U, Schild
19. Meyer G, Tamisier D, Sors H, Stern M, HH, Basche S, Hanrath P, Gnther RW.
Vouh P, Makowski S, et al. Pulmonary Fragmentation of massive pulmonary
embolectomy: a 20-year experience at one embolism using a pigtail rotation catheter.
center. Ann Thorac Surg. 1991;51:232-6. Chest. 1998;114:1427-36.
20. Leacche M, Unic D, Goldhaber SZ, Rawn JD, 29. Yoshida M, Inoue I, Kawagoe T, Ishihara
Aranki SF, Couper GS, et al. Modern surgical M, Shimatani Y, Kurisu S, et al. Novel
treatment of massive pulmonary embolism: percutaneous catheter thrombectomy in acute
results in 47 consecutive patients after rapid massive pulmonary embolism: rotational
diagnosis and aggressive surgical approach. J bidirectional thrombectomy (ROBOT).
Thorac Cardiovasc Surg. 2005;129(5):1018-23. Catheter Cardiovasc Interv. 2006;68:112-7.
21. Greenfield LJ, Proctor MC, Williams DM, 30. Lang EV, Barnhart WH, Walton DL, Raab
Wakefield TW. Long-term experience with SS. Percutaneous pulmonary thrombectomy. J
transvenous catheter pulmonary embolectomy. Vasc Interv Radiol. 1997;8:427-32.
J Vasc Surg. 1993;18:450-8. 31. Tajima H, Murata S, Kumazaki T, Nakazawa
22. Uflacker R, Strange C, Vujic B. Massive K, Kawamata H, Fukunaga T, et al. Manual
pulmonary embolism: preliminary results of aspiration thrombectomy with a standard
treatment with the Amplatz thrombectomy PTCA guiding catheter for treatment of acute
device. J Vasc Interv Radiol. 1996;7:519-28. massive pulmonary thromboembolism. Radiat
23. Mller-Hlsbeck S, Brossmann J, Jahnke Med. 2004;22:168-72.
T, Grimm J, Reuter M, Bewig B, et al. 32. Timsit JF, Reynaud P, Meyer G, Sors H.
Mechanical thrombectomy of major and Pulmonary embolectomy by catheter device
massive pulmonary embolism with use of the in massive pulmonary embolism. Chest.
Amplatz thrombectomy device. Invest Radiol. 1991;100:655-8.
2001;36:317-22. 33. Siablis D, Karnabatidis D, Katsanos K,
24. Rocek M, Peregrin J, Velimsky T. Mechanical Kagadis GC, Zabakis P, Hahalis G. AngioJet
thrombectomy of massive pulmonary rheolytic thrombectomy versus local
embolism using an Arrow Trerotola intrapulmonary thrombolysis in massive
percutaneous thrombolytic device. Eur Radiol. pulmonary embolism: a retrospective
1998;8:1683-5. data analysis. J Endovasc Ther. 2005;12:
25. Fava M, Loyola S, Flores P, Huete I. 206-14.
Mechanical fragmentation and pharmacologic 34. Zeni PT, Blank BG, Peeler DW. Use of
thrombolysis in massive pulmonary embolism. rheolytic thrombectomy in treatment of acute
J Vasc Interv Radiol. 1997;8:261-6. massive pulmonary embolism. J Vasc Interv
26. Schmitz-Rode T, Janssens U, Hanrath P, Radiol. 2003;14:1511-5.
Gnther RW. Fragmentation of massive 35. Chauhan MS, Kawamura A. Percutaneous
pulmonary embolism using a pigtail rotation rheolytic thrombectomy for large pulmonary
catheter: possible complication. Eur Radiol. embolism: a promising treatment option.
2001;11:2047-9. Catheter Cardiovasc Interv. 2007;70:
27. De Gregorio MA, Gimeno MJ, Mainar 123-30.
A, Herrera M, Tobio R, Alfonso R, et al. 36. Chechi T, Vecchio S, Spaziani G, Giuliani
Mechanical and enzymatic thrombolysis for G, Giannotti F, Arcangeli C, et al. Rheolytic

thrombectomy in patients with massive and 39. Skaf E, Beemath A, Siddiqui T, Janjua
submassive acute pulmonary embolism. M, Patel NR, Stein PD. Catheter-tip
Catheter Cardiovasc Interv. 2009;73:506-13. embolectomy in the management of acute
37. Dwarka D, Schwartz SA, Smyth SH, massive pulmonary embolism. Am J Cardiol.
OBrien MJ. Bradyarrhythmias during use 2007;99:415-20.
of the AngioJet system. J Vasc Interv Radiol. 40. Kuo WT, van den Bosch MA, Hofmann LV,
2006;17:1693-5. Louie JD, Kothary N, Sze DY. Catheter-
38. Eid-Lidt G, Gaspar J, Sandoval J. Combined directed embolectomy, fragmentation, and
clot fragmentation and aspiration in patients thrombolysis for the treatment of massive
with acute pulmonary embolism. Chest. pulmonary embolism after failure of systemic
2008;134:45-60. thrombolysis. Chest. 2008;134;250-4.

2
PA R T
Thoracic Outlet
Syndrome
V enous thoracic outlet re-

mains a niche practice in contemporary vascular surgery, and
management in this area generates some controversy. While all
agree that an upper extremity DVT requires anticoagulation in ac-
cordance with the current American College of Chest Physicians
(ACCP) criteria, the choice of patient who requires decompression
is practice dependent. Drs. Thompson and Johansen have kindly
provided a balanced and contemporary overview of the subject
and demonstrate the consensus and the differences in the patterns
of care, from resolution of the immediate problem with thrombolysis
to pharmacological or surgical intervention. There is no level I evi-
dence to support either the use of thrombolysis or decompression
similar to that used for the treatment of the lower extremity DVT.
one point that should be emphasized is that due to the anatomy,
endovascular interventions (angioplasty or stent placement) without
decompression have no role in the management of primary axil-
losubclavian DVT.
153

CHAPTER
13
Primary Axillosubclavian
Venous Thrombosis
observational Care
Kaj H. Johansen
T hrombosis of the deep veins of the

upper extremities may occur for diverse
reasons. Most commonly, axillosubclavian
ally young and active, develop this problem
after extensive or untoward physical exertion
with the affected arm.
venous thrombosis (ASVT) is secondary to In the lower extremities, deep venous
a pacemaker wire or a chronic indwelling thrombosis (DVT) commonly is followed by
catheter inserted for hemodialysis access or the development of chronic swelling, pain,
for the long-term administration of antibiot- prominent veins, and skin changes, all mani-
ics, chemotherapeutic agents, or blood prod- festations of the postphlebitic or postthrom-
ucts.1-3 botic syndrome (PTS). The condition can
Also presenting, like those with second- also follow untreated upper extremity DVT.
ary ASVT, with a swollen, discolored, some- To varying degrees, PTS can be vexing, dis-
times painful arm, are individuals who have figuring, or even disabling whether it occurs
suffered a spontaneous, or primary, ASVT. in the lower or the upper extremities.
Alternatively known as Paget-Schroetter syn- Because patients with primary ASVT
drome, the condition is also termed effort have been thought to be at substantial risk
thrombosis as a consequence of the observa- for chronic postthrombotic symptoms, they
tion that a number of such individuals, usu- are advised to undergo thoracic outlet de-
compression. If the axillosubclavian vein
is partially or completely occluded, various
venous reconstructive procedures are rec-
Davies md and Alan B. Lumsden md, eds. ommended so that such patients can enjoy a
Cardiotext Publishing, ISBN 978-1-935395-22-5. durable return of upper extremity function.
155

Anatomy and after presentation and catheter-directed ve-

nous thrombolysis is commonly instituted,
Pathophysiology of arm swelling rapidly returns to normal in
a large majority of patients with primary
Primary ASVT ASVT.
Upper extremity deep venous thrombosis

not related to a prior indwelling catheter or General Principles
electrode wire is generally thought to arise
consequent to extrinsic compression of the Only when the natural history of a disease
subclavian vein as it traverses the costoclavic- state is well understood can the advantages
ular space. It has been surmised that chronic and disadvantages of various therapies be
trauma to the subclavian vein at this site reevaluated. For example, for many decades
sults in a neointimal hyperplastic stenosis peptic ulcer disease was thought to be the
within the vein that, over time, narrows to consequence of a metabolic disorder lead-
the point that partial or complete thrombosis ing to gastric acid hypersecretion. It was
of the vein occurs, resulting in the character- assumed that if the condition was left un-
istic venous congestive symptoms and signs treated, recurrent foregut ulceration, per-
that characterize the clinical presentation of foration, bleeding, and obstruction were
primary ASVT. Substantial venous collater- certain and would be treatable only by
als then develop around the shoulder and the major operations that either altered the
clavicle, some of them becoming visible in neurophysiology of gastric acid secretion or
the subcutaneous tissues of the upper arm, rearranged the anatomy of the esophagus,
shoulder, and chest wall. Thrombus gener- stomach, and duodenum. Entire academic
ally extends throughout the subclavian vein careers rose and fell on whether a vagotomy
and may extend in a retrograde fashion out should be truncal or selective, and a gen-
into the axillary and even upper brachial and eration of medical students and residents
basilic veins: the cephalic vein commonly strained to differentiate among the different
remains patent and may be a major outflow postgastrectomy syndromes.
collateral for the congested upper extremity But because of the epochal discovery6
venous system. that peptic ulcers are almost always the result
Over time, as with deep vein clots else- of a bacterial colonization of the stomach,
where, the axillosubclavian thrombus may are easily treated by an oral medication, and
reduce in size, partially recanalize, or even are thereafter almost never associated with
lyse completely. Another explanation for a recurrent symptoms, the underlying condi-
reduction in size of the clot burden in the tion has all but disappeared from contem-
axillosubclavian vein is pulmonary emboli- porary surgical consciousness. Complicated
zation.4 peptic ulcer disease, treated properly, is only
Initial arm swelling in ASVT may be uncommonly diagnosed in contemporary
severe, and upon rare occasion, phlegmasia medical practice and is widely understood to
cerulea dolens has been reported.5 Because be a benign and self-limited condition.
of the aforementioned venous collateraliza- What do we truly know about the nat-
tion and clot resolution, as well as the fact ural history of deep venous disease of the
that such patients generally present early upper extremities?

Primary Axillosubclavian Venous Thrombosis 157
The Natural History of formation) show good to excellent results

relatively durable venous patency and resolu-
ASVT: Older Data tion of postthrombotic symptomsin 80%
to 90% of cases.11-15 While complications of
Beginning in the 1960s, numerous papers these major operations may be severe,16,17
regarding the natural history of primary they are relatively uncommon in series re-
ASVT were published.7-9 These series were ported from high-volume centers.11-15
of substantial size and had extended and
relatively complete follow-up. They dem-
onstrated that, once patients presented with Natural History of
upper extremity DVT, their course, in 40%
to 70% of cases, was quite morbid, with re- ASVT: Recent Data
current episodes of arm swelling, pain and
discoloration, and other manifestations of Beginning in the late 1990s, a series of re-
the postthrombotic syndrome. Many of ports began to appear suggesting that pa-
these patients were thought to be disabled. tients with primary ASVT who had been
Given the data, appropriate means to treated at the time of their initial presenta-
mitigate or even relieve patients chronic tion by venous thrombolysis (often associ-
PTS symptoms were contemplated, devel- ated with an effort at venous angioplasty)
oped, and assessed. Because ASVT has been and oral anticoagulation appeared to have a
thought by many to be a consequence of remarkably benign course, notwithstanding
extrinsic compression of the subclavian vein their having undergone no surgical inter-
in the costoclavicular space between the vention at all. These series often arose from
clavicle and the first rib, initial efforts were audits of vascular laboratory experiences
directed toward thoracic outlet decompres- following patients who had presented with
sion, generally by first rib resection or clavic- primary ASVT18 or from vascular medicine
ulectomy.9,10 Patients whose axillosubclavian or thrombosis specialists.19-22 But several sur-
veins had remained patent or only moder- gical series described patients in whom ini-
ately stenotic appeared to do well. tial thrombolysis and anticoagulation had
However, in ASVT patients whose ax- been performed but an operation had been
illosubclavian vein had been left occluded reserved for patients with recurrent or per-
or highly stenotic, simple thoracic outlet de- sistent, unremitting symptoms; a majority
compression was not adequate. In fact, the of these patients appeared to have done well
effort to remove the first rib potentially inter- without operation as well.23-25
rupted periclavicular venous collaterals and It could be readily concluded from
potentially worsening patients postthrom- these more recent studies that patients with
botic symptoms. So efforts to reconstruct primary ASVT treated by thrombolysis and
the blocked vein by thrombectomy, venous anticoagulation at the time of their initial
patch angioplasty, or bypass grafting began presentation ultimately do well, without de-
to be performed. Several relatively recent velopment of recurrent thrombosis or post-
series suggest these very large operations thrombotic symptoms, in 80% to 90% of
(which continue to include first rib resection cases.
besides axillosubclavian venous reconstruc-
tion and, sometimes, peripheral AV fistula

Comment the immediate anticoagulation in these

patientsnot only to prevent thrombus
The proper choice among differing treat- progression but also to eliminate the risk
ments for a particular condition should of pulmonary embolization4has become
revolve around relative safety, efficacy, dura- the standard of practice. Since the early
bility, and when possible, cost-effectiveness. 1980s, catheter-directed thrombolysis of the
The choice of therapy also depends upon an upper extremity veins has become routine.27
up-to-date understanding of the natural his- Indeed, when thrombolysis is partially or
tory of the condition. completely successful, the use of other endo-
But physicians understanding of the vascular techniques is commonplace. Some
natural history of a particular condition of these treatments, such as balloon angio-
should also be informed by new evidence. plasty of venous stenosis, may be effective,28
This is particularly true when a conditions whereas others (venous stenting) are not.29
subsequent course has been substantially al- Follow-up of ASVT patients subjected
tered by early treatment. As noted previously, to contemporary initial management sug-
the landscape for surgical management of gests a much more benign course than histor-
peptic ulcer disease has been transformed ical controls. Evidence from multiple series
by the contemporary demonstration that the of such patients undergoing contemporary
condition is predominantly an infectious initial treatment of their primary ASVT19-
disease6 that can be mostly prevented by the 25
suggests that the majority do not develop
administration of oral antimicrobial agents recurrent stenoses or significant ongoing
and antacids.26 evidence for postthrombotic symptoms. In-
Historical analyses of patients who had deed, while the optimal circumstance
suffered from primary deep venous throm- would obviously be complete restoration of
bosis of the upper extremity veins demon- axillosubclavian venous patency by either
strated that, if untreated, a large majority of endogenous or exogenous thrombolysis, the
patients had persistent arm swelling, pain, evidence is that a substantial number of in-
discoloration, and other manifestations of dividuals with persistent venous stenoses or
the postthrombotic syndrome.7-9 It was en- even occlusions collateralize around this
tirely reasonable that a surgical approach to relatively focal obstruction in such robust
diminishing the risk of recurrent thrombo- fashion that they are asymptomatic at rest
sis (by thoracic outlet decompression) and, and even after upper extremity exertion.23-25
later, persistently obstructed venous flow Thus, approximately the same percent-
(by venous reconstruction or bypass) were age of individuals followed by observation
developed and perfected. In recent series, a alone have an excellent, or at least as good
large majority of patients so treated have ap- as, outcome as individuals who undergo first
peared to enjoy durable relief of their prior rib resection and axillosubclavian venous
arm problems and have returned to normal reconstruction. With the presumption that
activities.11-15 these patient populations are similar, the
However, prior historical series evalu- routine recommendation for thoracic outlet
ated the outcomes of primary ASVT patients decompression/upper extremity vein recon-
who had received no initial treatment: no struction operation for patients who have
thrombolysis and sometimes not even con- suffered primary ASVT would seem to over-
sistent anticoagulation.7-10 More recently, reach.

Indeed, the major debate has been outcome following axillosubclavian vein
not whether to carry out such surgical re- decompression and reconstruction with an
construction for primary ASVT or not, but immediate operation vs. a period of observa-
whether to wait for a period for postthrombo- tion. Noting recent data,19-25 a skeptical ob-
sis inflammation to clear rather than operat- server might wonder whether, if observation
ing immediately.30 continued long enough, a substantial num-
It has been proposed that an aggressive ber of such patients would be asymptomatic
operative reconstructive approach to prima- and might not need an operation at all. One
ry ASVT may be particularly appropriate for study comparing outcomes of operations for
certain individuals who might be expected venous and neurogenic thoracic outlet syn-
to be sensitive to even minimal amounts of drome by means of a validated functional
arm swellingyounger individuals or elite tool 32 noted that primary ASVT patients
athletes,31 for example. This assertion may were minimally symptomatic preoperatively
be correct. However, it is impossible to prove and were not improved by operation.33
that a subset of primary ASVT patients exist Optimally, given the fact that this is
for whom aggressive operative reconstruc- a relatively stereotypic patient population
tion should be the norm because of the ab- that presents soon after the onset of upper
sence of a control group of such individuals extremity symptoms, a prospective random-
who have not been reconstructed in this ized trial of immediate operation vs. observa-
fashion. tion should be launched. An excellent venue
Indeed, the dilemma in what to advise for this could be the nascent Consortium
patients who have suffered primary ASVT, on Outcomes Research and Education for
after they have undergone thrombolysis and Thoracic Outlet Syndrome (CORE-TOS),
3 months of anticoagulation, is that no level chaired by Robert Thompson, MD, of St.
I, or even level II, evidence is available to Louis, and Julie Freischlag, MD, of Balti-
permit a truly informed decision by the pa- more. A multicenter enrollment of primary
tient. Instead, if a primary ASVT patient ASVT patients, all of them undergoing ini-
presents to a major academic medical center tial anticoagulation and thrombolysis and
in Baltimore, St. Louis, or Denver, the pa- then randomized either to operation or ob-
tient is likely to undergo thoracic outlet de- servation, should soon yield credible results
compression and, if necessary, major upper regarding whether there is an advantage
extremity venous reconstructive surgery. to operative intervention for such patients.
The same patient appearing at a similar in- Such a randomized control trial is under
stitution in Palo Alto, Seattle, or Vancouver discussion.
will likely simply be urged to exercise vigor-
ously and will be offered an operation only references
if, during long-term follow-up, persistent or 1. Linenburger ML. Catheter-related
recurrent postthrombotic symptoms persist thrombosis: risks, diagnosis and management.
or recuran extremely uncommon event. J Natl Compr Cancer Netw. 2006;4:889-901.
At the very least, the necessity for im- 2. Mandolfo S, Piazza W, Galli F. Central
mediately decompressing and reconstruct- venous catheter and the hemodialysis
ing primary ASVT patients axillosubclavian patient: a difficult symbiosis. J Vasc Access.
veins should be reexamined. There are no 2002;3:64-73.
compelling data that demonstrate improved 3. Rozmus G, Danbert JP, Huang DT, Rosero

S, Hall B, Francis C. Venous thrombosis and 14. Adelman MA, Stone DH, Riles TS,
stenosis after implantation of pacemakers and Lamparello PJ, Giangola G, Rosen RJ. A
defibrillators. J Interv Card Electrophysiol. multidisciplinary approach to the treatment
2005;13:9-19. of Paget-Schroetter syndrome. Ann Vasc Surg.
4. Monreal M, Lafoz E, Ruiz J, Valls R, 1997;11:149-54.
Alastrue A. Upper-extremity deep venous 15. Molina JE. Need for emergency treatment in
thrombosis and pulmonary embolism. Chest. subclavian vein effort thrombosis. J Am Coll
1991;99:280-3. Surg. 1995;181:414-20.
5. Kammen BF, Soulen MC. Phlegmasia 16. Melliere D, Becquemin JP, Etienne G,
cerulean dolens of the upper extremity. J Vasc Lecheviller B. Severe injuries resulting from
Interv Radiol. 1995;6:283-6. operations for thoracic outlet syndrome: can
6. Marshall BJ, Warren JR. Unidentified curved they be avoided? J Cardiovasc Surg (Torino).
bacilli in the stomach of patients with gastritis 1991;32:599-603.
and peptic ulceration. Lancet. 1984;1: 17. Chang DC, Lidor AO, Matsen SL, Freischlag
1311-5. JA. Reported in-hospital complications
7. Tilney NL, Griffiths HJG, Edwards EA. following rib resections for neurogenic
Natural history of major venous thrombosis of thoracic outlet syndrome. Ann Vasc Surg.
the upper extremity. Arch Surg. 1970;101: 2007;21:564-70.
792-6. 18. Hingorani A, Ascher E, Lorenson E, DePippo
8. Gloviczki P, Kazmier FS, Hollier LH. P, Salles-Cunha S, Scheinman M, et al.
Axillary-subclavian venous occlusion: the Upper extremity deep venous thrombosis and
morbidity of a non-lethal disease. J Vasc Surg. its impact on morbidity and mortality rates
1986;4:333-7. in a hospital-based population. J Vasc Surg.
9. Kunkel JM, Machleder HI. Treatment of 1997;26:853-60.
Paget-Schroetter syndrome. Arch Surg. 19. Heron E, Lozinguez O, Emmerich J, Laurian
1989;124:1153-8. C, Fiessinger JN. Long-term sequelae of
10. Green RM, Waldman D, Ouriel K, Riggs P, spontaneous axillary-subclavian venous
Deweese JA. Claviculectomy for subclavian thrombosis. Ann Intern Med. 1999;131:510-3.
venous repair: long-term functional results. 20. Lechner D, Wiener C, Weltermann
J Vasc Surg. 2000;32:315-21. A, Eischer L, Eichinger S, Kyrle PA.
11. Urschel HC, Razzuk MA. Improved Comparison between idiopathic deep vein
management of the Paget-Schroetter thrombosis of the upper and lower extremity
syndrome secondary to thoracic outlet regarding risk factors and recurrence. J
compression. Ann Thorac Surg. 1991;52: Thromb Haemost. 2008;6:
1217-21. 1269-74.
12. Machleder HI. Evaluation of a new treatment 21. Martinelli I, Battaglioli T, Bucciarelli
strategy for Paget-Schroetter syndrome: P, Passamonti SM, Mannucci PM. Risk
spontaneous thrombosis of the axillary- factors and recurrence rate of primary deep
subclavian vein. J Vasc Surg. 1993;17:305-17. vein thrombosis of the upper extremities.
13. Thompson RW, Schneider PA, Nelken NA, Circulation. 2004;110:566-70.
Skioldebrand CG, Stoney RJ. Circumferential 22. Elman EE, Kahn SR. The post-thrombotic
venolysis and paraclavicular thoracic outlet syndrome after upper extremity deep venous
decompression for effort thrombosis of the thrombosis in adults: a systematic review.
subclavian vein. J Vasc Surg. 1992;16:723-32. Thromb Res. 2006;117:609-14.

23. Lee WA, Hill BB, Harris EJ Jr, Semba CP, a Palmaz stent in the subclavian vein. AJR
Olcott C IV. Surgical intervention is not Am J Roentgenol. 1993;160:1123-4.
required for all patients with subclavian vein 30. Capparelli DJ, Freischlag J. A unified
thrombosis. J Vasc Surg. 2000;32:57-67. approach to axillosubclavian venous
24. Lokanathan R, Salvian AJ, Chen JC, Morris thrombosis in a single hospital admission.
C, Taylor DC, Hsiang YN. Outcome after Semin Vasc Surg. 2005;18:153-7.
thrombolysis and selective thoracic outlet 31. Melby SJ, Vedantham S, Narra VR, Paletta
decompression for primary axillary vein GA Jr, Khoo-Summers L, Driskill M, et al.
thrombosis. J Vasc Surg. 2001;33:783-8. Comprehensive surgical management of
25. Johansen K. Does axillosubclavian (effort) the competitive athlete with effort thrombosis
thrombosis oblige first-rib resection? Arch. of the subclavian vein (Paget-Schroetter
Surg., in press. syndrome). J Vasc Surg. 2008;47:809-20.
26. Chan FKL, Leung WK. Peptic ulcer disease. 32. Beaton DE, Wright JG, Katz JG. The Upper
Lancet. 2002;360:933-41. Extremity Collaborative Group. Development
27. Taylor LM, McAllister WR, Dennis DL, of the QuickDASH: comparison of three
Porter JM. Thrombolytic therapy followed by item-reduction approaches. J Bone Joint Surg.
first rib resection for spontaneous (effort) 2005;87A:1038-46.
subclavian vein thrombosis. Am J Surg. 33. Cordobes-Gual J, Lozano-Vilardell P,
1985;149:644-7. Torreguitart-Mirada N, Lara-Hernandez
28. Glanz S, Gordon DH, Lipkowitz GS, R, Riera-Vazquez R, Julia-Montoya J.
Butt KM, Hong J, Sclafani SJ. Axillary Prospective study of the functional recovery
and subclavian vein stenosis: percutaneous after surgery for thoracic outlet syndrome.
angioplasty. Radiology. 1988;168:371-3. Eur J Vasc Endovasc Surg. 2008;35:
29. Bjarnason H, Hunter DW, Crain MR, Ferral 79-83.
H, Miltz-Miller SE, Wegryn SA. Collapse of

CHAPTER
14
Axillary-Subclavian
Venous effort Thrombosis
Surgical Care
Valerie B. emery and robert W. Thompson
P rimary axillary-subclavian venous ef-

fort thrombosis, also known as the Pag-
et-Schroetter syndrome, is a relatively rare
Pathophysiology
Patients with venous TOS typically present
condition that affects young, active, other- with the axillary-subclavian venous effort
wise healthy individuals.1 Effort thrombosis thrombosis syndrome, characterized by the
is caused by compression and repetitive inju- abrupt, spontaneous swelling of the entire
ry of the subclavian vein between the first rib arm, often with cyanotic (bluish) discol-
and overlying clavicle, and it is therefore con- oration, heaviness, and pain.2 Rather than
sidered a form of thoracic outlet syndrome being caused by an underlying coagulation
(TOS). Axillary-subclavian venous effort disorder, the pathogenesis of effort thrombo-
thrombosis is distinct from other forms of sis involves extrinsic compression of the sub-
deep venous thrombosis (DVT) with respect clavian vein between the clavicle and first
to pathophysiology, clinical presentation, rib, particularly during activities involving
functional consequences, and treatment. In arm elevation or exertion (Figure 14.1). With
this chapter we review the management of repetition over many months, this focal type
axillary-subclavian venous effort thrombosis, of venous injury leads to progressive fibrous
with an emphasis on current approaches to stenosis of the vein at the level of the first rib,
surgical treatment. involving scar tissue formation and contrac-
tion around the outside of the vein, as well
as fibrosis and wall thickening within the
wall of the vein itself. This initial phase of
Davies md and Alan B. Lumsden md, eds. venous TOS is usually asymptomatic, due to
Cardiotext Publishing, ISBN 978-1-935395-22-5. the simultaneous expansion of dense collat-
163

Figure 14.1 Pathogenesis of axillary-subclavian venous thrombosis. A. Normal anatomy of the thoracic outlet
illustrating relationship of the internal jugular vein (IJV) and subclavian vein (SCV) to the clavicle and first rib.
B. Vigorous activities requiring overhead positions of the arm are associated with the development of axillary-
subclavian venous effort thrombosis. C. Subclavian vein compression between the clavicle and first rib results in focal
vein wall injury. D. Chronic repetitive compression injury of the subclavian vein leads to formation of circumferential
perivenous scar tissue, which can severely constrict the lumen. E. Thrombus formation within the lumen of the constricted
subclavian vein causes complete obstruction of the subclavian vein, with extension of thrombus to the axillary vein
causing obstruction of collateral veins. F. Symptomatic presentation of axillary-subclavian venous effort thrombosis.
Adapted and redrawn from Melby et al., Comprehensive surgical management of the competitive athlete with effort
thrombosis of the subclavian vein (Paget-Schroetter syndrome). J Vasc Surg 2008; 47:809-820, Supplement A Figure III.

Axillary-Subclavian Venous Effort Thrombosis 165
eral veins. Thrombotic occlusion eventually illary-subclavian venous effort thrombosis is

occurs due to stagnant and turbulent blood apparent from the stereotypical history and
flow in the narrowed segment of the subcla- physical examination findings.
vian vein. Peripheral extension of thrombus Some patients with venous TOS may
into the axillary vein can then result in fur- present with a more protracted history of arm
ther obstruction of critical collateral veins, swelling, fatigue, heaviness, and pain that oc-
resulting in the acute clinical presentation. curs only on an intermittent basis, especially
Pulmonary embolism from the subclavian following vigorous use of the arm. Such pa-
vein may also occur (currently estimated to tients may have nonthrombotic positional
occur in approximately 10% of patients with obstruction of the subclavian vein at the level
effort thrombosis), but this is infrequent of the first rib, which has not yet evolved to
compared to deep venous thrombosis in the produce axillary-subclavian venous throm-
lower extremities. bosis. Another subset of patients may present
with a chronic history of upper extremity ve-
nous insufficiency that has caused persistent
Diagnosis or progressive limitations in activity over a pe-
riod of many months to several years. These
individuals are often found to have had pre-
Clinical Presentation
vious axillary-subclavian venous thrombosis
In the absence of an indwelling central ve- that was unrecognized and/or untreated.
nous catheter, any young, healthy, active Finally, up to 20% of patients with venous
individual presenting with the sudden onset TOS describe symptoms of pain, numbness,
of arm swelling and cyanotic discoloration and/or tingling in the hand and fingers that
should be suspected of having axillary- suggest the concomitant presence of neu-
subclavian venous effort thrombosis. This rogenic TOS. This may occur as a result of
condition most frequently occurs in indi- a localized inflammatory response to subcla-
viduals between 15 and 35 years of age, with vian venous thrombosis that has extended
an equal distribution between males and to the surrounding tissues of the scalene
females. Most patients are physically ac- triangle, producing perineural fibrosis simi-
tive, with many engaged in work-related or lar to that seen in patients with neurogenic
recreational activities that involve vigorous TOS.
use of the upper extremities in repetitive
overhead positions and/or heavy lifting. The
magnitude of arm swelling is usually quite Vascular Laboratory Tests
substantial, with the diameter of the affected Upper extremity duplex imaging is most
extremity increased as much as twice that often used as the initial diagnostic study to
of the opposite side. Most patients also de- detect axillary-subclavian venous throm-
scribe fatigue, tightness, heaviness, and pain bosis, as it is noninvasive, inexpensive, and
in the arm, especially with use or overhead readily available. Duplex studies are of value
positioning. Many patients will exhibit vis- if they are positive for axillary-subclavian
ible distention of subcutaneous veins in the vein obstruction, helping to confirm the
upper arm, around the shoulder, or in the clinical diagnosis. However, duplex imaging
upper anterior chest wall. In the vast major- of the subclavian vein is complicated by the
ity of situations, the clinical diagnosis of ax- superimposed clavicle and the depth of the

vein in the neck, and is highly technician- Direct venography is also required to un-
dependent. Expanded collateral veins may dertake catheter-based venous thromboly-
also be mistaken for the subclavian vein, and sis, the preferred initial step in treatment
indirect hemodynamic measures of venous of almost all patients presenting with effort
flow may not accurately reflect the status of thrombosis. Taking all of these factors into
the proximal subclavian vein. Duplex imag- consideration, we believe the most practi-
ing studies have a false-negative rate as high cal, efficient, and cost-effective approach to
as 30% for effort thrombosis and are thereby evaluating the patient with suspected effort
not sufficiently accurate to exclude the diag- thrombosis is to go directly to catheter-based
nosis of venous TOS if negative. venography, rather than utilize duplex stud-
ies or other noninvasive imaging tests.
Radiologic Imaging
In current practice, either contrast-enhanced Blood Coagulation Tests
computed tomography (CT) or magnetic Although considered to be a mechani-
resonance (MR) angiography are being used cal anatomical problem unrelated to an
with greater frequency as the initial nonin- increased propensity toward thrombosis, it
vasive diagnostic studies for axillary-subcla- has been reported that up to 70% of patients
vian venous effort thrombosis. Both of these with venous TOS may have associated ab-
studies are highly accurate in detecting axil- normalities in coagulation tests.3 Since such
lary-subclavian vein occlusion and/or focal abnormalities may influence subsequent
stenosis at the level of the first rib (Figure patient management, a panel of coagula-
14.2A-B). They can also demonstrate the tion studies should be obtained either dur-
presence or absence of enlarged collateral ing the initial diagnostic evaluation or in
veins, may provide information on the age of follow-up, including tests for protein C and
any thrombus present, and can be performed protein S activities, antithrombin III levels,
with the arms in elevated positions as well plasma homocysteine, the presence of anti-
as at rest to elucidate evidence of positional cardiolipin antibodies and lupus anticoagu-
subclavian vein obstruction. Additional in- lant, and mutations in the genes encoding
formation can also be obtained by compari- prothrombin (G20210A), factor V (Leiden,
sons with the contralateral upper extremity. G1692A), plasminogen activator inhibitor-1
Since CT or MR venography provide more (PAI-1, 4G/5G), and methyltetrahydrofolate
anatomic information than venous duplex reductase (MTHFR, C677T). These tests
imaging, these studies can be used to accu- are usually negative and thereby add little to
rately exclude the diagnosis of venous TOS the initial diagnosis or management of effort
when negative. thrombosis.
The most direct and definitive means to
confirm the diagnosis of axillary-subclavian
venous effort thrombosis and venous TOS is Goals of Treatment and
through catheter-directed contrast venogra-
phy. Direct venography provides complete Initial Management
anatomic information regarding the site and
extent of thrombosis, and it allows definitive There are 4 principal goals of treatment for
evaluation of the collateral venous pathways. axillary-subclavian venous effort thrombo-

Figure 14.2 Imaging studies in venous TOS.AandB: Positional magnetic resonance venography. With the arms at
rest (A), there is a patent right subclavian vein and an occluded left subclavian vein. With the arms elevated overhead
(B) , there is moderate stenosis of the right subclavian vein and persistent occlusion of the left subclavian vein. Cand
D : Thrombolytic therapy for left-sided axillary-subclavian venous effort thrombosis. Initial venogram (C) demonstrates
a segmental occlusion of the subclavian vein with dense collaterals. Following thrombolytic therapy (D) , most of the
axillary and subclavian vein has been cleared of thrombus, with a residual focal high-grade stenosis at the level of the
first rib. EandF: Thrombolytic therapy for right-sided axillary-subclavian venous effort thrombosis. Initial venogram
(E) demonstrates a long occlusion of the entire axillary and subclavian veins, with relatively limited development of
collaterals. Following thrombolytic therapy (F) , most of the axillary and subclavian vein has been cleared of thrombus,
with a moderate residual stenosis at the level of the first rib.

sis: (1) provide prompt relief of acute symp- requiring monitoring in an acute-care set-
toms of upper extremity venous congestion ting (eg, intermediate-care or intensive-care
and prevention of pulmonary embolism; (2) unit) for several days. In recent years, it has
reduce the likelihood of recurrent venous become more typical to perform thromboly-
thrombosis following initial management; sis with catheter-based pharmacomechani-
(3) avoid the development of upper extrem- cal thrombectomy, in which a mechanical
ity postthrombotic syndrome; and (4) return device on the tip of the catheter is used to
to normal upper extremity activity without rapidly break up the clot, along with local-
medications. The first of these goals is best ized infusion of a much smaller amount of
met by prompt anticoagulation and throm- thrombolytic agent.5 The great advantage of
bolytic therapy. this approach is that it can usually be com-
pleted in a single stage, often within several
hours, thereby avoiding a long stay in a mon-
Anticoagulation itored hospital setting.
In the absence of any contraindications,
once the diagnosis of axillary-subclavian ve-
nous effort thrombosis is suspected, almost Balloon Angioplasty
all patients should be anticoagulated with The goal of thrombolysis is to clear any fresh
intravenous or subcutaneous heparin. This or recent clot from the axillary-subclavian
can be done while additional diagnostic and collateral veins. This usually results in
studies are being performed and/or prior to a marked improvement in the venographic
patient transfer from one hospital to another, appearance of the vein and a prompt reduc-
and is important to help prevent the exten- tion in symptoms of venous obstruction.
sion of thrombus within the axillary and sub- Following thrombolysis a focal occlusion or
clavian veins. Treatment with an antiplatelet high-grade stenosis of the proximal subcla-
agent, such as aspirin or clopidogrel (Plavix), vian vein is usually identified at the level of
is often included. the first rib; this is not composed of throm-
bus, but represents the underlying scar tis-
sue caused by subclavian vein compression,
Thrombolytic Therapy injury, and tissue repair. In some cases, bal-
Current approaches to venous TOS empha- loon angioplasty may be used at the same
size early diagnosis by contrast venography time in an attempt to reduce the degree of
and prompt use of catheter-based thrombo- stenosis in the subclavian vein. However,
lytic therapy to reduce the amount of throm- because the vein is usually obstructed by
bus within the axillary and subclavian veins scar tissue in the wall of the vein as well as
(Figure 14.2C-F).4 Venous thrombolysis has external compression between the clavicle
traditionally been performed by continuous and first rib, balloon angioplasty is often
infusion of the thrombolytic drug directly unsuccessful, and even when improvement
into a multihole catheter that has been is obtained it is usually short-lived. There is
placed within the axillary-subclavian vein at also abundant evidence demonstrating that
the time of the initial venogram. Infusion is vascular stents should not be placed in the
continued with repeat venograms performed subclavian vein, at least prior to surgical de-
at follow-up intervals for a period of 24 to 48 compression, due to an inevitably high rate
hours, until a maximum effect is achieved, of failure.6 We therefore rarely recommend

the use of balloon angioplasty for subclavian anatomy. It is notable that there is still a sig-
vein stenosis following thrombolysis. Follow- nificant risk of recurrent thrombosis follow-
ing thrombolysis, the patient should remain ing thrombolysis and anticoagulation alone,
on systemic anticoagulation. with published estimates ranging from 50%
to 70%. As summarized by Aziz et al., Med-
ical therapy, consisting either of anticoagu-
Nonsurgical lation or thrombolytic therapy, results in an
Management unsatisfactory clinical outcome because it
does not correct the underlying mechanical
Conservative treatment of axillary-subclavi- abnormality.12
an venous effort thrombosis has traditionally
consisted of chronic anticoagulation, inter-
mittent arm elevation, long-term restrictions Indications and
in arm activity, and the use of compression
sleeves, with the hope that increased collat- Protocols for Surgical
eral development will eventually compen-
sate for axillary-subclavian vein occlusion. Treatment
Many studies in the medical literature have Surgical treatment provides definitive man-
shown that this approach rarely results in agement for axillary-subclavian venous ef-
symptom-free use of the arm and is associat- fort thrombosis and venous TOS, and should
ed with a significant incidence of chronic ve- be considered in almost all patients with this
nous congestion, particularly with active use condition. Operative treatment is centered
of the arm, requiring considerable limita- on 2 goals: (1) decompression of the sub-
tions in young, active patients (Table 14.1).7- clavian vein and collateral venous pathways
11
Unlike lower extremity DVT, the proper through the thoracic outlet by removal of the
duration of anticoagulation treatment for first rib and associated scalene and subcla-
subclavian venous effort thrombosis is not vius muscles, and (2) restoration and main-
known. Because this condition is caused by tenance of normal blood flow through the
compression of the vein rather than a dis- subclavian vein, by removing constricting
order of blood clotting, many recommend scar tissue from around the vein, by adjunc-
lifelong anticoagulation unless there has tive balloon angioplasty, or by direct venous
been a defined alteration in the underlying reconstruction when necessary.
Table 14.1 ReportedOutcomesofConservativeManagement

forAxillary-SubclavianVenousEffortThrombosis
Author Year #Patients Outcomes
Adams et al.7 1965 NA Disabling symptoms >70%
Tilney et al.8 1970 48 Disabling symptoms in 74%
Donayre et al. 9
1986 41 Disabling symptoms >50%
Gloviczki et al. 10
1986 95 Late sequelae in 30%
Lindblad et al.11 1988 73 Moderate disability 25%

The vast majority of patients with recent Thrombolysis Anticoagulation

axillary-subclavian venous effort thrombosis
are excellent candidates for surgical treat- 1 Month Staged Surgery vs.
ment, particularly within the first several
Anticoagulation
weeks of undergoing successful thrombolyt-
ic therapy. However, patients with longstand- Reports from other centers indicate that the
ing untreated subclavian vein occlusion, or interval for reevaluation of patients on anti-
those that exhibit a long segment of residual coagulation can be effectively reduced to 1
venous occlusion extending into the axillary month following thrombolysis.16,17 In an ini-
vein despite thrombolysis, are often consid- tial report of 22 patients following this staged
ered to be unsuitable candidates for surgical approach, Lee et al. found that 13 (59%)
treatment. This judgment depends in large required surgery and 9 (41%) did not, with
part upon the surgical experience available uniformly successful outcomes in each treat-
and the surgical approaches to venous TOS ment group.16 However, in a follow-up report
preferentially used in a particular center. of a larger series (64 patients), 29 (45%) had
surgery within 3 months of thrombolysis and
35 (55%) continued with nonoperative man-
agement. It is notable that 8 (23%) of these
Thrombolysis Anticoagulation
conservatively managed patients also had
3 Months Staged Surgery vs. recurrent thrombosis requiring later surgical
Anticoagulation treatment (overall 58% requiring surgery).17
Over the past 3 decades since the advent

of thrombolytic therapy, protocols for Thrombolysis Anticoagulation
the management of axillary-subclavian
Early Surgery Interval Angioplasty
venous effort thrombosis have evolved
considerably. The first comprehensive ap- With the recognition that the vast majority
proach to this problem was described by of patients with axillary-subclavian venous
Machleder and colleagues at the Univer- effort thrombosis will require surgical treat-
sity of California-Los Angeles, involving ment within 1 to 3 months of initial throm-
a 3-month period of anticoagulation fol- bolytic therapy, others have sought to reduce
lowed by transaxillary first rib resection.13-15 the overall duration of treatment by proceed-
As reported in a series of publications, ap- ing with surgery within days to weeks after
proximately 70% of patients following the thrombolysis. As it has also become clearer
Machleder protocol required surgery for that long-term results for patients undergo-
persistent symptoms after 3 months of an- ing surgery are better for those with a patent
ticoagulation, with up to 30% having had subclavian vein, attempts to further improve
recurrent thrombosis. Clinical outcomes outcomes have coupled transaxillary first rib
were largely influenced by the status of the resection with postoperative balloon angio-
subclavian vein following decompression: plasty at various intervals after operation.18-20
at final assessment, 90% to 95% of those Although excellent results have been report-
with a patent subclavian vein were free of ed with these strategies, the actual outcomes
symptoms, as compared to 64% of those of delayed balloon angioplasty for venous
with an occluded subclavian vein. TOS remain unclear and there may remain

a substantial proportion of patients who have

residual subclavian vein obstruction refrac-
Selection of Surgical
tory to intervention, for which long-term an- Approach
ticoagulation may be recommended.
Transaxillary first rib resection remains one
of the most frequently employed approaches
Thrombolysis Early Surgery with to the treatment of venous TOS. This typi-
cally involves partial resection of the first
Subclavian Vein Reconstruction rib and division of its scalene muscle at-
The most recent step in the evolution of tachments. Because it is not feasible to fully
treatment protocols for axillary-subclavian expose or control the subclavian vein from
venous effort thrombosis has involved wider the transaxillary approach, direct evaluation
use of anterior approaches to thoracic out- and/or reconstruction of the subclavian vein
let decompression and direct reconstruction is not performed. Rather, transaxillary first
of the subclavian vein at the time of opera- rib resection is usually coupled with the sub-
tion.21-28 Molina and colleagues reported sequent use of intraoperative or postopera-
results with immediate surgery using sub- tive venography and performance of balloon
clavicular decompression and subclavian angioplasty and/or stent placement to deal
vein patch angioplasty in 114 patients.26 Of with any residual stenosis in the subclavian
97 (85%) patients treated within 2 weeks of vein.29 Current estimates indicate that 40%
symptoms, the outcomes were uniformly to 50% of patients will demonstrate a residu-
successful. However, of 17 (15%) patients al subclavian vein stenosis requiring balloon
treated more than 2 weeks after the onset angioplasty, even several weeks after first rib
of symptoms, all had developed progres- resection. Because these lesions are typically
sive subclavian vein fibrosis, with 12 (70%) composed of dense scar tissue within and
having postoperative restenosis and 5 (30%) around the wall of the vein, balloon angio-
being considered inoperable. In extend- plasty may be relatively ineffective in this
ing this strategy to offer definitive surgical setting. Although placement of subclavian
treatment to a broader group of patients, our vein stents may be considered in this situ-
group has long advocated use of paracla- ation, the long-term effectiveness of stents
vicular thoracic outlet decompression that in this position is limited. Long-term anti-
involves incisions above and below the clav- coagulation may therefore need to be con-
icle.21,24,28 This approach permits more com- sidered in an effort to reduce the potential
plete first rib resection and more thorough for recurrent venous thrombosis. For these
venous decompression than can be obtained reasons we prefer more direct and thorough
through alternative approaches, as well as approaches to the management of patients
optimal exposure to accomplish direct sub- with venous TOS.
clavian vein reconstruction when neces- The paraclavicular approach combines
sary. This approach also allows completion the advantages of the supraclavicular expo-
of these steps to be accomplished during a sure used for neurogenic and arterial forms
single operative procedure and hospital stay, of TOS with an infraclavicular incision that
with excellent functional outcomes in a permits complete resection of the medial first
large and ongoing clinical series of patients rib, as well as wide exposure of the subclavian
with venous TOS. vein to permit vascular reconstruction. With

this approach we offer operative decompres- The anterior scalene muscle is circum-
sion to virtually all patients with symptom- ferentially dissected at the level of its inser-
atic venous TOS or recent effort thrombosis, tion upon the first rib and sharply divided
regardless of the interval between initial under direct vision (Figure 14.2B). The re-
diagnosis and referral, previous treatment, maining muscle is lifted superiorly and de-
or adverse findings on contrast venography. tached from the underlying tissues, carrying
Operative procedures based on paraclavicu- the dissection superiorly to the level of its
lar exposure thereby provide the most versa- origin on the C6 transverse process, and the
tile, comprehensive, and safe approach to the muscle is divided and removed. Each of the
treatment of venous TOS.27,28 5 nerve roots comprising the brachial plexus
(C5, C6, C7, C8, and T1) are identified, mo-
bilized, and protected from injury, and any
Paraclavicular Thoracic aberrant fibrous bands, ligaments, or fascial
attachments that may contribute to neuro-
Outlet Decompression vascular compression are resected.
With the brachial plexus nerve roots
Thoracic outlet decompression for venous gently retracted in an anteromedial direc-
TOS begins with supraclavicular exposure tion, the attachment of the middle scalene
(Figure 14.3A). The patient is positioned su- muscle is carefully divided from the top of
pine with the head of the bed elevated 30 the first rib (Figure 14.2C). The intercostal
degrees, and the neck, chest, and affected muscles along the posterolateral aspect of
upper extremity are prepped into the field the first rib are divided, and the tip of a right-
to permit movement of the arm during the angle clamp is passed underneath the poste-
operation and access to the forearm and rior neck of the first rib. The posterior aspect
wrist. A transverse supraclavicular incision of the first rib is divided with a modified
is made, subplatysmal flaps are developed to Stille-Giertz rib cutter, and a Kerrison bone
expose the scalene fat pad, and the omohy- rongeur is used to smooth the posterior end
oid muscle is divided. The scalene fat pad of the bone to a level medial to the course
is detached and progressively elevated in a of the T1 nerve root (Figure 14.3D-E). The
medial to lateral direction, exposing the sur- proximal end of the first rib is elevated and
face of the anterior scalene muscle and the a fingertip is passed underneath the rib to
phrenic nerve. Small blood vessels and lym- bluntly separate additional extrapleural fas-
phatics are secured between ligatures and, if cia and intercostal muscle attachments. The
necessary, the thoracic duct is ligated and di- anterior portion of the first rib is not yet di-
vided. Lateral mobilization of the scalene fat vided at this stage.
pad continues until there is ample exposure To accomplish complete resection of
of the anterior scalene muscle and phrenic the anteromedial portion of the first rib, a
nerve, the brachial plexus nerve roots, and second transverse skin incision is made one
the middle scalene muscle. The long tho- fingerbreadth below the medial clavicle
racic nerve is observed emerging from the (Figure 14.3F). A plane of separation is cre-
middle scalene muscle and crossing the pos- ated between the upper and middle portions
terolateral aspect of the first rib. The scalene of the pectoralis major muscle, and the carti-
fat pad is held in position with several retrac- laginous portion of the first rib is identified.
tion sutures. This is facilitated by applying downward

INFRACLAVICULAR
INCISION
Figure 14.3 Paraclavicular thoracic outlet decompression.A. Paraclavicular decompression begins with a
supraclavicular incision. B. The anterior scalene muscle is divided from the first rib and removed. C. The middle
scalene muscle is divided from the first rib and removed. DandE. The posterior aspect of the first rib is divided (D)
with protection of the C8 and T1 brachial plexus nerve roots under direct vision, and the remaining edge of the rib is
trimmed to a level proximal to the nerve roots (E). The anterior first rib is not yet divided at this stage of the procedure,
as it is in operations for neurogenic or arterial TOS. F. An infraclavicular incision is made overlying the anteromedial
aspect of the first rib and the rib is exposed, facilitated by pressure on the posterior end of the rib to separate the
costoclavicular space. G. The anteromedial aspect of the first rib is divided at the edge of the sternum and the entire
first rib is removed, allowing complete dissection of the axillary-subclavian vein throughout the thoracic outlet, to its
junction with the internal jugular and innominate veins. Adapted and redrawn from Thompson RW, Venous Thoracic
Outlet Syndrome: Paraclavicular Approach. Operative Techniques in General Surgery 2008; 10:113-121 (Figures 1-7).

pressure to the divided posterior segment of constricted segment of the vein, and if the
the first rib with a finger placed within the vein is soft and easily compressible to pal-
supraclavicular incision, placing the attach- pation, and shows evidence of rapid filling
ments between the medial first rib and clav- and emptying with respiratory variation, it
icle under tension and allowing the medial is likely that no further venous reconstruc-
portion of the first rib to be dissected from tion is necessary (in our experience, this is
its soft tissue attachments through the infra- the case in approximately 50% of patients
clavicular incision. with venous TOS, even in those with long-
The subclavius muscle tendon, the cos- segment stenosis prior to operation). When
toclavicular ligament, and the muscles of external venolysis does not alleviate subclavi-
the first intercostal space are divided under an vein obstruction, or when intraoperative
direct vision and the cartilaginous portion venography demonstrates a residual stenosis
of the first rib is divided adjacent to the ster- despite the apparent success of external ve-
num, with the first rib removed from the nolysis, additional venous reconstruction is
operative field as a single specimen (Figure performed. Following systemic anticoagula-
14.3G). The axillary-subclavian vein is iden- tion (Dextran and heparin), clamp control
tified underneath the clavicle and carefully is obtained of the distal subclavian and in-
separated from the subclavius muscle, with ternal jugular veins and a pediatric Satinsky
ligation and division of any collateral vein clamp is passed around the upper portion of
branches that enter the subclavian vein, and the innominate vein. A longitudinal venoto-
the subclavius muscle is resected. Further my is created along the superior aspect of the
exposure of the subclavian vein is undertak- subclavian vein and the lumen is thoroughly
en through the supraclavicular exposure and inspected. If the luminal surface is smooth
continued medially toward the junction of and free of thrombus and/or irregularity, a
the subclavian and internal jugular veins to vein patch angioplasty is performed using
form the innominate vein. The internal jug- greater saphenous vein or a segment of cryo-
ular vein is fully exposed several centimeters preserved femoral vein (Figure 14.4B). It is
superior to its junction with the subclavian considered important to construct the patch
vein, and the innominate vein is exposed for to span the entire length of the affected sub-
several centimeters into the upper mediasti- clavian vein, with an extension into the lat-
num. The course of the phrenic nerve into eral aspect of the internal jugular vein or the
the upper mediastinum is also noted, and anteromedial aspect of the innominate vein.
the nerve is protected where it passes under- When dense fibrosis remains within
neath the subclavian vein. the wall of the subclavian vein despite ex-
Any pathological changes in the central ternal venolysis, the affected segment of the
portion of the subclavian vein are assessed subclavian vein is replaced by interposition
visually and by digital palpation. As the bypass (Figure 14.4C). The intervening seg-
subclavian vein is typically found to harbor ment of the native subclavian vein is excised
a focal area with fibrous wall thickening re- (Figure 14.4D), and an interposition graft
sulting from chronic repetitive injury, any is constructed using a widely beveled end-
residual scar tissue surrounding the vein is to-end anastomosis to the unaffected distal
completely excised (circumferential exter- axillary-subclavian vein. The proximal anas-
nal venolysis) (Figure 14.4A). This often tomosis is constructed in a wide end-to-side
results in reexpansion of the previously anastomosis, using an extension of the graft

Figure 14.4 Subclavian vein reconstruction in venous TOS.A. Circumferential external venolysis to remove
constricting fibrous scar tissue from around the subclavian vein, which may allow the vein to re-expand to a normal
diameter. B. When subclavian vein reconstruction is necessary and the lumen of the vein is found to be smooth
and free of chronic thrombus, vein patch angioplasty is sufficient for reconstruction. C. When subclavian vein
reconstruction is necessary but the luminal surface is unsuitable for patch angioplasty, replacement of the subclavian
vein is accomplished with an interposition bypass using a saphenous vein panel graft. D. Microscopic cross-sectional
appearance of the chronically occluded subclavian vein in effort thrombosis. The lumen is occluded with chronic
organized thrombus and inflammatory cell infiltrates, and the vein wall is markedly thickened with fibrosis, loss of the
normal elastic architecture, and previous intramural hemorrhage (Verhoeff van Giesen stain). EandF. Venography in
right-sided axillary-subclavian venous effort thrombosis. The initial venogram (E) demonstrates long occlusion of the
axillary and subclavian veins with widespread collaterals. Venogram performed 12 weeks after surgical treatment (F),
which had involved creation of an axillary-innominate subclavian vein interposition bypass reconstruction. Adapted and
redrawn from Melby et al., Comprehensive surgical management of the competitive athlete with effort thrombosis of the
subclavian vein (Paget-Schroetter syndrome). J Vasc Surg 2008; 47:809-820, Supplement A Figure III; and Thompson
RW, Venous Thoracic Outlet Syndrome: Paraclavicular Approach. Operative Techniques in General Surgery 2008;
10:113-121 (Figures 1-7).

onto the lateral aspect of the internal jugu- Pneumothorax or pleural effusion
lar vein or the anteromedial aspect of the Postoperative lymph leak
innominate vein. Because the caliber of the Residual subclavian vein obstruction
saphenous vein is usually too small to match or early postoperative rethrombosis
the subclavian vein, use of the saphenous Postoperative bleeding/wound
vein requires creation of a panel graft with hematoma/excessive anticoagulation
twice the diameter of the native saphenous Late postoperative axillary or
vein. Alternatively, subclavian vein interposi- subclavian vein obstruction or
tion bypass can be performed using a suit-
rethrombosis
able segment of cryopreserved femoral vein.
Finally, intraoperative venography is
used to confirm satisfactory subclavian vein The expected hospital stay is 5 days, with the
reconstruction, typically performed through closed-suction drain removed approximately
the cephalic vein in the distal forearm. Our 7 days after operation. Inpatient physical
operative approach also includes frequent therapy is started the day after operation
construction of a temporary radiocephalic to maintain range of motion, with postop-
arteriovenous (AV) fistula between the end erative rehabilitation then overseen by a
of the distal cephalic vein and the side of the physical therapist with expertise in the man-
radial artery, used as an adjunct to increase agement of TOS, and no restrictions placed
upper extremity venous blood flow during on upper extremity activity beyond 12 weeks
the first several months after operation. This after operation. Therapeutic anticoagulation
is subsequently ligated under local anesthe- (heparin/warfarin plus clopidogrel) is initiat-
sia at 12 weeks after surgical treatment, at ed several days after operation and then dis-
which time a follow-up contrast venogram is continued at 12 weeks. Recovery is typically
also performed (Figure 14.4E-F). complete within 3 months of operation, and
Postoperative care includes ample use a full return to previous levels of function
of pain medications, muscle relaxants, and can usually be expected.
anti-inflammatory agents. The potential
complications of surgery are similar to those
considered in other operations for thoracic references
outlet syndromes, as well as those related to 1. Hughes ESR. Venous obstruction in the upper
venous reconstruction, as follows: extremity (Paget-Schroetters syndrome). Int
Abstr. Surg. 1949;88:89-127.
Subclavian artery injury/intraoperative 2. Sanders RJ. Thoracic Outlet Syndrome:
hemorrhage A Common Sequelae of Neck Injuries.
Subclavian vein injury/intraoperative Philadelphia: J. B. Lippincott Company; 1991.
hemorrhage 3. Cassada DC, Lipscomb AL, Stevens SL,
Brachial plexus nerve injury or Freeman MB, Grandas OH, Goldman MH.
postoperative paresis The importance of thrombophilia in the
Phrenic nerve injury or postoperative treatment of Paget-Schroetter syndrome. Ann
paresis Vasc Surg. 2006;20:596-601.
Long thoracic nerve injury or 4. Rutherford RB. Primary subclavian-axillary
postoperative paresis vein thrombosis: the relative roles of
thrombolysis, percutaneous angioplasty, stents,

and surgery. Semin Vasc Surg. 1998;11: spontaneous thrombosis of the axillary-
91-5. subclavian vein. J Vasc Surg. 1993;17:305-15.
5. Schneider DB, Curry TK, Eichler CM, 16. Lee WA, Hill BB, Harris EJ Jr, Semba CP,
Messina LM, Gordon RL, Kerlan RK. Olcott CI. Surgical intervention is not
Percutaneous mechanical thrombectomy for required for all patients with subclavian vein
the management of venous thoracic outlet thrombosis. J Vasc Surg. 2000;32:57-67.
syndrome. J Endovasc Ther. 2003;10:336-40. 17. Lee JT, Karwowski JK, Harris EJ, Haukoos
6. Urschel HC Jr, Patel AN. Paget-Schroetter JS, Olcott C IV. Long-term thrombotic
syndrome therapy: failure of intravenous recurrence after nonoperative management
stents. Ann Thorac Surg. 2003;75:1693-6. of Paget-Schroetter syndrome. J Vasc Surg.
7. Adams JT, McEvoy RK, DeWeese JA. 2006;43:1236-43.
Primary deep venous thrombosis of the upper 18. Urschel HC Jr, Razzuk MA. Improved
extremity. Arch Surg. 1965;91:29-42. management of the Paget-Schroetter
8. Tilney NL, Griffiths HJG, Edwards E. Natural syndrome secondary to thoracic outlet
history of major venous thrombosis of the compression. Ann Thorac Surg. 1991;52:1217-
upper extremity. Arch Surg. 1970;101:792-5. 21.
9. Donayre CE, White GH, Mehringer SM, 19. Angle N, Gelabert HA, Farooq MM, Ahn SS,
Wilson SE. Pathogenesis determines late Caswell DR, Freischlag JA, et al. Safety and
morbidity of axillosubclavian vein thrombosis. efficacy of early surgical decompression of the
Am J Surg. 1986;152:179-84. thoracic outlet for Paget-Schroetter syndrome.
10. Gloviczki P, Kazmier FJ, Hollier LH. Axillary- Ann Vasc Surg. 2001;15:37-42.
subclavian venous occlusion: the morbidity of 20. Caparrelli DJ, Freischlag J. A unified approach
a nonlethal disease. J Vasc Surg. 1986;4:333-7. to axillosubclavian venous thrombosis in a
11. Lindblad B, Tengborn L, Bergqvist D. Deep single hospital admission. Semin Vasc Surg.
vein thrombosis of the axillary-subclavian 2005;18:153-7.
veins: epidemiologic data, effects of different 21. Thompson RW, Schneider PA, Nelken NA,
types of treatment and late sequelae. Eur J Skioldebrand CG, Stoney RJ. Circumferential
Vasc Surg. 1988;2:161-5. venolysis and paraclavicular thoracic outlet
12. Aziz S, Straehley CJ, Whelan TJ Jr. Effort- decompression for effort thrombosis of the
related axillosubclavian vein thrombosis. A subclavian vein. J Vasc Surg. 1992;16:723-32.
new theory of pathogenesis and a plea for 22. Molina JE. Surgery for effort thrombosis of the
direct surgical intervention. Am J Surg. 1986; subclavian vein. J Thorac Cardiovasc Surg.
152:57-61. 1992;103:341-6.
13. Kunkel JM, Machleder HI. Spontaneous 23. Molina JE. Need for emergency treatment in
subclavain vein thrombosis: a successful subclavian vein effort thrombosis. J Am Coll
combined approach of local thrombolytic Surg. 1995;181:414-20.
therapy followed by first rib resection. Surgery. 24. Thompson RW, Petrinec D, Toursarkissian
1989;106:114. B. Surgical treatment of thoracic outlet
14. Kunkel JM, Machleder HI. Treatment compression syndromes. II. Supraclavicular
of Paget-Schroetter syndrome: a staged, exploration and vascular reconstruction. Ann
multidisciplinary approach. Arch Surg. Vasc Surg. 1997;11:442-51.
1989;124:1153-7. 25. Azakie A, McElhinney DB, Thompson RW,
15. Machleder HI. Evaluation of a new treatment Raven RB, Messina LM, Stoney RJ. Surgical
strategy for Paget-Schroetter syndrome: management of subclavian vein effort

thrombosis secondary to thoracic outlet (Paget-Schroetter syndrome). J Vasc Surg.

compression. J Vasc Surg. 1998;28:777-86. 2008;47:809-20.
26. Molina JE, Hunter DW, Dietz CA. 28. Melby SJ, Thompson RW. Supraclavicular
Paget-Schroetter syndrome treated with (paraclavicular) approach for thoracic outlet
thrombolytics and immediate surgery. J Vasc syndrome. In: Pearce WH, Matsumura JS, Yao
Surg. 2007;45:328-34. JST, editors. Operative Vascular Surgery in
27. Melby SJ, Vedantham S, Narra VR, Paletta the Endovascular Era. Evanston: Greenwood
GA, Jr., Khoo-Summers L, Driskill M, Academic; 2008. p434-45.
Thompson RW. Comprehensive surgical 29. Urschel HC, Jr., Razzuk MA. Paget-Schroetter
management of the competitive athlete with syndrome: what is the best management? Ann
effort thrombosis of the subclavian vein Thorac Surg. 2000;69:1663-8.

3
PA R T
Central Venous
Disease
C entral venous diseases

(DVT and stenosis) are a frequent area of consultation for vascular
specialists as the interest in direct intervention on iliofemoral and
innominate disease is growing. We have sufficient level I evidence
to support anticoagulation in cases of DVT. Interventions for these
categories of acute occlusive disease have become more common-
place due to the rapid resolution of symptoms with thrombolytic
therapy. May-Thurner syndrome is well treated with lysis and stent
placement. Introduction of intravascular ultrasonogrraphy in the ve-
nous system has increased the understanding and incidence of ste-
notic disease in the system. In a different area of vascular surgery,
central venous occlusive disease has developed as the Achilles heel
of dialysis access and is posing a difficult area to treat with either
endovascular or open interventions. no intervention is proving suc-
cessful. renal vein thrombosis has become recognized as a unique
problem in renovascular disease and requires intensive imaging,
good clinical judgment, and therapy. We have only limited case
179

series to guide clinical decision-making in these areas. Mesenteric
venous thrombosis and Budd-Chiari syndrome are distinct and im-
portant gastrointestinal venous diseases.
180

CHAPTER
15
Iliofemoral
Mark g. Davies
I liofemoral deep venous thrombosis (il-

iofemoral DVT) is a distinct entity with
distinct anatomic and functional implica-
included a venous duplex scan to rule out
deep venous thrombosis and an abdominal
computed tomography (CT) scan to rule
tions for the patient. Isolated left lower ex- out pelvic mass.
tremity swelling secondary to left iliac vein
compression was first described by McMur-
rich in 1908, and defined anatomically by
May and Thurner in 1957 and clinically Medical Management
by Cockett and Thomas in 1965. The left
iliac vein is usually located posterior to the and Open Surgical
right iliac artery and can be compressed
between the artery and the fifth lumbar
Thrombectomy
vertebrae. Symptoms include left lower ex- Medical therapy comprising appropriate
tremity edema, pain, varicosities, venous identification by imaging, adequate imme-
stasis changes, and deep venous thrombosis. diate and long-term anticoagulation, and
Historically, the evaluation of these patients mobilization with compression are now
recommended for all patients with DVT,
including iliofemoral DVT. Until recently,
this was the only standard management of
an iliofemoral DVT except in extreme limb
Davies md and Alan B. Lumsden md, eds. salvage situations where open venous throm-
Cardiotext Publishing, ISBN 978-1-935395-22-5. bectomy was considered appropriate. The
181

goal of venous thrombectomy is to remove was 82%, and the assisted patency rate was
thrombus from the iliofemoral segments of 91%, which remained unchanged over a
a lower extremity in patients who are un- mean follow-up of 22 months. There was
suitable for catheter-directed thrombolysis mild reflux with few clinical symptoms of
(CDT) and in danger of limb loss.1 The postthrombotic syndrome.3 In the only pub-
chapter by Bo Eklf clearly illustrates the lished randomized trial, Plate et al. showed
role of surgical venous thrombectomy in the that, although the operated group (venous
modern management of iliofemoral DVT. thrombectomy) had better outcomes than
Surgical extraction of venous thrombi should the group treated with anticoagulation, nei-
be used in the management of patients with ther of the 2 groups did very well in the long
extensive iliofemoral disease in which limb run.4
loss is imminent, such as in phlegmasia
cerulea dolens. In these patients, there are
benefits to avoiding limb loss, preventing Endovascular
pulmonary embolism, and reducing the se-
verity of postthrombotic syndromes. Surgi-
Management
cal thrombectomy should not be performed The management of iliofemoral DVT is un-
in DVT that does not involve the iliofemoral dergoing an evolution with increased recog-
segment, in nonambulatory patients or in nition that early intervention may benefit the
high-risk surgical patients, and is generally well-selected patient. The indications for en-
reserved for young patients. Early mortality dovascular intervention are phlegmasia ceru-
is generally 1% if proper selection criteria lea dolens, acute/subacute inferior vena cava
are applied. Early iliac vein rethrombosis is (IVC) thrombosis, acute iliofemoral DVT,
13%. Cumulative patency is 75% at 4 years. acute femoropopliteal DVT, and subacute/
In the presence of nonadherent clots, the pa- chronic iliofemoral DVT. This evolution is
tency is 92%, while in patients with adherent prompted by accruing evidence of benefit
clots, the patency is 45%.2 In patients with and spurred on by the 2008 ACCP guide-
successful venous thrombectomy, 37% are lines as discussed by Comerota in chapter 5.
symptom-free compared to 18% in patients As described in previous chapters by Henke,
treated conservatively. Those patients with Wakefield, and Bush in this book, the etiol-
no symptoms had significantly lower am- ogy of DVT is multifactorial and must be
bulatory venous pressures, improved venous individualized for each patient. Iliofemoral
emptying, and a better calf pump function. DVT may be due to extrinsic compression,
These physiological changes portend to a wall damage secondary to catheters, and
lower incidence of chronic venous insuffi- transient and permanent hypercoaguable
ciency in the longer term. In a study that ex- states. Iliac vein compression is the most
amined the efficacy of stent placement after probable cause of iliofemoral DVT. One-
infrainguinal loco-regional thrombolysis half to two-thirds of patients with left-sided
and iliac thrombectomy of acute DVT in pa- iliofemoral DVT have intraluminal webs or
tients with May-Thurner syndrome, techni- spurs from chronic extrinsic compression of
cal success defined as complete vein patency the left iliac vein at the crossing point of the
and normal valve function was documented right common iliac artery. Approximately
in all patients. The calculated cumulative 2% to 5% of those with chronic deep ve-
primary patency rate for venous iliac stents nous insufficiency of the left leg may have

Iliofemoral Deep Venous Thrombosis 183
iliac vein compression syndrome. Iliac vein Patients suitable for catheter-directed
compression syndrome occurs when com- thrombolysis include young, active individu-
pression of the common iliac vein is severe als who have an acute (<10 days) deep venous
enough to inhibit the rate of venous outflow. thrombosis and those with isolated infrain-
In its more severe manifestation, iliac vein guinal disease. Additionally, any patient who
compression syndrome is known to cause has signs/symptoms of phlegmasia cerulea
acute iliofemoral DVT. Iliac vein compres- dolens, regardless of their clinical condition
sion syndrome is caused by the combination or age, should be considered for thrombolyt-
of compression and the vibratory pressure of ic therapy. The algorithm for care is shown
the right iliac artery on the iliac vein, which in Figure 15.1. The incidence of major clini-
is pinched between the artery and the pel- cal hemorrhage after fibrinolysis for DVT is
vic bone. In oncology patients requiring between 6% and 30%, a 3-fold increase com-
staging CT scans of the thorax, abdomen, pared to standard heparin therapy. Infusion
and pelvis, there is a prevalence of 6.8% of thrombolysis is associated with increased
unsuspected iliofemoral, 1.2% unsuspected risk of local hematoma formation at the site
common iliac, and 0.3% unsuspected infe- of catheter insertion and a lower risk of dis-
rior vena cava DVT.5 In a series of patients tant bleeding. The National Multicenter
treated medically for iliofemoral DVT, 50% Venous Registry recommends that any acute
of the limbs had a pathological (deep reflux deep venous thrombosis without a prior his-
or obstructive change) finding in the poplite- tory of thrombosis will yield the greatest de-
al segment after a 20-month follow-up. The gree of lysis, which is predictive of long-term
rate of recanalization was high. There was benefit and a significant decrease in the like-
no difference between calf and more proxi- lihood of the development of chronic venous
mal DVTs. Pain (62%), edema (46%), and insufficiency. Endovascular management
pigmentation (35%) were common, and only of iliofemoral DVT due to May-Thurner
27% of the legs with DVT were asymptomat- syndrome in patients with protein C and/
ic. One study suggests that the development or S deficiency is clinically effective in the
of the postthrombotic syndrome begins quite short term.11 The incidence of clinical signs
early. The frequency of the subjective symp- and symptoms of venous insufficiency and
toms is high.6 The introduction of catheter- duplex-scan findings of valvular reflux was
directed thrombolysis, rheolytic catheters, significantly lower in the patients in which
and the combination pharmacomechanical lytic therapy succeeded and patency was
thrombectomy (PMT) were initiated with kept, compared with patients experiencing
the goals of (1) eliminating iliofemoral ve- acute therapeutic failure or rethrombosis (P
nous thrombus, (2) providing unobstructed <.01).12
venous drainage from the affected limb, and Catheter-directed thrombolytic therapy
(3) preventing recurrent thrombosis; these for the treatment of acute extensive iliofem-
approaches appear to be associated with bet- oral DVT due to May-Thurner syndrome is
ter clinical outcome compared with either an effective method for restoring venous pa-
systemic fibrinolysis and standard anticoag- tency, preventing valvular insufficiency, and
ulation (Table 15.1).7-10 The specifics of the providing relief of the acute symptoms.13 In
catheters, techniques, and results are illus- a study by Dake et al., initial technical suc-
trated in chapters by Davies, Meissner, and cess was achieved in 34 of 39 patients (87%).
Arko in this book. The overall patency rate at 1 year was 79%.

Table 15.1. OutcomesofTrialsofIliofemoralDVTInterventions

Author Study Type Patients Drug Anatomic Clinical
Catheter-DirectedTherapy
Semba 1994 Observational 21 UK 85 85
Bjarnason 1997 Prospective Cohort 77 UK 79 79
Verhaeghe 1997 Observational 25 TPA 76 76
Raju 1998 Observational 24 UK 88 -
Mewissen 1999 Prospective 287 UK 83 -
Registry
Patel 2000 Observational 10 UK 100 100
OSullivan 2000 Observational 39 UK 87 -
Kasirajan 2001 Observational 17 UK/TPA/RPA 82 82
Chang 2001 Observational 10 TPA 100 100
Ouriel 2001 Prospective 11 RPA 91 -
Registry
Shortell 2001 Observational 31 UK/tPA 80 -
AbuRahma 2001 Prospective 51 UK/tPA 89 -
Controlled
Castaneda 2002 Prospective Cohort 25 rPA 92 -
Vedantham 2002 Observational 20 UK/tPA/rPA 89 82
Razavi 2002 Prospective Cohort 31 TNK 89 -
Elsharawy 2002 Randomized Trial 35 SK 100 -
Sugimoto 2003 Observational 54 UK/tPA - 85
Grunwald 2004 Observational 74 UK/tPA/rPA 98 -
Vedantham 2004 Observational 18 RPA 100 96
Lin 2006 Observational 46 tPA 64 -
Kim 2006 Observational 23 UK 81 -
Martinez 2008 Observational 21 tPA 60 -
PharmacomechanicalTherapy
Kasirajan 2001 Observational 17 tPA 80 82
Vedantham 2002 Observational 20 UK/rPA/RPA 62 -
Bush 2004 Observational 22 UK/rPA/RPA 65 100
Jackson 2006 Observational 28 TNK 100 80
Lin 2006 Observational 52 tPA 68 -
Kim 2006 Observational 14 UK 84 -
Arko 2007 Observational 14 TNK 80 90
Martinez 2008 Observational 22 tPA 80 -
Rao 2009 Observational 43 tPA 95 93
Shi 2009 Observational 16 UK 89 -
Li 2010 Observational 36 tPA 80 83
EKOS
No large studies
Abbreviations: EKOS = EkoSonic Endovascular System; rPA = reteplase; SK = streptokinase; TNK = tenecteplase;
tPA = tissue plasminogen activator; UK = urokinase. Adapted from Vedantham S et al. J Vasc Interv Radiol. 2009
Jul;20(7 suppl):S227-39.

Symptomatically, 85% of patients were com- was achieved in 30% for medical therapy
pletely or partially improved compared with vs. 78% endovascular therapy.15 Kwak et al.
findings before treatment. There were no demonstrated a technical success rate of
deaths, pulmonary embolus, cerebral hem- 96% (26 of 27 stents) and a clinical success
orrhage, or major bleeding complications.14 rate of 95% (21 of 22 patients). The causes
AbuRahma et al. reported a technical suc- of common iliac vein obstruction were May-
cess in 16 of 18 patients (89%) receiving Thurner syndrome (n = 16), pelvic mass (n
CDT and primary iliofemoral venous pa- = 2), and unknown (n = 4). Overall, the
tency rates at 1, 3, and 5 years of 24%, 18%, 1-year and 2-year primary patency rates were
and 18% and 83%, 69%, and 69% for medi- both 95%, and the 1-year and 2-year second-
cal therapy and endovascular therapy, re- ary patency rates were both 100%.16 From a
spectively. Long-term symptom resolution national registry of patients (n = 473) with
IliofemoralDVT
Asymptomatic Assessmentof Symptomatic

Contraindications
Risk-BenefitAnalysis
Anticoagulation Percutaneous open Thrombectomy

Thrombolysis
PMT eKoS
Stent
Ambulation Anticoagulation Compression Stockings
Figure 15.1 Algorithm for iliofemoral DVT prevention. Patients diagnosed with an iliofemoral DVT should be
stratified into symptomatic and asymptomatic and then assessed for contraindications to thrombolysis. Concomitant
with this, a risk-benefit analysis should also be performed. If in the clinicians opinion declotting needs to be
done and there are no contraindications for lysis, percutaneous lysis is performed with use of simple lyse and
wait, mechanical thrombectomy, pharmacomechancial (PMT), or ultrasound-assisted lysis (EKOS). If the patient
is symptomatic with a significant clot burden and contraindications to thrombolysis, open thrombectomy should
be performed. Venography and correction of underlying lesions with stent placement is recommended. Standard
therapy thereafter is anticoagulation, ambulation, and compression stockings.

symptomatic lower limb DVT, results of that all of those who initially had complete
312 urokinase infusions in 303 limbs of 287 lysis remained patent at 20 months, whereas
patients were analyzed. After thrombolysis, those with only partial lysis had a lower pa-
grade III (complete) lysis was achieved in 96 tency rate.21,22
(31%) infusions; grade II (50%99% lysis) Use of rheolytic mechanical throm-
in 162 (52%); and grade I (< 50% lysis) in bectomy catheters has been shown to be ef-
54 (17%). For acute thrombosis, grade III fective in the treatment of ilifemoral deep
lysis occurred in 34% of cases of acute and venous thrombosis. Additional infusion of
in 19% of cases of chronic DVT (P < .01). thrombolytic agents via the device creates
Major bleeding complications occurred in a novel treatment strategy of pharmacom-
54 (11%) patients, most often at the punc- echanical thrombectomy, which further en-
ture site. At 1 year, the primary patency hances thrombectomy efficacy (Figure 15.2).
rate was 60%. Lysis grade was predictive of In a preliminary experience, pharmacome-
1-year patency rate (grade III, 79%; grade II, chanical catheter-directed iliofemoral DVT
58%; grade I, 32%; P < .001).17 There was thrombectomy with early stent placement
no difference found in physical functioning was safe and effective.23 The Amplatz device
and well-being between the groups before is reported as having removal of thrombus at
the development of deep venous thrombo- 75% to 83% in lower extremity acute DVT
sis. Following treatment, patients receiving within 6 months and a patency of 77%.24,25
catheter-directed thrombolysis reported bet- The Arrow-Trerotola device, when used clini-
ter overall physical functioning, less stigma, cally in addition to thrombolytic therapy and
less health distress, and fewer postthrom- angioplasty with stents, has been reported to
botic symptoms compared to those patients have technical and clinical success in 100%
treated with anticoagulation alone.18 A 10- of patients with a 16-month clinical success
year study by Klbel et al. demonstrated a of 92%. Concerns about valve and intimal
technical success of 100% and a clinical suc- damage, although justified, have not been
cess of 96%. While cumulative patency was reported.26 In a study without adjunctive
89%, clinically 68% of limbs were asymp- preprocedural thrombolytic therapy, Kasira-
tomatic, 18% of limbs were moderately im- jan et al27 reported that half of the patients
proved, and the remainder were unchanged treated with the AngioJet device had 50%
or moderately worse.19 In another study to of their thrombus removed. Patency was re-
assess venous reflux and the obstruction stored in 77% of those patients with 50%
pattern after catheter-directed and systemic thrombus removal. Improved results have
thrombolysis of deep iliofemoral venous been observed in a similar study in which
thrombosis, valvular competence was pre- the AngioJet was used without preprocedural
served in 44% of patients treated with cath- thrombolytics. Sixty-five percent of patients
eter-directed thrombolysis compared with had complete thrombus removal while par-
13% of those treated with systemic throm- tial thrombus removal was observed in the
bolysis (P = 0.049). Reflux in any deep vein remaining 35%.28 The Trellis catheter can
was present in 44% of patients treated by provide segmental and controlled pharma-
catheter-directed lysis compared with 81% comechanical thrombectomy. It is associ-
of patients receiving systemic thrombolysis ated with a greater technical success rate, a
(P = 0.03).20 Bjarnason et al. have reported lower rate of bleeding, and a lower cost than
a 2-year patency of 78% and Ly et al. found that reported for CDT.29,30 The emission of

Figure 15.2 32-year-old patient presenting with symptomatic iliofemoral DVT. She is 6 weeks postpartum.
She underwent transpopliteal venogram in the prone position. The venograms show iliofemoral DVT ( A-
D ). The final result after PMT and 24 hours of EKOS-supplemented tPA infusion, which reveals a stenosis
consistent with May-Thurner syndrome ( EandF ). Final image shows a patent vein and an intact, fully
deployed stent indicated by the arrows ( G ).

ultrasound waves from an infusion catheter DVT treated with thrombolysis and best
delivering lytic agent is an interesting new medical therapy, a significantly greater pro-
adjunct to catheter-directed thrombolysis. A portion of iliofemoral patients (73%) than
recent study evaluated the success of lysis and femoral patients (31%) remained asymptom-
clinical outcomes in patients treated with ul- atic at the end of their follow-up (P <0.025);
trasound (US)-accelerated thrombolysis for 82% of iliofemoral limbs showed partial or
DVT using an EKOS device. Complete lysis complete lysis 4 weeks after diagnosis of clot.
(90%) was seen in 70% cases and overall No significant difference in reflux develop-
lysis (complete plus partial) was seen in 91%. ment was observed between the 2 groups.
No lysis occurred in 5 cases (9%), 4 of which Although the extent of reflux development
were chronic. It appears that the addition of was similar in both groups, iliofemoral pa-
ultrasound reduces total infusion time and tients still showed fewer clinical symptoms
provides a greater incidence of complete after follow-up.38
lysis.31 In another retrospective study, patients In a small randomized trial of local
undergoing CDT showed complete or partial thrombolysis and anticoagulation vs. antico-
thrombus removal in 32 (70%) and 14 (30%) agulation alone in patients with iliofemoral
cases, respectively. When compared to CDT DVT, the patency rate at 6 months was bet-
in this study, PMT provides similar treat- ter in cases treated with thrombolysis (13/18
ment success with reduced ICU and total [72%] vs. 2/17 [12%], P < 0.001). Venous
hospital length of stay and hospital costs.32 reflux was higher in patients treated with
Subclinical thrombus embolization during anticoagulant (7 patients [41%] vs. 2 [11%],
CDT/PMT is a common phenomenon in P = 0.04).39 In an open multicenter, random-
patients with iliofemoral DVT.33 A study by ized, controlled trial, 103 patients (64 men,
Protack et al. demonstrated that when using mean age 52 years) were allocated to CDT
a mixed modality (CDT/PMT) approach, (n = 50) or standard treatment alone (n =
83%, 83%, and 75% of patients were free of 53). After CDT, grade III (complete) lysis
recurrent DVT at 1, 2, and 3 years, respec- was achieved in 24 and grade II (50%90%)
tively. Furthermore, it appears that CDT/ lysis in 20 patients. After 6 months, iliofemo-
PMT without universal prophylactic IVC fil- ral patency was found in 32 patients (64.0%)
ter placement is safe and effective in treating in the CDT group vs. 19 (35.8%) controls,
acute iliofemoral DVT. Selective rather than corresponding to an absolute risk reduction
routine IVC filter placement is a safe and ap- (RR) of 28.2% (95% CI: 9.7%46.7%; P =
propriate approach.34 0.004). Venous obstruction was found in 10
Several reports suggest that long-term patients (20.0%) in the CDT group vs. 26
results of CDT are not satisfactory because (49.1%) controls; absolute RR 29.1% (95%
of the high recurrence rate of DVT, with CI: 20.0%38.0%; P = 0.004). Femoral ve-
stent length more than 6 cm being a poor nous insufficiency did not differ between
prognostic factor35 and only a marginal re- the two groups.40
duction in the incidence and severity of
postthrombotic limb syndromes.36 Further-
more, extent of thrombolysis is a statistically Conclusion
significant factor affecting the freedom of
rethrombosis and chronic change.37 In a ret- CDT reduced clot burden and DVT recur-
rospective study of iliofemoral and femoral rence, and it may prevent the formation of

postthrombotic syndrome. Indications for disease found on oncologic staging CT. Am J

its use include younger individuals with a Roentgenol. 2007;189(1):162-70.
long life expectancy and few comorbidities, 6. Saarinen J, Kallio T, Lehto M, Hiltunen
limb-threatening thromboses, and proximal S, Sisto T. The occurrence of the post-
iliofemoral DVTs. There is a marked lack thrombotic changes after an acute deep
of randomized controlled trials examining venous thrombosis. A prospective two-year
CDT-related mortality and long-term out- follow-up study. J Cardiovasc Surg (Torino).
comes compared to anticoagulation alone. 2000;41(3):441-6.
Published studies on endovascular DVT 7. Comerota AJ, Aldridge SC, Cohen G,
treatments have been limited by nonstan- Ball DS, Pliskin M, White JV. A strategy
dardized reporting, lack of long-term follow- of aggressive regional therapy for acute
up, and use of surrogate outcomes measures. iliofemoral venous thrombosis with
The Society of Interventional Radiology has contemporary venous thrombectomy or
published reporting and quality improve- catheter-directed thrombolysis. J Vasc Surg.
ment guideline documents that should be 1994;20(2):244-54.
followed in any future trials. 41,42 The effec- 8. Comerota AJ, Gravett MH. Iliofemoral venous
tiveness of combined pharmacomechanical thrombosis. J Vasc Surg. 2007;46(5):1065-76.
thrombectomy, although promising, needs 9. Comerota AJ, Paolini D. Treatment of acute
to be investigated further, as does the role iliofemoral deep venous thrombosis: a strategy
of caval filters in preventing DVT-associated of thrombus removal. Eur J Vasc Endovasc
pulmonary emboli.43 Surg. 2007;33(3):351-60; discussion 361-2.
10. Semba CP, Dake MD. Catheter-directed
thrombolysis for iliofemoral venous
references thrombosis. Semin Vasc Surg. 1996;9(1):26-33.
1. Comerota AJ, Gale SS. Technique of 11. Cho YP, Ahn JH, Choi SJ, Han MS, Jang HJ,
contemporary iliofemoral and infrainguinal Kim YH, et al. Endovascular management
venous thrombectomy. J Vasc Surg. of iliofemoral deep venous thrombosis due to
2006;43(1):185-91. iliac vein compression syndrome in patients
2. Elliott G. Thrombolytic therapy for venous with protein C and/or S deficiency. J Korean
thromboembolism. Curr Opin Hematol. Med Sci. 2004;19(5):729-34.
1999;6(5):304-8. 12. Casella IB, Presti C, Aun R, Benabou JE,
3. Husmann MJ, Heller G, Kalka C, Savolainen Puech-Leo P. Late results of catheter-
H, Do DD, Schmidli J, Baumgartner I. directed recombinant tissue plasminogen
Stenting of common iliac vein obstructions activator fibrinolytic therapy of iliofemoral
combined with regional thrombolysis and deep venous thrombosis. Clinics (Sao Paulo).
thrombectomy in acute deep vein thrombosis. 2007;62(1):31-40.
Eur J Vasc Endovasc Surg. 2007;34(1):87-91. 13. Patel NH, Stookey KR, Ketcham DB, Cragg
4. Plate G, Akesson H, Einarsson E, Ohlin P, AH. Endovascular management of acute
Eklof B. Long-term results of venous throm- extensive iliofemoral deep venous thrombosis
bectomy combined with temporary arterio- caused by May-Thurner syndrome. J Vasc
venous fistula. Eur J Vasc Surg. 1990;4:483-9. Interv Radiol. 2000;11(10):1297-302.
5. Cronin CG, Lohan DG, Keane M, Roche 14. OSullivan GJ, Semba CP, Bittner CA, Kee
C, Murphy JM. Prevalence and significance ST, Razavi MK, Sze DY, et al. Endovascular
of asymptomatic venous thromboembolic management of iliac vein compression (May-

Thurner) syndrome. J Vasc Interv Radiol. bekkenvenetrombose. Tidsskr Nor Lgeforen.

2000;11(7):823-36. 2004;124:478-80.
15. AbuRahma AF, Perkins SE, Wulu JT, Ng 23. Vedantham S, Vesely TM, Sicard GA,
HK. Iliofemoral deep vein thrombosis: Brown D, Rubin B, Sanchez LA, et al.
conventional therapy versus lysis and Pharmaco-mechanical thrombolysis and
percutaneous transluminal angioplasty early stent placement for iliofemoral deep
and stenting. Ann Surg. 2001;233(6): vein thrombosis. J Vasc Interv Radiol.
752-60. 2004;15(6):565-74.
16. Kwak HS, Han YM, Lee YS, Jin GY, Chung 24. Gandini R, Maspes F, Sodani G, Masala S,
GH. Stents in common iliac vein obstruction Assegnati G, Simonetti G. Percutaneous ilio-
with acute ipsilateral deep venous thrombosis: caval thrombectomy with the Amplatz device:
early and late results. J Vasc Interv Radiol. preliminary results. Eur Radiol. 1999;9(5):
2005;16(6):815-22. 951-8.
17. Mewissen MW, Seabrook GR, Meissner 25. Delomez M, Beregi JP, Willoteaux S,
MH, Cynamon J, Labropoulos N, Haughton Bauchart JJ, Janne DB, Asseman P, et al.
SH. Catheter-directed thrombolysis for lower Mechanical thrombectomy in patients with
extremity deep venous thrombosis: report of a deep venous thrombosis. Cardiovasc Intervent
national multicenter registry. Radiology. 1999 Radiol. 2001;24(1):42-8.
Apr;211(1):39-49. 26. Lee KH, Han H, Lee KJ, Yoon CS, Kim SH,
18. Comerota AJ. Quality-of-life improvement Won JY, et al. Mechanical thrombectomy
using thrombolytic therapy for iliofemoral of acute iliofemoral deep vein thrombosis
deep venous thrombosis. Rev Cardiovasc Med. with use of an Arrow-Trerotola percutaneous
2002;3 suppl 2:S61-7. thrombectomy device. J Vasc Interv Radiol.
19. Klbel T, Lindh M, Holst J, Uher P, 2006;17(3):487-95.
Eriksson KF, Sonesson B, et al. Extensive 27. Kasirajan K, Gray B, Ouriel K. Percutaneous
acute deep vein thrombosis of the iliocaval AngioJet thrombectomy in the management
segment: midterm results of thrombolysis of extensive deep venous thrombosis. J Vasc
and stent placement. J Vasc Interv Radiol. Interv Radiol. 2001;12(2):179-85.
2007;18(2):243-50. 28. Bush RL, Lin PH, Bates JT, Mureebe L,
20. Laiho MK, Oinonen A, Sugano N, Harjola Zhou W, Lumsden AB. Pharmacomechanical
VP, Lehtola AL, Roth WD, et al. Preservation thrombectomy for treatment of symptomatic
of venous valve function after catheter- lower extremity deep venous thrombosis:
directed and systemic thrombolysis for deep safety and feasibility study. J Vasc Surg.
venous thrombosis. Eur J Vasc Endovasc Surg. 2004;40(5):965-70.
2004;28(4):391-6. 29. Hilleman DE, Razavi MK. Clinical and
21. Bjarnason H, Kruse JR, Asinger DA, economic evaluation of the Trellis-8 infusion
Nazarian GK, Dietz CA, Caldwell MD, catheter for deep vein thrombosis. J Vasc
et al. Iliofemoral deep venous thrombosis: Interv Radiol. 2008;19(3):377-83.
safety and efficacy outcome during 5 years of 30. OSullivan GJ, Lohan DG, Gough N,
catheter-directed thrombolytic therapy. J Vasc Cronin CG, Kee ST. Pharmacomechanical
Interv Radiol. 1997;8:405-18. thrombectomy of acute deep vein thrombosis
22. Ly B, Njaastad AM, Sandbk G, with the Trellis-8 isolated thrombolysis
Solstrand R, Rosales A, Slagsvold CE. catheter. J Vasc Interv Radiol. 2007;18(6):
Kateterbasert trombolytisk behandling af 715-24.

31. Parikh S, Motarjeme A, McNamara T, always prevent post-thrombotic damage. Eur J

Raabe R, Hagspiel K, Benenati JF, et al. Vasc Endovasc Surg. 2008 Dec;36(6):
Ultrasound-accelerated thrombolysis for the 725-30.
treatment of deep vein thrombosis: initial 38. Singh H, Masuda EM. Comparing short-
clinical experience. J Vasc Interv Radiol. term outcomes of femoral-popliteal and
2008;19(4):521-8. iliofemoral deep venous thrombosis: early lysis
32. Lin PH, Zhou W, Dardik A, Mussa F, and development of reflux. Ann Vasc Surg.
Kougias P, Hedayati N, et al. Catheter-direct 2005;19(1):74-9.
thrombolysis versus pharmacomechanical 39. Elsharawy M, Elzayat E. Early results
thrombectomy for treatment of symptomatic of thrombolysis vs anticoagulation in
lower extremity deep venous thrombosis. Am J iliofemoral venous thrombosis. A randomised
Surg. 2006;192(6):782-8. clinical trial. Eur J Vasc Endovasc Surg.
33. Klbel T, Alhadad A, Acosta S, Lindh 2002;24(3):209-14.
M, Ivancev K, Gottster A. Thrombus 40. Enden T, Klw NE, Sandvik L, Slagsvold
embolization into IVC filters during CE, Ghanima W, Hafsahl G, et al; CaVenT
catheter-directed thrombolysis for proximal Study Group. Catheter-directed thrombolysis
deep venous thrombosis. J Endovasc Ther. vs. anticoagulant therapy alone in deep vein
2008;15(5):605-13. thrombosis: results of an open randomized,
34. Protack CD, Bakken AM, Patel N, Saad controlled trial reporting on short-term
WE, Waldman DL, Davies MG. Long-term patency. J Thromb Haemost. 2009;7(8):
outcomes of catheter directed thrombolysis 1268-75.
for lower extremity deep venous thrombosis 41. Vedantham S, Grassi CJ, Ferral H, Patel NH,
without prophylactic inferior vena cava filter Thorpe PE, Antonacci VP, et al; Technology
placement. J Vasc Surg. 2007;45(5):992-7; Assessment Committee of the Society of
discussion 997. Interventional Radiology. Reporting standards
35. Park YJ, Choi JY, Min SK, Lee T, Jung IM, for endovascular treatment of lower extremity
Chung JK, et al. Restoration of patency deep vein thrombosis. J Vasc Interv Radiol.
in iliofemoral deep vein thrombosis with 2009;20(7 suppl):S391-408.
catheter-directed thrombolysis does not 42. Vedantham S, Thorpe PE, Cardella JF, Grassi
always prevent post-thrombotic damage. CJ, Patel NH, Ferral H, et al; CIRSE and SIR
Eur J Vasc Endovasc Surg. 2008;36(6): Standards of Practice Committees. Quality
725-30. improvement guidelines for the treatment of
36. Comerota A, Aldridge S. Thrombolytic lower extremity deep vein thrombosis with
therapy for deep vein thrombosis: a clinical use of endovascular thrombus removal. J Vasc
review. Can J Surg. 1993;36:359-64. Interv Radiol. 2009;20(7 suppl):S227-39.
37. Park YJ, Choi JY, Min SK, Lee T, Jung IM, 43. Gogalniceanu P, Johnston CJ, Khalid U, Holt
Chung JK, et al. Restoration of patency PJ, Hincliffe R, Loftus IM, et al. Indications
in iliofemoral deep vein thrombosis with for thrombolysis in deep venous thrombosis.
catheter-directed thrombolysis does not Eur J Vasc Endovasc Surg. 2009;38(2):192-8.

CHAPTER
16
Management of
Central Venous Stenosis
Michael J. reardon and Mark g. Davies
W ith increasing use of both short-

term and long-term central ve-
nous access systems and with the growing
noses, it is 41% vs. 3.3%; for asymptomatic
jugular vein stenoses, it is 9% vs. 1.6%.1 In
about 20% of the cases, however, the cause
dialysis access population, central venous is benign.
stenosis and/or occlusion are becoming a Etiology of extrathoracic central venous
more frequent event in clinical vascular obstruction is most often anatomic (May-
practice. There is also renewed interest in Thurner and Paget-Schroetter), iatrogenic
Budd-Chiari syndrome, Paget-Schroetter or traumatic (thrombosis or fibrosis related
syndrome, and May-Thurner syndrome. The to indwelling dialysis catheters, pacemakers,
global incidence of subclavian vein stenoses defibrillators, or central venous lines), while
is 15.6%, jugular vein stenoses is 2.7%, and intrathoracic central venous obstruction
femoral vein stenoses is 0% to 3.8%. Asymp- is commonly an extension of bronchopul-
tomatic subclavian vein stenoses, detected monary neoplasm or mediastinal disease.2
by venograms, represent only 23% to 33% Onset of symptoms is often slow and insidi-
of all subclavian vein stenoses. Most reports ous, with tolerance in the early stages being
show a higher incidence of asymptomatic vs. explained by the development of an effec-
symptomatic lesions: for subclavian vein ste- tive collateral circulation. Symptoms usually
regress after medical treatment, sometimes
requiring thrombolysis; however, in 10% of
patients, major functional impairment may
Davies md and Alan B. Lumsden md, eds. require bypass surgery or transluminal an-
Cardiotext Publishing, ISBN 978-1-935395-22-5. gioplasty.3 The surgical treatment of central
193

venous stenosis has been reserved for signifi- malignancy.10 When replacing the superior
cantly symptomatic patients who have an ob- vena cava combined with resection of medi-
struction or stenosis that is not amenable to astinal malignancies, reconstruction of a left
an endovascular approach, who have failed brachiocephalic vein alone results in a sig-
endovascular intervention, or who have ve- nificant rate of occlusion and development
nous resection as part of a planned en bloc of superior vena cava syndrome. Single right
tumor resection.4 brachiocephalic vein reconstruction or bilat-
eral brachiocephalic vein reconstruction in
this setting, and separate reconstruction of
Intrathoracic the veins, is preferable to use of a Y graft.11
Replacement grafts can include PTFE,1 spi-
Malignant Disease ral vein graft,12 or pericardium, although no
data exist to support the superiority of one
In the presence of known malignant disease over the other. Our preference for superior
of the thorax, percutaneous management of vena cava reconstruction has been a self-
complete superior vena cava (SVC) occlu- constructed pericardial tube graft (Figure
sion with thrombolysis and/or clot aspiration 16.1). Vessel involvement by soft tissue sarco-
followed by stent insertion is safe and effec- ma can classified as type I, artery and vein;
tive, giving sustained symptomatic relief.5 type II, artery only; type III, vein only; and
Primary patency in malignant and benign type IV, neither artery nor vein. In patients
cases at 1 year was 64% and 76%, respec- with retroperitoneal soft tissue sarcoma, the
tively. Overall symptom-free survival ranged most common vascular involvement pattern
from 1 to 34 months.6 Resection for malig- was vein only (type III, 64%). The inferior
nant tumor involvement was long considered vena cava (6 ePTFE tube grafts, 3 ePTFE
an absolute contraindication to resection. In- patches, 2 venoplasties), iliac vein (1 ePTFE
creasingly, resection and graft replacement bypass, 1 Dacron bypass, 1 venous patch),
have been used in selected cases in which en and superior mesenteric vein (1 anastomo-
bloc resection could be achieved.7,8 Infiltra- sis, 1 Dacron bypass) were restored in 80%
tion of the SVC due to advanced non small of the patients (n = 16). Morbidity was 36%
cell lung cancer (NSCLC) or thymoma can and mortality was 4%. At a median follow-up
be treated by prosthetic replacement or tan- of 19.3 months, the venous patency rate was
gential resection. It should not be considered 93.8% (primary and secondary).13
as palliative treatment because of the peri-
operative risks. SVC tangential resection
involves fewer surgical problems. However, Intrathoracic
since this procedure is used mostly for N2
NSCLC subjects, patients have a low mean Benign Disease
survival in spite of adjuvant therapy.9 Surgi-
cal reconstruction of the superior vena cava The need for intervention for benign etiolo-
with an ePTFE (expanded polytetrafluo- gies is increasing as the use of indwelling
roethylene) prosthesis provided immediate catheters for dialysis and cardiac therapy
and long-term relief of symptoms of superior such as pacemakers and implantable defi-
vena cava obstruction with a low surgical brillators increases. Endovascular treatment
morbidity, even in patients with unresectable of benign iliocaval occlusive disease is a safe

Management of Central Venous Stenosis 195
A B C
Figure 16.1 Central venous bypass. A.Superior vena cava reconstruction with a self-constructed pericardial
tube graft (arrow).B. Innominate vein tandom stenosis on venography resistant to angioplasty or stent placement.
C. Venogram of a superior vena cava reconstruction with a self-constructed pericardial tube graft.
and efficient minimally invasive technique insertion. No pacemaker function dysfunc-

with good midterm patency rates. Moreover, tion was encountered in several case series.16
it improves cases with obstruction only, as Reconstruction is usually limited to patients
well as cases with associated reflux and ob- who are significantly symptomatic and have
struction. Primary stenting should always failed endovascular attempts at correction.
be performed by using self-expanding stents Reconstruction can be done via mediaster-
deployed under general anesthesia to avoid notomy or upper hemisternotomy. Proximal
lumbar pain. In case of failure, the endovas- access is generally not difficult, and replace-
cular procedure does not preclude further ment of the SVC using the right atrium as
surgical reconstruction.14 Endovascular re- the outflow site is easily accomplished. For
pair is emerging as a first-line treatment for isolated brachiocephalic obstruction, re-
patients with superior vena cava syndrome construction has generally been done with
of benign etiology. Open surgical repair of ringed PTFE. The major technical issue is
benign SVC syndrome is effective, with du- vein exposure and control in the arm and
rable long-term relief from symptoms. Endo- thoracic outlet.
vascular repair is less invasive but is equally
effective in the midterm, albeit at the cost of
multiple secondary interventions. It is an ap-
propriate primary treatment for benign SVC
Infradiaphragmatic
syndrome. Open surgical repair remains an
excellent choice for patients who are not suit-
Disease
ed for endovascular repair or in whom the Occlusive disease of the inferior vena cava
endovascular repair fails.15 Pacemaker wires (IVC) can arise from old thrombotic events,
can result in stenosis of the superior vena fibrosis, and extrinsic compression. Endo-
cava and other central veins. SVC stenting is venous stent therapy has emerged as an ef-
safe and effective in patients who develop the fective, minimally invasive discipline for
SVC obstruction after cardiac pacemaker restoring patency in chronic iliofemoral/

caval vein obstruction or the May-Thurner of 8.3%. The variable prognosis of the vari-
syndrome. Technical success of 90% or ous IVC tumor lesions depends on tumor
greater can be achieved with this approach, entity, stage, resection status, and individual
and a 3-year assisted primary patency ex- risk factors.18 Palliative relief can be achieved
ceeds 80%. Venous reconstructions for il- with covered stents to obtain short-term de-
iofemoral or IVC obstruction offers 3-year compression, while other therapies (radia-
patency rates of 62%.14 The Palma procedure tion or chemotherapy) are considered or the
with autologous saphenous vein has the best natural course of the disease occurs (Figure
long-term patency, whereas long-term suc- 16.2).
cess with ePTFE is more moderate.17 Tumor
lesions of the inferior vena cava can originate
from the vein or can develop by malignant Budd-Chiari Syndrome
tumor infiltration from the surrounding tis-
sue. The resection rate is 83%, with surgical Primary Budd-Chiari syndrome (BCS) is
reconstruction of the IVC achievable in all characterized by a blocked hepatic venous
cases. The perioperative morbidity for such outflow tract at various levels from small
a strategy is 33%, with a hospital mortality hepatic veins to inferior vena cava, resulting
Figure 16.2.Stent grafting of the IVC. A and B demonstrate occlusion of the IVC and the common iliac veins.
Crepresents the patency of the IVC and common iliac veins after lytic therapy.

from thrombosis or its fibrous sequellae.19 of transjugular intrahepatic portosystemic

This is a rare disease that affects mainly shunts (TIPS).
young adults. Multiple risk factors have Disruption of a portal vein thrombus
been identified and are often combined in can also be done during the same session.
the same patient. Myeloproliferative dis- Surgical shunts have been superseded by the
eases of atypical presentation account for use of transjugular intrahepatic portosystem-
nearly 50% of patients; their diagnosis can ic shunts. Liver transplantation is reserved
be made by showing the V617F mutation for fulminant and progressive chronic forms
in Janus tyrosine kinase-2 gene of periph- of BCS. Anticoagulation therapy must be
eral blood granulocytes and, should this used routinely before and after specific ther-
mutation be absent, by showing clusters of apy regardless of whether a thrombophilic
dystrophic megakaryocytes at bone marrow disorder is diagnosed. First-line strategy is
biopsy. Diagnosis can be difficult because anticoagulation, correction of risk factors,
of the wide spectrum of presentation of the diuretics, and prophylaxis for portal hyper-
disease and the varying severity of liver dam- tension. Second-line strategy consists of an-
age. The traditional classification of Budd- gioplasty for short-length venous stenoses21
Chiari syndromeas fulminant, acute, or followed by TIPS. The final strategy is liver
chronicis not prognostically useful. This transplantation. Treatment progression is
makes assessing the benefit of therapy dif- dictated by the response to previous therapy.
ficult, especially because there is no evi- This tiered strategy has achieved 5-year sur-
dence from randomized studies. Portal-vein vival rates approaching 90%. Medium-term
thrombosis occurs in 20% to 30% of cases, prognosis depends on the severity of liver dis-
and acute presentation reflects an acute or ease. Patients with Budd-Chiari syndrome
chronic syndrome in 60% of Budd-Chiari and portal venous system thrombosis consti-
syndrome cases.20 Presentation and mani- tute a unique group with limited therapeutic
festations are extremely varied, so that the options and poor prognosis (median survival
diagnosis must be considered in any patient = 1 month).22 Long-term outcome might be
with acute or chronic liver disease. Dop- jeopardized by transformation of underlying
pler ultrasonography, computed tomogra- conditions and hepatocellular carcinoma.
phy (CT), or magnetic resonance imaging
(MRI) of hepatic veins and inferior vena
cava can demonstrate lesions in the hepatic Renal Vein Thrombosis
veins and/or inferior vena cava.
The disease is considered to be spon- (RVT)
taneously lethal within 3 years of first
symptoms. Budd-Chiari syndrome can be Renal vein occlusion in adults is usually
diagnosed and treated on a single occasion a result of the vein thrombosis, which is
in the setting of the radiology department, frequently associated with the nephrotic
with hepatic venography, transjugular liver syndrome. The anatomy of renal vascular-
biopsy, retrograde CO2 portography, and in- ization is of primary importance in under-
ferior vena cava pressure measurements per- standing its pathophysiological responses
formed simultaneously with therapies such and the clinical and diagnostic presenta-
as dilation or stenting of webs in the inferior tion of patients with this condition.23 The
vena cava or hepatic veins, and placement reaction of the kidney to its vein occlusion

is determined by the balance between the found in the asymptomatic general popula-
acuteness of the disease, extent of the de- tion. However, the clinical syndrome, vari-
velopment of collateral circulation, involve- ously known as May-Thurner syndrome,
ment of one or both kidneys, and the origin Cockett syndrome, or iliac vein compression
of the underlying disease. Renal vein oc- syndrome, is thought to be a relatively rare
clusion is generally a complication of some contributor of chronic venous disease. Iliac
other condition but may also be a primary vein compression syndrome (May-Thurner
disease. Renal vein thrombosis is relatively syndrome) is the most probable cause of il-
rare. CT angiography is considered the in- iofemoral deep venous thrombosis (DVT).
vestigation of choice; alternatives include One-half to two-thirds of patients with left-
MR angiography or renal venography in sided iliofemoral DVT have intraluminal
highly selected patients. webs or spurs from chronic extrinsic com-
As the condition is relatively uncom- pression of the left iliac vein at the crossing
mon, consensus on the best form of therapy point of the right common iliac artery. Ap-
for this condition has been slow to evolve. proximately 2% to 5% of those with chronic
The trend in management has shifted to deep venous insufficiency of the left leg may
nonsurgical therapies, particularly systemic have May-Thurner syndrome. May-Thurner
anticoagulation, except in a highly select- syndrome occurs when compression of the
ed group of patients. The principal mode common iliac vein is severe enough to in-
of treatment includes correction of fluid hibit the rate of venous outflow. In its more
and electrolyte imbalance, dialysis, antihy- severe manifestation, May-Thurner syn-
pertensive drugs, anticoagulation, and in drome is known to cause acute iliofemoral
certain cases, thrombolysis. Percutaneous DVT.
catheter-directed thrombectomy with or May-Thurner syndrome is caused by
without thrombolysis for acute RVT is as- the combination of compression and the vi-
sociated with a rapid improvement in renal bratory pressure of the right iliac artery on
function and low incidence of morbidity. It the iliac vein that is pinched between the
is feasible for native and allograft renal veins artery and the pelvic bone. With the advent
and should be considered in patients with of catheter-directed thrombolytic therapy for
acute RVT, particularly in the setting of de- patients presenting with iliofemoral DVT,
teriorating renal function.24 the underlying cause has been unveiled, and
May-Thurner syndrome is gaining recogni-
tion. Patients presenting with symptoms of
May-Thurner chronic venous insufficiency often fail con-
servative treatment, and because of their
Syndrome crippling symptoms, they may have a high
rate of work absence or are on permanent
The obstruction of the left common iliac disability. If May-Thurner syndrome can be
vein by the pressure of the anteriorly posi- identified as the cause and corrected, pa-
tioned right common iliac artery, with in- tients quality of life would improve. With
timal changes, was first described by May the advent of endovascular stenting, the un-
and Thurner in 1956.25 Nonthrombotic derlying cause can be easily corrected, and
iliac vein lesions, such as webs and spurs de- long-term patency is acceptable26,27 (Figure
scribed by May and Thurner, are commonly 16.3).

Figure 16.3. May-Thurner syndrome. A. Stenosis of the common iliac vein with collaterals. B. An angioplasty
balloon inflated in the area. C. Final result: a patent common iliac vein without collaterals.
Paget-Schroetter the central location of the venous abnormal-

ity (Figures 16.4 and 16.5). The underlying
Syndrome pathophysiology of this disorder is felt to be
repetitive venous trauma owing to arm mo-
Paget-Schroetter syndrome refers to spon- tion in the narrow anatomic space between
taneous thrombosis of the axillary and sub- the clavicle and first rib. The treatment of
clavian venous segments in young, healthy Paget-Schroetter syndrome remains contro-
individuals. It is a rare but potentially dis- versial. Prompt anticoagulation is generally
abling affliction. The diagnosis should be accepted as the minimal treatment offered.
suspected in any young patient presenting In the symptomatic individual, a more ag-
with unilateral arm swelling. Typically, the gressive endovascular treatment is currently
dominant arm is affected, and frequent, re- undertaken, with thrombolysis undertaken
petitive arm use is a common component in the acute situation.28 Many centers pro-
of the patients history. While venous du- ceed with relief of the anatomic compression
plex may make or infer the diagnosis, cross- of the subclavian vein by first rib resection
sectional imaging or contrast venography in all patients, whereas others perform this
is used to confirm the diagnosis because of procedure selectively in cases of persistent

Figure 16.4. Venous thoracic outlet syndrome.

A. Venogram showing an occluded subclavian
vein with multiple collaterals. B. Demonstrates the
result of lysis after a 24-hour infusion; a residual
stenosis remains. C. Final venogram after balloon
angioplasty of a residual stenosis and TOS
decompression.
Figure 16.5. Venous thoracic outlet syndrome. Example of a patent subclavian vein with extrinsic
compression and collateral formation representing TOS.

venous stenosis or ongoing symptoms.29,30 dialysis access sites.39 Thus, patients with a
Angioplasty with or without stenting is dis- reasonable life expectancy or who are unable
couraged without concomitant anatomic to return for subsequent procedures should
decompression but does have an adjunctive be considered for alternative therapy.40 The
role in patients undergoing first rib resection right atrial bypass grafting has been used to
and can avoid the need for jugular turn- restore central venous patency in the careful-
down or patch angioplasty. ly selected patient, in whom all other access
sites are exhausted and in whom percutane-
ous dilation and/or stenting has failed.41
Dialysis Access-Related
Central Venous Stenosis Conclusion
Central venous obstruction is a common Central venous disease remains a small but
problem in patients with chronic renal fail- significant area of venous disease. The data
ure who undergo maintenance hemodialysis. in the literature are composed of small series
Incidences of central vein stenosis reported and case reports. In most cases, treatment
within hemodialysis patients have ranged has not been validated by level I evidence.
from 11% to 40%.31-35 Significant stenosis or There remains significant potential for clini-
occlusion of the subclavian vein is known to cal research and reporting. The practicing
occur in 20% to 50% of patients who have vascular specialist should be cognizant of
had central venous catheters inserted into each of the entities and be prepared to offer
the subclavian vein or the internal jugular the spectrum of medical, radiological, and
vein.36 Stents provide a temporary benefit surgical therapy.
in most patients with central or peripheral
upper extremity stenosis and obstruction.
Incidence of central vein stenosis reported
within hemodialysis patients have ranged references
from 11% to 40%.31-33,37,38 Regular follow-up 1. Huu TC. Central venous stenosis: review of
and reinterventions are required to maintain the literature from 1980 to 2000. Nephrologie.
patency and achieve long-term clinical suc- 2001;22(8):479-85.
cess. Stents used for central venous lesions 2. Schifferdecker B, Shaw JA, Piemonte TC,
have higher clinical success rates than stents Eisenhauer AC. Nonmalignant superior
used for peripheral venous lesions. However, vena cava syndrome: pathophysiology
endovascular therapy for central venous ste- and management. Catheter Cardio Inte.
nosis in the dialysis access patient, whether 2005;65:416-23.
via angioplasty or stenting, is safe, with low 3. Marlier S, Bonal J, Cellarier G, Bouchiat C,
rates of technical failure. Multiple addi- Talard P, Dussarat GV. Superior vena cava
tional interventions are the rule with both syndromes of benign etiology. Presse Med.
treatment modalities. Although neither of- 1996;25(26):1203-7.
fers truly durable outcomes, stent placement 4. Picquet J, Blin V, Dussaussoy C, Jousset Y,
does not improve upon the patency rates Papon X, Enon B. Surgical reconstruction of
provided by balloon angioplasty and does the superior vena cava system: indications and
not add to the longevity of ipsilateral hemo- results. Surgery. 2009;135:93-9.

5. Crowe MT, Davies CH, Gaines PA. 14. Hartung O, Otero A, Boufi M, Decaridi
Percutaneous management of superior G, Barthelemy P, Juhan C, et al. Mid-
vena cava occlusions. Cardiovasc Inter Rad. term results of endovascular treatment for
1995;18(6):367-72. symptomatic chronic nonmalignant iliocaval
6. Barshes NR, Annambhotla S, El Sayed venous occlusive disease. J Vasc Surg.
HF, Huynh TT, Kougias P, Dardik A, et al. 2005;42(6):1138-44.
Percutaneous stenting of superior vena cava 15. Rizvi AZ, Kalra M, Bjarnason H, Bower TC,
syndrome: treatment outcome in patients Schleck C, Gloviczki P. Benign superior vena
with benign and malignant etiology. Vascular. cava syndrome: stenting is now the first line of
2007;15(5):314-21. treatment. J Vasc Surg. 2008;47(2):372-80.
7. Suzuki K, Asamura H, Watanbe S, Tsuchiya 16. Teo N, Sabharwal T, Rowland E, Curry P,
R. Combined resection of superior vena cava Adam A. Treatment of superior vena cava
for lung carcinoma: prognostic significance of obstruction secondary to pacemaker wires with
patterns of superior vena cava invasion. Ann balloon venoplasty and insertion of metallic
Thoracic Surg. 2004;78:1184-9. stents. Eur Heart J. 2002;23(18): 1465-70.
8. Thomas P, Magnan PE, Moulin G, Giudicelli 17. Jost CJ, Gloviczki P, Cherry KJJ, McKusick
R, Fuentes P. Extended operation for the lung MA, Harmsen WS, Jenkins GD, et al. Surgical
cancer invading the superior vena cava. Eur J reconstruction of iliofemoral veins and the
Cardiothorac Surg. 1994;8:177-82. inferior vena cava for nonmalignant occlusive
9. Politi L, Crisci C, Montinaro F, Andreani disease. J Vasc Surg. 2001;33(2):320-8.
M, Podzemny V, Borzellino G. Prosthetic 18. Eder F, Halloul Z, Meyer F, Huth C, Lippert
replacement and tangential resection of H. Surgery of inferior vena cava associated
the superior vena cava in chest tumors. malignant tumor lesions. Vasa. 2008;37(1):68-
J Cardiovasc Surg (Torino). 2007;48(3):363-8. 80.
10. Magnan PE, Thomas P, Giudicelli R, Fuentes 19. Valla DC. Primary Budd-Chiari syndrome.
P, Branchereau A. Surgical reconstruction J Hepatol. 2009;50(1):195-203.
of the superior vena cava. Cardiovasc Surg. 20. Senzolo M, Cholongitas EC, Patch D,
1994;2(5):598-604. Burroughs AK. Update on the classification,
11. Shintani Y, Ohta M, Minami M, Shiono assessment of prognosis and therapy of
H, Hirabayashi H, Inoue M, et al. Long- Budd-Chiari syndrome. Nat Clin Pract Gastr.
term graft patency after replacement of 2005;2(4):182-90.
the brachiocephalic veins combined with 21. Lee BB, Villavicencio L, Kim YW, Do YS,
resection of mediastinal tumors. J Thorac Koh KC, Lim HK, et al. Primary Budd-
Cardiovasc Surg. 2005;129(4):809-12. Chiari syndrome: outcome of endovascular
12. Erbella J, Hess PJ, Huber TS. Superior vena management for suprahepatic venous
cava bypass with superficial femoral vein for obstruction. J Vasc Surg. 2006;43(1):101-8.
benign superior vena cava syndrome. Ann 22. Mahmoud AE, Helmy A, Billingham L, Elias
Vasc Surg. 2006;20:834-8. E. Poor prognosis and limited therapeutic
13. Schwarzbach MH, Hormann Y, Hinz U, options in patients with Budd-Chiari syndrome
Leowardi C, Bckler D, Mechtersheimer and portal venous system thrombosis. Eur
G, et al. Clinical results of surgery for J Gastr Hepat. 1997;9(5):485-9.
retroperitoneal sarcoma with major 23. Asghar M, Ahmed K, Shah SS, Siddique MK,
blood vessel involvement. J Vasc Surg. Dasgupta P, Khan MS. Renal vein thrombosis.
2006;44(1):46-55. Eur J Vasc Endovasc Surg. 2007;34(2):217-23.

24. Kim HS, Fine DM, Atta MG. Catheter- RH, Saeed M, Dennis VW. Hemodialysis-
directed thrombectomy and thrombolysis for associated subclavian vein stenosis. Kidney Int.
acute renal vein thrombosis. J Vasc Interv 1988;33:1156-9.
Radiol. 2006;17(5):815-22. 34. Surratt RS, Picus D, Hicks ME, Darcy MD,
25. May R, Thurner J. Em Gefsssporn in Kleinhoffer M, Jendrisak M. The importance
der v.iliaca corn. sin. als wahrscheinliche of preoperative evaluation of the subclavian
Ursacheder uberwiegend liknsseitigen vein in dialysis access planning. Am J
Beckenvenenthrombose. Z Kreisl-Forsch. Roentgenol. 1991;156:623-5.
1956;45:912. 35. Lumsden AB, MacDonald MJ, Isiklar H,
26. Mussa FF, Peden EK, Zhou W, Lin PH, Martin LG, Kikeri D, Harker LA, et al.
Lumsden AB, Bush RL. Iliac vein stenting for Central venous stenosis in the hemodialysis
chronic venous insufficiency. Tex Heart Inst J. patient: incidence and efficacy of endovascular
2007;34(1):60-6. treatment. Cardiovasc Surg. 1997;5:504-9.
27. Lin PH, Zhou W, Dardik A, Mussa FF, 36. Chandler NM, Mistry BM, Garvin PJ.
Kougias P, Hedayati N, et al. Catheter-direct Surgical bypass for subclavian vein occlusion
thrombolysis versus pharmacomechanical in hemodialysis patients. J Am Coll Surg.
thrombectomy for treatment of symptomatic 2002;194(4):416-21.
lower extremity deep venous thrombosis. Am J 37. Lumsden AB, MacDonald MJ, Isiklar H,
Surg. 2006;192(6):782-8. Martin LG, Kikeri D, Harker LA, et al.
28. Landry GJ, Liem TK. Endovascular Central venous stenosis in the hemodialysis
management of Paget-Schroetter syndrome. patient: incidence and efficacy of endovascular
Vascular. 2007;15(5):290-6. treatment. Cardiovasc Surg. 1997;5:504-9.
29. Molina JE, Hunter DW, Dietz CA. 38. Surratt RS, Picus D, Hicks ME, Darcy MD,
Protocols for Paget-Schroetter syndrome Kleinhoffer M, Jendrisak M. The importance
and late treatment of chronic subclavian of preoperative evaluation of the subclavian
vein obstruction. Ann Thorac Surg. vein in dialysis access planning. Am J
2009;67(2):416-22. Roentgenol. 1991;156:623-5.
30. Molina JE, Hunter DW, Dietz CA. 39. Bakken AM, Protack CD, Saad WEA, Lee
Paget-Schroetter syndrome treated with D, Waldman DL, Davies MG. Longterm
thrombolytics and immediate surgery. J Vasc outcomes of primary angioplasty versus
Surg. 2007;45(2):328-34. primary stenting in the management of central
31. Schumacher KA, Wallner B, Weidenmaier venous stenoses in the hemodialysis patient.
W, Friedrich JM. Venous occlusions J Vasc Surg. 2007;45:776-83.
distant to the shunt as malfunction factors 40. Oderich GS, Treiman GS, Schneider P,
during hemodialysis. Fortschr Rntgenstr. Bhirangi K. Stent placement for treatment of
1989;150:198-201. central and peripheral venous obstruction: a
32. Haage P, Vorwerk D, Piroth W, Schuermann long-term multi-institutional experience.
K, Guenther RW. Treatment of hemodialysis- J Vasc Surg. 2000;32(4):760-9.
related central venous stenosis or occlusion: 41. El-Sabrout RA, Duncan JM. Right atrial
results of primary wallstent placement bypass grafting for central venous obstruction
and follow-up in 50 patients. Radiology. associated with dialysis access: another
1999;212:175-80. treatment option. J Vasc Surg. 1999;29(3):
33. Schwab SJ, Quarles LD, Middleton JP, Cohan 472-8.

CHAPTER
17
Mesenteric Venous Thrombosis

Hosam F. el Sayed
M esenteric venous thrombosis

(MVT) is a distinct clinical entity
that is responsible for 5% to 15% of all cases
mately 0.01% of all emergency surgical ser-
vice admissions. Studies have shown that
the incidence of the disease has not changed
of mesenteric ischemia.1 The low incidence over the past 3 decades; however, there was
of the disease, the nonspecific symptoms, a higher proportion of elderly population in
the insidious onset, and the low awareness recent years.6 There is no sex predilection for
among clinicians may be responsible for the the disease and the median age at presenta-
relatively high mortality rate of 32% to 44% tion is 65 years old, with the highest inci-
reported in recent series.2 The disease was dence in patients between 70 and 79 years of
first described by Elliot in 18953 and was age.6 In a recent report by Rhee et al.,7 MVT
later recognized in 1926 by Cokkinis.4 How- comprised only 6.2% of all patients treated
ever, Warren and Eberhard in 1935 were the for mesenteric ischemia.
first to recognize it as a cause of intestinal The diagnosis has often been delayed,
infarction distinct from mesenteric arterial and most cases were identified either at lapa-
occlusion.5 rotomy or at autopsy.1 The recent improve-
The disease represents 2 per 100,000 ments in imaging techniques have led to
hospital admissions and comprises approxi- both earlier diagnosis and a better idea of the
etiology of mesenteric venous thrombosis,
which has in turn led to significant changes
in the treatment and perhaps reduction of
Davies md and Alan B. Lumsden md, eds. the morbidity and mortality associated with
Cardiotext Publishing, ISBN 978-1-935395-22-5. this clinical syndrome.
205

Etiology creatitis, inflammatory bowel disease, and

intra-abdominal infections.
Mesenteric venous thrombosis can be clas- Local venous congestion or stasis in the
sified as either primary or secondary. When portal venous system is another significant
an etiologic factor is found, the patient is risk factor for MVT. This is most commonly
classified as having secondary mesenteric represented in cases with portal hyperten-
venous thrombosis, and when no obvious sion, especially in cases of liver cirrhosis. It
cause is found, it is said to be a primary is also manifested in cases with severe con-
case. With continued improvements in the gestive heart failure, causing a form of pos-
recognition of various etiologic factors for thepatic portal hypertension. Patients with
mesenteric venous thrombosis, the propor- morbid obesity (BMI >40) are also prone to
tion of primary cases continues to decline. develop MVT.6
Currently, the etiologic factor for mesenteric
venous thrombosis is found in about three-
quarters of patients.1 Many studies have also
shown that a significant portion of patients
with secondary MVT have multiple risk fac- Table 17.1 ConditionsAssociated
tors.1,6,8 withMVT
The conditions associated with MVT DirectInjury
fall into 3 main categories (Table 17.1) 6 : di-
Abdominal trauma
rect injury, local venous congestion or stasis, Postsurgical trauma (eg, post-splenectomy)
and hypercoagulable states or thrombophil- Pancreatitis
ia. These are essentially the same factors that Inflammatory bowel disease
were explained by Rudolph Virchow when
LocalVenousCongestionorStasis
he described his triad in 1856. His work was
Portal hypertension/liver cirrhosis
central to the evolution of the concept of
Severe congestive heart failure (EF < 20%)
thrombo-hemorrhagic balance. This theory Morbid obesity (BMI > 40)
states that there is a fine balance between fi-
HypercoagulableStates(Thrombophilia)
brin degradation and fibrin production, and
dominance of the latter leads to a hyperco- Inherited
agulable state.9 Activated protein C resistance

Factors that cause direct endothelial Protein C deficiency
injury include major abdominal trauma, es- Protein S deficiency
Antithrombin III deficiency
pecially blunt traumas, and after major ab-
Prothrombin gene mutation
dominal surgeries, especially splenectomy.
Acquired
This is much more common in the case of
large splenomegaly and splenectomy for he- Malignancy (pancreatic cancer, metastatic
abdominal cancer)
matologic malignancies possibly secondary
Hematologic disorders:
to thrombocytosis. MVT affects almost 5% Polycytemia rubra vera
of patients with these disorders undergoing Essential thrombocytosis
splenectomy.10 Local inflammatory intra- Leukemia
abdominal conditions are also associated Hormonal therapy (oral contraceptive pills)
with increased risk of mesenteric venous Lupus anticoagulants
thrombosis.11 This includes cases of pan- Heparin-induced thrombocytopenia

Mesenteric Venous Thrombosis 207
However, the most common cause of Mutation of the prothrombin gene has
secondary MVT remains thrombophilia or also been associated with increased risk of
hypercoagulable conditions. In fact, more thrombotic events. This mutation leads to
than half of all cases of MVT have an higher plasma prothrombin levels and is 4
identifiable hypercoagulable condition.6,12 times as common in patients with a history
Hypercoagulable conditions can be either of venous thrombosis compared to their con-
inherited when the subject is born with a ge- trols.15 In one small study, this mutation was
netic defect or acquired. The acquired types found in 40% of cases with idiopathic portal
of hypercoagulable conditions refer to those vein thrombosis.16
systemic diseases that have been known and
proven to be associated with thrombosis.
However, the true frequency of these dis- Acquired Hypercoagulable Conditions
orders in patients with MVT is difficult to Multiple risk factors are associated with ac-
estimate since most studies in the literature quired hypercoagulable states. The most
included patients with all forms of deep ve- common of these conditions are malignan-
nous thrombosis. One study suggested that cy, hormonal therapy, and antiphospholipid
an acquired or inherited hypercoagulable antibodies.
state was more likely in patients with isolated Malignancy has been a known risk
MVT compared with those with concomi- factor for hypercoagulable state and venous
tant involvement of the portal or splenic thrombosis. This is particularly more com-
veins.13 A list of the hypercoagulable condi- mon in older patients and in those malig-
tions associated with MVT can be found in nancies associated with advanced disease at
Table 17.1. the time of diagnosis (eg, pancreas).17 Cer-
tain cancers are particularly more associated
with MVT. Myeloproliferative disorders,
Inherited Hypercoagulable Conditions especially polycythemia vera and essential
The available data suggest that the most thrombocythemia, are strongly associated
common inherited hypercoagulable condi- with MVT.1 In a small series of patients with
tion associated with MVT is the activated MVT, 10% had associated polycythemia
protein C resistance, secondary to factor vera.18 Pancreatic cancer and metastatic ab-
V Leiden mutation. In one epidemiologic dominal cancers are also strongly associated
study of cases with documented MVT, fac- with MVT. In one series, abdominal cancer
tor V Leiden mutation was present in 45% of was present in 24% of all cases of document-
those cases.6 Activated protein C resistance ed MVT, with pancreatic cancer present in
can also occur due to other forms of muta- half of those cases.6
tions in another 10% of cases.14 Hormonal therapy is also highly associ-
Another cause of inherited hyperco- ated with venous thrombosis in general and
agulable conditions is the deficiency of the MVT in particular. Historically, oral con-
natural anticoagulant proteins that maintain traceptive use accounted for 9% to 18% of
the balance between the coagulant and anti- the episodes of MVT in young women.12,19
coagulant activity of the blood. Deficiencies In a more recent study, oral contraceptives
of protein C, protein S, and antithrombin or estrogen hormonal replacement therapy
III (AT III), collectively, are present in about was present in 12% of women diagnosed
10% of cases with MVT.6,14 with MVT.6 Antiphospholipid antibodies

are directed against plasma proteins bound ing mesenteries, supplying the wall of the
to anionic phospholipids and are associated gut. The venous drainage follows a similar
with venous thrombosis. The antiphospho- pattern, with the venae rectae forming the
lipid antibodies, including the cardiolipin venous arcades, which drain the gut into the
antibodies and the lupus anticoagulants, are inferior mesenteric and superior mesenteric
present in another 5% to 10% of cases.6,8 veins. The inferior mesenteric vein joins the
splenic vein before the confluence with the
superior mesenteric vein behind the neck of
Pathophysiology the pancreas to form the portal vein that sup-
plies the liver parenchyma (Figure 17.1).
The celiac, superior mesenteric, and inferior There are multiple extrinsic and intrin-
mesenteric arteries are the visceral branches sic mechanisms that control the blood flow
of the aorta supplying the gut. The branches to the bowel wall. This is mainly controlled
from these arteries terminate as the arteriae via changes in the resistance of the mesenter-
rectae in the distal parts of the correspond- ic arterioles, which accounts for the marked
Figure 17.1 Normal mesenteric venous circulation. Reprinted with permission from Agur AM, Dailey, AF, Boileau
Grant JC. Grants Atlas of Anatomy. 11th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2004.

variation in the splanchnic blood flow, be- keeps flowing into the bowel wall in patients
tween 10% and 35% of the cardiac output, with venous compromise. This leads to in-
depending on variable conditions such as creased intravascular hydrostatic pressure,
fasting and postprandial states and systemic which dilates the blood vessels and widens
hypotension. The bowel can withstand an the fenestrations between the vascular en-
acute 75% reduction of the mesenteric blood dothelial cells. This results in extravasation
flow for up to 12 hours without significant of plasma and red blood cells into the bowel
damage occurring.20 When ischemic injury wall or lumen. Tension in the submucosal
of the intestine develops, progressive vaso- extravascular compartment or prolonged
constriction develops in the affected vas- stasis-induced thrombosis of the microvas-
cular bed, thereby reducing collateral flow. culature may result in interruption of the ar-
Vasoconstriction can persist even after blood terial blood flow.23 Grossly, the bowel wall is
flow has been restored, leading to continued swollen and looks cyanotic, with intramural
bowel ischemia and providing the rationale hemorrhage. These changes also affect the
for therapeutic infusion of the vasodilator nearby mesentery. The transition from the
papaverine for treatment of acute mesenteric affected bowel segment to the normal bowel
ischemia.20,21 is usually gradual, unlike that seen with arte-
The location of the initial thrombosis rial occlusion, which makes it more difficult
within the mesenteric venous circulation to accurately evaluate viable from nonviable
varies with the etiology. Thrombosis due bowel during surgery. The arterial pulsations
to intra-abdominal causes such as cirrhosis, are present up to the bowel wall, but arte-
neoplasm, or operative injury starts at the rial vasoconstriction frequently intervenes,
large vessels (eg, superior mesenteric vein which accentuates the compromised intra-
or portal vein) and progresses peripherally mural blood flow in the congested bowel.
to involve the smaller venous arcades and Later on, transmural infarction of the bowel
arcuate channels. However, thrombosis due sets in, and at that point it may be impos-
to an underlying hypercoagulable state be- sible to differentiate venous from arterial
gins in the small vessels and progresses to occlusion as a cause of the bowel infarction.
involve the larger vessels.1,13 Intestinal infarc- Transmural bowel infarction happens in
tion rarely occurs unless the branches of the 16% of patients suffering from acute MVT.24
peripheral arcades and the microcirculation Serosanguineous peritoneal fluid accompa-
are thrombosed, even when the junction of nies early hemorrhagic infarction. When
the portal vein and the superior mesenteric a large segment of the bowel is affected, it
vein is occluded. In a recent large series of leads to extensive third space fluid loss with
patients diagnosed with MVT, bowel infarc- systemic hypovolemia.13,20,21
tion occurred in 33% of them that required Two factors contribute to the bowel in-
bowel resection.22 Inferior mesenteric vein jury when affected by acute MVT, the first
thrombosis leading to infarction has been being transmural ischemia secondary to ces-
reported in fewer than 6% of cases with mes- sation of the blood flow. The second factor
enteric venous thrombosis.21 is reperfusion injury if restoration of flow is
When acute MVT happens, the ve- achieved. Most of the damage is due to reper-
nous drainage of the affected segment of fusion injury after a brief period of ischemia,
the bowel is compromised. The intramural while the detrimental effects of hypoxia pre-
blood volume increases as the arterial blood dominate with longer periods of ischemia.25

MVT can lead to presinusoidal portal is present in almost all patients with acute
hypertension in the long run and is classi- and subacute MVT, no other symptoms are
fied as chronic mesenteric venous throm- pathognomonic of MVT. The pain is usu-
bosis. This usually happens when patients ally central abdominal and colicky in nature.
have no symptoms at the time of the ini- The duration of pain varies, but 75% of cases
tial thrombosis but secondary pathologic had the symptoms for more than 48 hours,
changes develop over the time. This usually a very important distinction from mesenter-
manifests as gastrointestinal bleeding from ic arterial causes where the presentation is
esophageal and/or intestinal varices. Most of much earlier.7 Mathews and White27 found
these patients have associated thrombosis of that approximately 50% of their patients had
the major veins such as the portal or splenic pain from 5 to 30 days before seeking medi-
vein. It was reported that portal hypertension cal attention, and 27% reported abdominal
develops in 25% of patients who suffered pain for more than 1 month. It was also
from acute MVT.24 reported that patients who presented with
symptoms more than 5 days had a better
prognosis than patients presenting earlier,
Clinical Presentation possibly due to a more benign disease form
that allowed them to delay seeking medical
MVT can present with acute onset within attention for so long.21
hours, subacute within days to even weeks, Anorexia, nausea, vomiting, and di-
or chronic that is usually asymptomatic until arrhea are also common in acute MVT,
late complications occur. Patients with the occurring in more than 50% of cases. Clini-
acute presentation are at the highest risk for cal gastrointestinal bleeding in the form of
developing bowel infarction and peritoni- bloody diarrhea or melena is present in 15%
tis. On the other hand, the subacute form of cases. Hematemesis is also present in an
is more indolent, where abdominal pain is additional 13% of cases. Actually, the pres-
prominent but neither infarction nor vari- ence of hematemesis, as well as bleeding per
ceal bleeding is likely. In less common situa- rectus, should alert the physician to the pos-
tions, the subacute form will evolve over the sibility of mesenteric ischemic catastrophe.21
period of days or weeks into intestinal infarc- However, occult blood in stool is found in
tion and peritonitis, blurring the distinction nearly 50% of cases.12
between the acute and the subacute presen- Initial physical findings in acute MVT
tation.1 In the chronic form of the disease, vary greatly, reflecting both different stages
patients are essentially asymptomatic, and and degrees of bowel injury. Most of the pa-
their presentation can be upper or lower gas- tients will have some degree of abdominal
trointestinal bleeding secondary to esopha- tenderness and decreased bowel sounds with
geal or intestinal varices. Those patients can abdominal distention. However, a minor-
also develop hypersplenism with pancytope- ity of patients will develop peritoneal signs
nia or secondary thrombocytopenia.21,26 in the form of rebound tenderness and ab-
The hallmark of mesenteric ischemia, dominal rigidity, which signifies the pres-
whether it is due to arterial or to venous ence of intestinal infarction with peritonitis.
thrombosis, is abdominal pain that is out In a large series of patients with MVT, this
of proportion to the physical findings. With occurred in 33% of cases.22 Hemodynamic
the exception of the abdominal pain, which instability can also occur if a large segment

of the bowel is affected due to fluid seques- suggest the diagnosis of intestinal ischemia.21
tration within the bowel lumen or the ab- A useful clinical guideline is that any patient
dominal cavity. It also appears later in the with acute abdominal pain and metabolic
disease as a manifestation of septic shock acidosis is suspected of having acute mesen-
secondary to the bowel infarction and septic teric ischemia until proven otherwise.
peritonitis. It was consistently shown that the
presence of hemodynamic instability with a
systolic blood pressure of less than 90 mm Workup
Hg denotes a poor prognosis.28 About half
of the patients have a personal or family his- In the absence of any reliable and specific
tory of deep venous thrombosis, pulmonary symptoms, signs, and laboratory studies,
embolism, or a heritable hypercoagulable the preoperative diagnosis of acute MVT is
state, and this is an important point to in- difficult. Also, the severe variability in the
quire about as it can guide the diagnosis and duration and severity of the manifestations
therapy of the case.7 significantly adds to the difficulty. The aim
Laboratory studies in all forms of intes- of the workup is to reach the correct diag-
tinal ischemia have low sensitivity and speci- nosis of MVT before bowel infarction devel-
ficity. While abnormal laboratory values may ops, in which case the prognosis becomes
be helpful in bolstering suspicion for acute significantly worse. In the past, the correct
mesenteric ischemia, normal laboratory val- diagnosis of MVT was reached in more than
ues do not exclude it and do not justify delay- 90% of cases at laparotomy for patients with
ing management when clinical suspicion is peritoneal signs secondary to bowel infarc-
present. There is no specific laboratory test tion, signifying that the diagnosis was severe-
that is pathognomonic of MVT. The pres- ly delayed. In recent years, there has been
ence of increased serum lactate levels and enormous improvement of diagnostic radio-
metabolic acidosis may serve to identify pa- graphic modalities that helped in achieving
tients with established bowel infarction, but this goal. Nowadays, most of the patients are
this is a late finding.1 Elevated serum amy- diagnosed prior to surgery, with significant
lase levels have been observed in half the pa- improvement in morbidity and mortality.
tients with intestinal ischemia. An elevated
phosphate level has also been observed in Projectional Radiography
80% of cases. Creatine kinase (CK)-BB iso-
enzyme elevations have been found to cor- Plain films of the abdomen are nonspecific
relate with bowel infarction while total CK and can be normal in 25% of patients. Posi-
levels were not found to be useful. Lactate tive findings include intestinal dilatation,
dehydrogenase (LDH) elevation was present small bowel pseudo-obstruction pattern,
in 75% of cases with bowel infarction but did or paralytic ileus. More specific but far less
not differentiate between ischemia and in- common findings include thumbprinting,
farction.29-31 In fact, the simple leukocytosis in which multiple round, smooth, soft tissue
with a white cell count above 12,000/cu mm densities project into the intestinal lumen as
with an increase in the proportion of poly- a result of mucosal and submucosal edema
morphonuclear cells is the most clinically and hemorrhage. Specific but usually late
helpful lab test and is present in two-thirds signs include the presence of air in the wall
of cases. Currently, laboratory tests can only of the bowel (peumatosis intestinalis) (Fig-

ure 17.2), portal venous gas or free intra-

peritoneal air, which usually signify bowel
Multidetector Computed Tomography
infarction.32 It is to be remembered that the Multidetector computed tomography
use of intraluminal dye studies of the bowel (MDCT) has become the preferred imag-
are contraindicated because they interfere ing modality for the evaluation of suspected
with the visualization using angiogram and MVT cases. It can be performed quickly in
add little information to the plain film.21 critically ill patients and is less dependent on
operator skill and patient factors than other
imaging modalities.23 In fact, the sensitivity
Doppler Flow Ultrasonography of CT scan in diagnosing MVT was found to
Doppler flow ultrasonography of the mes- be 90% in several studies.7 CT scan produc-
enteric veins can be used to demonstrate es volume data that can be manipulated and
thrombi, especially in the large veins like the viewed in different projections evaluating
superior mesenteric vein and portal vein. The the bowel wall, the mesentery, the surround-
findings include lucent filling defects, thick- ing fat, the omenta, and the mesenteric ves-
ened mesentery, and bowel wall and intra- sels themselves, visualizing both the arteries
peritoneal free fluid. However, like all acute and veins and any pathologic abnormalities
mesenteric ischemia disorders, because of including stenosis, atherosclerotic plaques,
body habitus, the overlying bowel gas that in- and intravascular thrombi.34 CT scan can
terferes with proper visualization and patient exclude other cases of acute abdominal
cooperation, ultrasonography is not typically pain. Specifically, in cases of MVT, the CT
used in the initial evaluation of acutely ill pa- scan will show failure to opacify the mesen-
tients with suspected acute MVT.33 teric veins with intravenous filling defects,
Figure 17.2 Plain abdominal film showing extensive pneumatosis

intestinalis.

enlarged superior mesenteric vein and per- agents.23 MR, however, is best used in the
sistent enhancement of the bowel wall due nonacute setting. The critically ill patient
to congestion, as well as thickened mesen- with suspected acute mesenteric ischemia
tery (Figure 17.3). In late cases, it will show usually has life support apparatus that is in-
pneumatosis, portal venous gas, and possibly compatible with the MR scanner. Those pa-
intraperitoneal free air and fluid.21 tients are best scanned with MDCT, which
also offers better spatial resolution.35
Magnetic Resonance Imaging/

Selective Mesenteric Angiography
Magnetic Resonance Angiography Selective mesenteric angiography was the
Magnetic resonance imaging (MRI) and diagnostic modality of choice before the
magnetic resonance angiography (MRA) 1990s, when CT scan became available. In
can also be used, providing information that fact, a selective mesenteric angiogram is
is similar to CT scan, with the theoretical much less sensitive than CT scan in diag-
advantage of not using ionizing irradiation nosing MVT, where it was able to diagnose
and using gadolinium-based intravascular only 55.5% compared to the above 90%
contrast rather than iodinated contrast used sensitivity of CT scan.22 Nowadays, mesen-
in CT scans in patient with renal insufficien- teric angiogram is performed primarily for
cy, which has been negated by the increased transcatheter therapeutic interventions both
incidence of nephrogenic systemic fibrosis for arterial and venous mesenteric ischemic
related to the gadolinium-based MR contrast disorders. Those interventions include in-
Figure 17.3 CT scan with intravenous contrast (venous phase) in a case

showing partial occlusion of the superior mesenteric vein by a thrombus
(shown by arrow).

fusion of vasodilator agents (papaverine), the pain persist for several days before pre-
thrombolytic agents (streptokinase, uroki- sentation, should have MDCT as their initial
nase, recombinant tissue plasminogen acti- evaluation after starting aggressive resuscita-
vator), and angioplasty.36 The angiographic tion. In cases where MVT was diagnosed,
findings of MVT have been determined the patient should immediately be started
experimentally and clinically and include21: on heparin. If the patient has obvious peri-
toneal signs, the patient should immediately
go the operating room for exploratory lapa-
Demonstration of a thrombus in the rotomy. A mesenteric angiogram is needed
superior mesenteric vein (SMV) on the in case of a negative CT scan for MVT or in
venous phase of the study with partial cases where endovascular interventions are
or complete occlusion. needed. At any time, if the patient starts de-
Failure to visualize the SMV or portal veloping peritoneal signs, the patient has to
vein on delayed views. undergo exploratory laparotomy.
Slow or absent filling of the mesenteric
veins.
Mesenteric arterial spasm with back Management
flushing into the aorta on intra-arterial
contrast injection. The principles of management of MVT are
Failure of the arterial arcades to empty. 3-fold:
Prolonged blush of the wall of the
involved bowel segment. Prevention of further development of
The angiography can also show slow venous thrombosis,
reconstitution of the mesenteric resection of necrotic bowel, and
venous blood flow proximal to the prevention of recurrence of venous
thrombus, which is important in the thrombosis.
therapeutic decisions.
Historically, the treatment of acute and sub-
acute MVT included anticoagulation with
The workup of patients with suspected exploratory laparotomy with resection of the
MVT is summarized in the guidelines of necrotic bowel. Laparotomy was an essential
management of intestinal ischemia from the part of the management both for diagnostic
American Gastroenterological Association as and therapeutic reasons, as preoperative di-
illustrated in the algorithm in Figure 17.4.37 agnosis of MVT was not possible. Recently,
Patients presenting with abdominal pain that with the advent of MDCT, the preopera-
is severe enough to call to the attention of a tive diagnosis of MVT was possible in most
physician, whose pain persists for more than cases. Adding to that the rapid and extensive
2 or 3 hours, and whose clinical picture does development of endovascular techniques
not suggest other abdominal problems (eg, as minimally invasive therapy options, cur-
cholecystitis or diverticulitis) should be eval- rently the use of exploratory laparotomy is
uated and treated for acute mesenteric isch- limited to patients who have persistent peri-
emia. In those patients, suspicion of MVT is toneal signs for possible bowel resection. In
signaled by a history of DVT or familial or a one study, that was only needed in 33% of
personal hypercoagulability state, or having their patients with acute MVT.22

laparotomy
resuscitate the patient and no yes
correct any predisposing or Plain film History of DVT or familial
Peritoneal findings Abdominal
precipitating factors hypercoagulable state no angiogram
other yes
cause
Dynamic CT scan
Mesenteric Venous Thrombosis
no persistent Persistent
peritoneal findings peritoneal findings
Heparin with or
without thrombolytic laparotomy
agents
Short ischemic
extensive ischemic Involvement
segment
resect nonviable Viable
Heparin resect Close Main vein open Main vein

or reconstituted occluded
Coumadin long term
parenteral
Thrombectomy,
nutrition Heparin, papaverine
heparin, papaverine
Second look
+/ resection
Thrombolytic Agents
Figure 17.4 Diagnosis and treatment of acute mesenteric venous thrombosis. Solid lines indicate accepted
management plan; dashed lines indicate alternate management plan. DVT, deep venous thrombosis. (From American
Gastroenterological Association Medical Position Statement: guidelines on intestinal ischemia. Gastroenterology
118(5): 951-3.) With permission from Elsevier.
Currently, the management of acute MVT. The theoretical aim of heparin in such
and subacute MVT includes: a situation is to prevent progression of the
thrombosis and limit the amount of bowel
Anticoagulation resection, if needed. Contrary to heparin use
Exploratory laparotomy in acute mesenteric arterial thrombosis and
Endovascular options for recanalizing embolism, where the use is delayed for 48
the thrombosed veins hours postoperatively to avoid postoperative
Adjunctive supportive measures bleeding, in acute and subacute MVT it is
started immediately with the diagnosis of the
condition, continues intraoperatively, and in
Anticoagulation the immediate postoperative period. Antico-
Heparin is an essential component of ther- agulation should be started even in the pres-
apy in all patients with acute and subacute ence of gastrointestinal bleeding, as the risk

of bleeding is outweighed by the benefit of is any doubt of involvement of the remain-

preventing bowel infarction.1 Anticoagula- ing bowel, continuity may not be restored
tion has been shown to decrease the rate of in the original operation, and the patient is
recurrence from 25% to 13%. It also reduced sent back to the intensive care with a second-
mortality from 50% to 13%.28 Standard an- look laparotomy performed 18 to 24 hours
ticoagulation with Coumadin is continued after the initial exploration to check for the
for at least 6 months. In cases where there remaining bowel condition.21 An adjunctive
is an underlying hypercoagulable state, the measure at that time can be continued in-
therapy can continue indefinitely.19,38 fusion of papaverine, a potent vasodilator,
in the superior mesenteric artery (SMA)
through a catheter inserted via the femoral
Exploratory Laparotomy artery approach. This is because there is a
Historically, this was needed in almost all component of mesenteric arterial spasm in
cases because most cases of acute MVT were the pathologic process of acute MVT and
diagnosed at laparotomy, due to the inabil- papaverine can help alleviate the vasospasm
ity to diagnose the condition preoperatively. to improve blood flow, which may avert re-
Nowadays, with the advent of the advanced section of reversibly ischemic bowel.28 If no
radiologic diagnostic modalities including further affection of the bowel is found, the
CT, MRA, and mesenteric angiogram, most continuity is restored and the abdomen is
of the cases are diagnosed preoperatively closed. It is to be stressed that anticoagula-
and many are treated nonoperatively, using tion should be continued throughout the
the advanced endovascular techniques now process, including intraoperatively.
available. Exploratory laparotomy is reserved In some cases, there is extensive in-
for those cases with persistent peritoneal volvement of the bowel with obvious bowel
signs (eg, abdominal tenderness, rebound necrosis, in which case and at the discretion
tenderness, and rigidity). It is also indicat- of the surgeon 1 of 2 options are available, ei-
ed in cases with obvious manifestations of ther to perform major resection of the bowel
bowel infarction including free intraperito- with implementation of lifelong parenteral
neal air and portal venous gas. nutrition due to short loop syndrome, or
At laparotomy, the bowel is evaluated closing the abdomen as the resection would
for extent of involvement and viability, both be incompatible with life. However, if there
clinically and, if necessary, by administra- is extensive involvement of the bowel that is
tion of fluorescein with examination under of doubtful viability, in which case the en-
ultraviolet light. In cases in which there is dovascular options are extremely helpful to
a localized involvement of the bowel, the attempt to recanalize the thrombosed mes-
segment involved is resected and continuity enteric veins with saving as much bowel as
is restored if the remaining bowel appears possible and a second laparotomy to excise
absolutely normal. Historically, it was rec- the obviously nonviable bowel segments.
ommended to perform wider resection than These techniques can help save some of
the apparently involved bowel because the those patients with extensive disease, which
mesenteric venous thrombosis process ex- was not possible using the conventional anti-
tends further in the mesentery; however, re- coagulation and resection alone.
cent evidence suggests that there is no need Historically, there have been some re-
to sacrifice viable bowel. However, if there ports of open mesenteric venous thrombec-

tomy using Fogarty balloon thrombectomy Thrombolysis

catheters, which was performed for cases Multiple case reports and small case series
with extensive bowel involvement with have shown thrombolytic therapy to be ef-
doubtful viability. These were indicated fective in treating acute MVT. Theoreti-
when a long segment of questionable bowel cally, heparin prevents further thrombosis
was found and the angiogram or opera- and thrombus propagation; thrombolytic
tive findings indicate complete thrombosis therapy with active clot dissolution may be
of the SMV at its junction with the portal more effective than traditional treatment of
vein, with or without extension into the por- venous thrombosis with anticoagulation and
tal vein. A second-look operation was always clot stabilization.39 There are several routes
performed after the venous thrombectomy. through which the thrombolytic agent is
This technique has been reported in only a delivered. Systemic intravenous administra-
few patients in the literature, which makes tion of urokinase or tissue plasminogen ac-
it impossible to define its role in the treat- tivator (tPA) has been reported; 40 however,
ment of acute MVT.21 Additionally, the it does not deliver a high concentration of
endovascular options have largely replaced the thrombolytic agent into the mesenteric
the need to perform open mesenteric ve- venous circulation, in addition to the in-
nous thrombectomy. creased bleeding complications with sys-
temic use as well as the need for a higher
dose to be effective. The 2 commonly used
Endovascular Interventions routes of administration are the transarte-
Recently, with the extensive development rial route via a catheter placed in the SMA
of minimally invasive endovascular inter- and direct mesenteric venous administra-
ventions, several therapeutic options have tion through a catheter placed in the portal
evolved in the management of acute and sub- venous system via the transjugular trans-
acute MVT. These interventions are mostly hepatic approach,41,42 or via direct percutane-
used when nonoperative therapy for acute or ous trans-hepatic approach.43,44 Whereas the
subacute MVT is performed to reverse the trans-hepatic approaches are successful in
clinical course and avoid bowel infarction. clearance of thrombus in the larger mesen-
It is also used in cases when at exploratory teric veins by delivering direct and high con-
laparotomy there is an extensive and diffuse centrations of the thrombolytics and gaining
affection of the bowel with large areas of direct access to the mesenteric venous system
doubtful viability. However, in cases where allowing the performance of percutaneous
exploratory laparotomy is used and there is mechanical thrombectomy, they may not be
a localized segment of nonviable bowel with ideal in addressing thrombus in the smaller
no doubtfully viable loops, the therapy re- secondary and tertiary venous branches. In
mains the standard localized resection and those cases, the intra-arterial administration
anticoagulation. These interventions can be of thrombolytics via the SMA is more effec-
grouped into 3 options: tive, as it enables the thrombolytic agent to
permeate the capillaries and small venules,
Thrombolysis something that cannot be achieved using
Percutaneous mechanical the trans-hepatic approach.39 SMA admin-
thrombectomy istration of the thrombolytic also avoids the
Intra-arterial vasodilator infusion added risk of bleeding secondary to punc-

turing the liver parenchyma both through thrombectomy device (ev3, Plymouth, MN),
the transjugular approach and especially and the Arrow Trerotola device (Arrow In-
through the direct percutaneous approach. ternational Inc, Reading, PA). Aspiration
Very few case reports of thrombolysis via a thrombectomy using a large angulated cath-
surgically placed mesenteric venous catheter (at least 8 Fr) has also been described.42
eter into one of the small venous tributar- The aim of these devices is to debulk the
ies during exploratory laparotomy are found mesenteric venous clot, thereby reducing
in the literature.45 Most of the earlier stud- the dose and duration needed for throm-
ies of mesenteric venous thrombolysis used bolysis, especially in patients with high risk
urokinase. Nowadays, because urokinase of bleeding complications. Also, balloon an-
is no longer available on the market, tPA is gioplasty has also been described for cases
typically used. The usual dose is an initial in which there have been venous stenotic le-
bolus of 2 mg followed by infusion of 1 to 2 sions or large clot burden not responding to
mg/h for 24 hours. At that time, a catheter thrombolysis or mechanical thrombectomy
check and repeat angiogram to evaluate the because of a large chronic component in the
response are performed. If there is still re- clot. The angioplasty is used to relieve the
sidual thrombus left, continued infusion for stenosis with or without use of stents or cov-
another 24 hours is performed, after which ered stents. They are also useful to fragment
time the catheter is removed. In the percuta- and displace the clot to restore flow within
neous trans-hepatic approach, the track has the portal venous system to relieve the con-
to be sealed with haemostatic glue and coils gestion in the affected bowel.42
to avoid intraperitoneal bleeding from the
liver track, which can be life-threatening. Intra-arterial Vasodilator Infusion
The patient should be closely moni- Mesenteric arterial vasospasm is commonly
tored for manifestations of bleeding and present in cases of acute and subacute MVT.
fibrinogen levels to avoid overconsumption It is specifically useful in cases following ex-
and excessive thrombolysis. Heparin infu- ploration where there is doubtful bowel left
sion continues throughout the thrombolytic behind for a second-look laparotomy. In this
infusion, however, at a smaller dose (500 situation, the use of the vasodilators can im-
units/h), to resume full dose heparinization prove the blood flow to the affected bowel,
once thrombolytic infusion has stopped. which in addition to anticoagulation and
other endovascular interventions, can help
Percutaneous Mechanical Thrombectomy limit the amount of bowel to be resected.1
This is another endovascular adjunct proce- Multiple vasodilators can be used for this
dure for clearing acute MVT. It can be used indication, including tolazoline, nitroglyc-
only when there is trans-hepatic access into erin, glucagon, and isoproterenol. However,
the portal venous system. It is also indicated the most-used vasodilator for this purpose
when there is a large clot burden in the main is papaverine. Papaverine is a nonaddictive
mesenteric veins. Several commercially opium derivative extracted from the poppy
available percutaneous mechanical throm- plant. It acts by inhibiting the phosphdiester-
bectomy devices have been reportedly used ase enzyme that metabolizes cyclic adenos-
for that indication including the AngioJet ine monophosphate (cAMP). The net effect
rheolytic thrombectomy device (MEDRAD, is to increase the tissue levels of cAMP,
Inc., Warrendale, PA), the Helix Clot Buster which is a direct vascular smooth muscle re-

laxant. Ninety percent of papaverine under- require surgery and bowel resection are sick-
goes first-pass metabolism by the liver, which er and have longer hospital stays and a more
limits its systemic toxicity and side effects.38 complicated course than those who do not
The usual dose is a bolus of 30 to 60 mg, fol- require surgery.1 However, there has been no
lowed by infusion at a rate of 30 to 60 mg/h. correlation between the length of resected
The duration of therapy is usually 24 to 48 bowel and the mortality rate, although it has
hours and is guided by clinical and angio- a detrimental effect on long-term morbidity
graphic response. Patients receiving papaver- in the form of short bowel syndrome and its
ine infusion should be closely monitored in sequelae.21
an ICU setting for hypotension. The sudden In recent years, the mortality rate has
onset of hypotension in such patients indi- been slightly lower than those from earlier
cates that the catheter has been dislodged series, now ranging between 13% and 30%.22
from the SMA and that the drug is being in- This may be due, in part, to early diagnosis
fused systemically and requires another trip and aggressive management. However, this
to the angiography suite for catheter check might also be a selection bias, because we
and replacement.38 are catching more patients with milder and
more benign disease due to the diagnosis
of the condition using CT scan instead of
Supportive Measures the previous reports in which the diagnosis
All patients with acute and subacute MVT was mostly made at laparotomy, indicating
should be managed in an ICU setting. Ade- a more severe and advanced disease.22 Only
quate resuscitation is mandatory. They must a few studies have described the long-term
have complete bowel rest with NPO and outcome in patients who developed MVT.
nasogastric tube suction. Broad-spectrum One of the largest series included 60 patients
antibiotic coverage is not indicated in the who were followed for a median of 3.5 years.
absence of bowel perforation or peritonitis.1 The overall survival at 1 and 5 years was 82%
and 78%, respectively. These values were
86% and 82%, respectively, after excluding
Outcome patients with an underlying malignancy.46
Long-term survival depends primarily on
The true incidence of MVT is unknown, the cause of the thrombosis. If cancer is the
mainly because it is not diagnosed in many underlying cause, survival is short and deter-
patients because of the vague, mild, and mined by the nature of the cancer.1 The use
nonspecific manifestations. Because of that, of endovascular techniques in the manage-
the true outcomes of the condition are es- ment of acute MVT may contribute to the
sentially unknown, as well. Historically, the improving mortality rates; however, the use
mortality rate associated with acute MVT of those techniques remains limited, and
ranged between 20% and 50%.1 Although their true value remains to be evaluated
high, it was still better than other cases of when more experience is accumulated.
acute arterial mesenteric ischemia. Survival Mesenteric venous thrombosis is a dis-
depends on multiple factors, including age, ease with a high rate of recurrence, and re-
the presence or absence of coexisting comor- currences are most common within the first
bid conditions, and the timing of the diag- 30 days after presentation.47 Coumadin ther-
nosis and surgical intervention. Patients who apy significantly reduces the risk of recur-

rence from 20%25% to 13%15%. There intestine due to occlusion of the mesenteric
is no evidence that long-term Coumadin vessels. Ann Surg. 1895(21):9-23.
therapy is beneficial except in cases where 4. Cokkinis A. Mesenteric Vascular Occlusions.
there is a persistent hypercoagulable state London: Bailliere, Tindall and Cox; 1926. p
when lifelong therapy is indicated.19,38 1-93.
The natural history of chronic MVT is 5. Warren S, Eberhard TP. Mesenteric venous
not known, but most of the patients appear thrombosis. Surg Gynecol Obstet. 1935;141:
to be asymptomatic. The percentage of pa- 740-2.
tients with chronic MVT who develop late 6. Acosta S, Alhadad A, Svensson P, Ekberg O.
gastrointestinal bleeding or hypersplenism Epidemiology, risk and prognostic factors
has not been determined but is probably in mesenteric venous thrombosis. Br J Surg.
small. As MVT is recognized more frequent- 2008;95(10):1245-51.
ly on CT scans done for other disorders, our 7. Rhee RY, Gloviczki P. Mesenteric
understanding of the natural history of this venous thrombosis. Surg Clin North Am.
disorder should improve.21 1997;77(2):327-38.
MVT remains an underdiagnosed and 8. Martinelli I, Mannucci PM, De Stefano V,
understudied condition that we need to Taioli E, Rossi V, Crosti F, et al. Different
learn more about. Although there has been risks of thrombosis in four coagulation defects
significant improvement in the diagnosis associated with inherited thrombophilia:
and management of the disease in recent a study of 150 families. Blood.
years, this has not been translated into sig- 1998;92(7):2353-8.
nificant reduction in mortality and morbid- 9. Bagot CN Arya R. Virchow and his triad:
ity, which tells us that there is still a long way a question of attribution. Br J Haematol.
to go to achieve that goal. More is needed 2008;143(2):180-90.
to study the natural history of the disease. 10. Stamou KM, Toutouzas KG, Kekis PB, Nakos
Also, we need to evaluate the efficacy and S, Gafou A, Manouras A, et al. Prospective
refine the new modalities of therapy, espe- study of the incidence and risk factors of
cially the minimally invasive endovascular postsplenectomy thrombosis of the portal,
techniques, as well as developing new thera- mesenteric, and splenic veins. Arch Surg.
pies that can prevent or limit bowel injury, 2006;141(7):663-9.
thus reducing the need for open surgery with 11. Hatoum OA, Spinelli KS, Abu-Hajir M, Attila
bowel resection. T, Franco J, Otterson MF, et al. Mesenteric
venous thrombosis in inflammatory bowel
disease. J Clin Gastroenterol. 2005;39(1):27-31.
references 12. Harward TR, Green D, Bergan JJ, Rizzo RJ,
1. Kumar S, Sarr MG, Kamath PS. Mesenteric Yao JS. Mesenteric venous thrombosis. J Vasc
venous thrombosis. N Engl J Med. 2001; Surg. 1989;9(2):328-33.
345(23):1683-8. 13. Kumar S, Kamath PS. Acute superior
2. Schoots IG, Koffeman GI, Legemate DA, mesenteric venous thrombosis: one disease
Levi M, van Gulik TM. Systematic review or two? Am J Gastroenterol. 2003;98(6):
of survival after acute mesenteric ischaemia 1299-304.
according to disease aetiology. Br J Surg. 14. Provan D, OShaughnessy DF. Recent
2004;91(1):17-27. advances in haematology. BMJ. 1999;318
3. Elliot J. The operative relief of gangrene of (7189):991-4.

15. Margaglione M, Brancaccio V, Giuliani N, 25. Zimmerman BJ, Granger DN. Reperfusion
DAndrea G, Cappucci G, Iannaccone L, et injury. Surg Clin North Am. 1992;72(1):65-83.
al. Increased risk for venous thrombosis in 26. Warshaw AL, Jin GL, Ottinger LW.
carriers of the prothrombin GA20210 gene Recognition and clinical implications of
variant. Ann Intern Med. 1998;129(2):89-93. mesenteric and portal vein obstruction
16. Chamouard P, Pencreach E, Maloisel F, in chronic pancreatitis. Arch Surg
Grunebaum L, Ardizzone JF, Meyer A, et 1987;122(4):410-5.
al. Frequent factor II G20210A mutation 27. Mathews JE, White RR. Primary mesenteric
in idiopathic portal vein thrombosis. venous occlusive disease. Am J Surg 1971;
Gastroenterology. 1999;116(1):144-8. 122(5):579-83.
17. Elting LS, Escalante CP, Cooksley C, 28. Boley SJ, Kaleya RN, Brandt LJ. Mesenteric
Avritscher EB, Kurtin D, Hamblin L, et al. venous thrombosis. Surg Clin North Am.
Outcomes and cost of deep venous thrombosis 1992;72(1):183-201.
among patients with cancer. Arch Intern Med. 29. Lange H, Jackel R. Usefulness of plasma
2004;164(15):1653-61. lactate concentration in the diagnosis of acute
18. Alvi AR, Khan S, Niazi SK, Ghulam M, abdominal disease. Eur J Surg. 1994;160(6-
Bibi S. Acute mesenteric venous thrombosis: 7):381-4.
improved outcome with early diagnosis and 30. Jamieson WG, Marchuk S, Rowsom J,
prompt anticoagulation therapy. Int J Surg. Durand D. The early diagnosis of massive
2009;7(3):210-3. acute intestinal ischaemia. Br J Surg. 1982;69
19. Abdu RA, Zakhour BJ, Dallis DJ. Mesenteric suppl:S52-3.
venous thrombosis1911 to 1984. Surgery. 31. Fried MW, Murthy UK, Hassig SR, Woo J,
1987;101(4):383-8. Oates RP. Creatine kinase isoenzymes in the
20. Rosenblum JD, Boyle CM, Schwartz LB. diagnosis of intestinal infarction. Dig Dis Sci.
The mesenteric circulation. Anatomy 1991;36(11):1589-93.
and physiology. Surg Clin North Am. 32. Baker SR, Cho KC. The Abdominal Plain
1997;77(2):289-306. Film with Correlative Imaging. 2nd ed.
21. Boley SJK. Mesenteric ischemic disorders. Stamford (CT): Appleton & Lange; 1999. p
In: Zinner MJ, ed. Maingots Abdominal 315-29.
Operations, 10th ed. Upper Saddle River, NJ: 33. Dietrich CF, Jedrzejczyk M, Ignee A.
Appleton & Lange; 1997. Sonographic assessment of splanchnic arteries
22. Morasch MD, Ebaugh JL, Chiou AC, and the bowel wall. Eur J Radiol. 2007;64(2):
Matsumura JS, Pearce WH, Yao JS. 202-12.
Mesenteric venous thrombosis: a changing 34. Horton KM, Fishman EK. Computed
clinical entity. J Vasc Surg. 2001;34(4):680-4. tomography evaluation of intestinal ischemia.
23. Gore RM, Thakrar KH, Mehta UK, Berlin Semin Roentgenol. 2001;36(2):118-25.
J, Yaghmai V, Newmark GM. Imaging 35. Michaely HJ, Dietrich O, Nael K, Weckbach
in intestinal ischemic disorders. Clin S, Reiser MF, Schoenberg SO. MRA of
Gastroenterol Hepatol. 2008;6(8): 849-58. abdominal vessels: technical advances. Eur
24. Brunaud L, Antunes L, Collinet-Adler S, Radiol. 2006;16(8):1637-50.
Marchal F, Ayav A, Bresler L, et al. Acute 36. Kim ST, Nemcek AA, Vogelzang RL.
mesenteric venous thrombosis: case for Angiography and interventional radiology of
nonoperative management. J Vasc Surg. the hollow viscera. In: Gore RM, Levine MS,
2001;34(4):673-9. eds. Textbook of Gastrointestinal Radiology.

3rd ed. Philadelphia: Saunders; 2008. thrombectomy, and direct thrombolysis in

p117-40. the treatment of acute portal and superior
37. American Gastroenterological Association mesenteric vein thrombosis. Cardiovasc Interv
Medical Position Statement: guidelines Radiol. 2007;30(5):1070-4.
on intestinal ischemia. Gastroenterology. 43. Rosen MP, Sheiman R. Transhepatic
2000;118(5):951-3. mechanical thrombectomy followed by
38. Frishman WH, Novak S, Brandt LJ, Spiegel infusion of TPA into the superior mesenteric
A, Gutwein A, Kohi M, et al. Pharmacologic artery to treat acute mesenteric vein
management of mesenteric occlusive disease. thrombosis. J Vasc Interv Radiol. 2000;11(2 Pt
Cardiol Rev. 2008;16(2):59-68. 1):195-8.
39. Henao EA, Bohannon WT, Silva MB Jr. 44. Ryu R, Lin TC, Kumpe D, Krysl J, Durham
Treatment of portal venous thrombosis with JD, Goff JS, et al. Percutaneous mesenteric
selective superior mesenteric artery infusion of venous thrombectomy and thrombolysis:
recombinant tissue plasminogen activator. successful treatment followed by liver
J Vasc Surg. 2003;38(6):1411-5. transplantation. Liver Transpl Surg. 1998;4(3):
40. Suzuki S, Nakamura S, Baba S, Sakaguchi 222-5.
S, Ohnuki Y, Yokoi Y, et al. Portal vein 45. Ozdogan M, Gurer A, Gokakin AK,
thrombosis after splenectomy successfully Kulacoglu H, Aydin R. Thrombolysis via
treated by an enormous dosage of fibrinolytic an operatively placed mesenteric catheter
agent in a short period: report of two cases. for portal and superior mesenteric vein
Surg Today. 1992;22(5): 464-9. thrombosis: report of a case. Surg Today. 2006;
41. Wang MQ, Lin HY, Guo LP, Liu FY, 36(9):846-8.
Duan F, Wang ZJ. Acute extensive portal 46. Orr DW, Harrison PM, Devlin J, Karani
and mesenteric venous thrombosis after JB, Kane PA, Heaton ND, et al. Chronic
splenectomy: treated by interventional mesenteric venous thrombosis: evaluation
thrombolysis with transjugular approach. and determinants of survival during long-
World J Gastroenterol. 2009;15(24): term follow-up. Clin Gastroenterol Hepatol.
3038-45. 2007;5(1):80-6.
42. Ferro C, Rossi UG, Bovio G, Dahamane 47. Jona J, Cummins GM Jr, Head HB, Govostis
M, Centanaro M. Transjugular intrahepatic MC. Recurrent primary mesenteric venous
portosystemic shunt, mechanical aspiration thrombosis. JAMA. 1974;227(9):1033-5.

4
PA R T
Practice
Management
T he success of any practice

resides in the ability to accurately capture its activity and collect
revenue with the minimum cost while still maintaining customer
satisfaction. Appropriate coding will allow rapid bill processing,
reduce denials, and lower accounts receivable. Improved cash flow
cannot help but enhance practice profitability.
223

CHAPTER
18
Current Venous Coding

and Billing
gary Burns
Three Phases of Phase 1: Venous

Peripheral Interventional Catheter Placement
Billing and Coding There are a few factors to remember when
There are 3 phases of reporting venous in- reporting venous catheter placement. First,
terventional procedures accurately. They only the most distal placement in the ve-
are venous catheter placement (phase 1), di- nous system is reported. For example, if a
agnostic venogram imaging (phase 2), and right common femoral vein is accessed with
venous interventions (phase 3). There are placement of the catheter to the inferior
specific reporting requirements for accurate vena cava (IVC), only the IVC placement is
representation of peripheral interventional reported. The final destination of the cath-
procedures. Included here is a summary of eter in the IVC includes the initial access
the key components to each of the phases and all catheter movement to its end posi-
with regard to representing venous proce- tion. Second, only the most distal diagnostic
dures accurately and in compliance with catheter or interventional device will be re-
regulators. ported. For example, if a diagnostic catheter
is placed for venography but an intervention-
al device is advanced further in the venous
vascular system, the interventional device
Davies md and Alan B. Lumsden md, eds. will be reported as the most distal catheter
Cardiotext Publishing, ISBN 978-1-935395-22-5. advancement. It then will include all other
225

diagnostic catheter advancements to get it to Venous Catheter Advancement to the

the vessel of intervention.
It is extremely important to note that Vena Cava (Superior or Inferior)
a guidance wire placement or end position When a catheter (diagnostic or interven-
is never reported as a catheter placement. tional) is advanced from the access site to
Also, a sheath is not reported as a catheter the vena cava (inferior or superior), this end
position. Again, only report the final end po- position will prevail and include all pre-
sition of a diagnostic catheter or an interven- vious work. There is a code for vena cava
tional device. One exception here is when a catheter placement (36010). This code
sheath is used for injection of contrast and will include a venous access (36000) and/
venography. It can be classified as a diagnos- or venous access with contrast injection for
tic catheter position. extremity venography (36005). Remem-
ber that the end position of the diagnostic
catheter or interventional device prevails
Venous Access and includes all other catheter work com-
There is a code for access to the venous sys- ponents to this end path. For example, if a
tem (36000).* It represents nothing more right common femoral vein access was ac-
than a needle puncture to the venous system. complished followed by a contrast injection
This excludes dialysis fistula/graft puncture, venography with eventual movement of a
which we will discuss later in this document. inferior vena cava filter, the interventional
If more than one access is created in the ve- device would prevail as the most distal end
nous system, each would be reported sepa- position (36010). This position would in-
rately as its own work component. clude the venous access and the contrast
There is a separate code for access to injection venogram.
the venous system with a contrast injec-
tion for extremity venography (36005). It
represents the venous needle puncture and Selective Venous Catheter
a contrast injection for extremity venogra- Advancement to a Superior Vena
phy. From one access with an extremity ve-
nography contrast injection, you can only Cava or Inferior Vena Cava Branch
report the 36005. It includes the venous When venous diagnostic catheters or in-
access (36000). However, if you access the tervention devices are advanced out of the
right lower extremity for contrast injec- vena cava to branched vessels arising from
tion venogram (36005) and the left lower the vena cava, there are codes to reflect the
extremity for contrast injection venogram complexity of these movements. Any branch
(36005), it is appropriate to report each vessel that arises off of the vena cava (superi-
work component. or or inferior) is defined as a vascular family.
The first branches that arise off of the vena
cava are considered first-order selective cath-
eter advancements (36011). If the catheter is
*Current Procedural Terminology (CPT) copyright 2009 American Medical manipulated past a first-order bifurcation,
Association (AMA). All rights reserved. The AMA assumes no liability for the selective movement is considered second
the data contained herein. CPT is a trademark of the American Medical
Association. Any CPT code listed herein should be referenced to the complete
order branches (36012).
description by the AMA. As with the venous access site (36000)

Current Venous Coding and Billing 227
or venous extremity contrast injection Selective Venous Catheter

(36005) being bundled into the vena cava
placement, the SVC or IVC placement will Advancement to Multiple Branches
be bundled into the selective catheter ad- Within a Vascular Family
vancement to a first-order branch (36011).
Also, when a selective catheter is advanced As discussed earlier, a vascular family is a
to a second-order branch (36012), it is not network of venous vessels that arise from the
appropriate to report the first-order selec- vena cava (superior or inferior). We have also
tive branch (36011). The first-order branch noted that each vascular family with a selec-
(36011), the vena cava placement (36010), tive diagnostic catheter or interventional de-
and the venous access (36000 or 36005) are vice would be reported separately. Once in
all bundled into the second-order end posi- a vascular family a physician will encounter
tion of the diagnostic catheter or interven- the first-order branch (36011) that arises off
tional device placement. of the vena cava. This first-order branch ves-
sel will bifurcate into a second-order branch
(36012). Again, we already established that
Selective Venous Catheter only the end position can be reported.
Advancement to Multiple Superior When a physician works to manipulate
catheters or interventional devices selec-
Vena Cava or Inferior Vena Cava tively into multiple families arising off of the
Branches vena cava, then each is reported separately.
However, it is important to note that any
If the interventional physician manipulates further advancement of a selective catheter
the catheter or interventional device to or interventional device past second-order bi-
multiple branches that arise off of the vena furcations does not have a code to reflect this
cava, each would be reported separately. For higher level of work component. Remember,
example, if a right common femoral access one distal end position path represents only
was accomplished with advancement of the one code assignment. The second-order ad-
diagnostic catheter to the right brachioce- vancement includes all distal placements in
phalic vein (first-order branch off of the a vascular family. For example, if a physician
SVC, 36011) and left brachiocephalic vein accesses the right common femoral vein with
(36011), each would be reported separately. advancement to the inferior vena cava and
Again, remember that the venous access and then superior vena cava to the right brachio-
the vena cava placement would be bundled cephalic vein vascular family with final se-
into the brachiocephalic placements and not lective manipulation to the right basilic vein,
reported separately regardless if diagnostic only the second-order placement (36012) is
venograms (ie, SVC gram) or interventions reported. This represents one pathway and
(ie, IVC filter) were performed. only one end position of a diagnostic cath-
eter placement or interventional device.
A noted exception to the above is when
a physician selectively advances a diagnos-
tic catheter or an interventional device to
multiple second-order branches within a
vascular family, it is appropriate to each end

pathway separately. For example, if a physi- nature of your complex work components.
cian accesses the right common femoral For example, when performing a right leg
vein with advancement to the inferior vena venogram and a left leg venogram from two
cava and then superior vena cava to the right separate venous access sites, both are report-
brachiocephalic vein vascular family with ed (36005 + 36005). However, a payor will
final selective manipulation to the right sub- rarely recognize both access sites with in-
clavian vein and right internal jugular vein, jection codes without a modifier to show 2
each of the 2 separate pathways would be access sites were performed. This is primar-
reported. The second-order right subclavian ily because the codes are relatively generic.
vein (36012) and separate pathway end posi- Code 36005 is a venous access for extremity
tion to the right internal jgular (36012) are contrast injection venogram. The code is not
reported. This represents 2 pathways and 2 specific to right or left leg access and injec-
end positions of a diagnostic catheter place- tion. Thus, a modifier 59 (distinct procedure
ment or interventional device. Again, report or separate access site with contrast injection
both work components. in this case) must be appended. The modifier
It is important to note that only inten- will only need appended to one of the codes
tional catheter positions are reported. that are duplicated (ie, 36005 + 36005-59).
Application of Modifiers for Catheter High-Value Codes with lower-Value Code

When reporting codes of a high-value nature
Placements (second-order selective catheter placement,
To accompany the complex rules listed in 36012) with a lower-valued code (first-order
phase 1, it is important to note that there selective catheter placement, 36011) a modi-
are some key factors relating to payor sub- fier 59 (distinct procedureseparate branch
mission requirements of your work compo- placement) is necessary to report the com-
nents. Any guidance offered here should be plexity of a procedure. The modifier will only
validated with each of your payors. All payor be needed on the lower-valued code (36012
rules for modifiers should be adhered to, as- + 36011-59). For example, if a right common
suring accurate representation of your work. femoral vein access was accomplished and
We can report a few basic guidelines related the catheter was advanced to the right bra-
to phase 1. chiocephalic vein (first-order branch, 36011)
Modifiers are code attachments to fur- and the left subclavian vein (second-order
ther recognize your complex work. It will be branch, 36012separate vascular family) a
necessary to report a modifier when proce- modifier is necessary to reflect the 2 separate
dure codes are duplicated as well as when a vascular families selectively catheterized.
higher value code is reported with a lower-
valued code.
Duplicate Codes
Phase 2: Diagnostic
When reporting the same procedure code Venogram Imaging
twice on a claim, you will invariably need a
modifier appended to one of your codes to There are 3 key components to diagnostic
ensure that a payor recognizes the duplicate venography imaging. First, contrast must be

injected. Second, images must be captured gram, and thus, can be used as an additional
and stored. Third and finally, the physician imaging component if the 3 components for
must document an interpretation of the vas- diagnostic venography imaging are met (ie,
cular vessels. contrast injection, image capture, physician
Guiding injections to ensure that an in- interpretation).
terventional device is in position prior to the
interventional technique is not considered
diagnostic imaging. Road-mapping imaging
Venous Imaging Performed at the
techniques are also not considered diagnos-
tic imaging. Postinterventional follow-up im- Same Time/Same Session as the
aging is considered part of the interventional Venous Intervention
technique and again not diagnostic imaging
because these imaging techniques tend not Guidelines have been established in the
to have revealed the initial diagnostic inter- past few years by payors to limit diagnos-
pretation of the endovascular disease. tic imaging reimbursements at the same
A physician must create an initial dis- time of the interventional procedure. Thus,
ease assessment interpretation of a contrast when a physician performs a diagnostic ve-
injection whether patent, totally occluded, nography at the same time as the interven-
and anything in between to qualify for re- tional procedure the physician must append
porting a code for diagnostic imaging. And a modifier to reveal that venous imaging
remember, a stored image will validate your was not done previously. The modifier es-
work components. tablished to reveal that the venogram was a
Venography reporting is not impacted diagnostic study that leads to the interven-
by the amount of contrast injected or the tion is 59 (distinct procedure). For example,
number of images stored. It is solely based if a right lower extremity unilateral veno-
on the disease assessment of the vessel. gram (75820) was performed and a venous
angioplasty followed at the same session,
the physician must append the modifier 59
Selective Venography to the venography (75820-59) to show that
There are some venography codes that re- the interpretation lead to the interventional
flect the narrative of selective venography in procedure.
the narrative of the procedure. For example, If the procedure is staged where the
selective bilateral renal venography reflects venography was performed at a separate
this narrative. When the narrative of the ve- session than the intervention, the repeated
nography code reflects the selective termi- venogram should not be reported a second
nology, a catheter position within the vessel time. However, if there is a change in the
is necessary to report the imaging procedure. clinical findings, it would be appropriate to
As in the selective bilateral renal venography report the venography with the modifier 59
(75833) procedure, 2 catheter positions are to establish it as a distinct imaging compo-
required within the renal veins (ie, right nent at the time of intervention.
renal vein and left renal vein) to accompany
the imaging code. It is noted that the IVC
gram (75825) is not included in the code
narrative for selective bilateral renal veno-

Phase 3: Venous superior vena cava (35476 + 75978) it would

be established as its own zone of treatment
Interventions and added as a separate work component.
Thus, a physician would report the superior
The third phase of accurate billing and code vena cava, right brachiocephalic, and right
capture is the reporting of venous interven- subclavian as separate work zones for venous
tions (angioplasty, stent, IVC filter place- angioplasty.
ment, IVC filter removal, thrombectomy, It is important to note that each of the
ablation, and dialysis graft/fistula declot). other phases of procedure capture are not in-
Guidelines have been established for each cluded in this interventional technique and
intervention separately, and then a complex should be reported separately. Add the cath-
set when multiple interventional techniques eter placement to the interventional zone as
are used for clinical treatment of venous dis- well as any diagnostic imaging performed to
ease. When reporting most venous interven- assess disease prior to intervention in the ve-
tions, a companion imaging component will nous vessels.
accompany the surgical code assignment.
For example, the venous balloon dilation Modifier Adjustments
is labeled as a surgical work component (ie, As discussed earlier, when a code is du-
35476). There is a supervision and interpre- plicated but reported correctly it must be
tation imaging component that accompanies appended with the modifier 59 (distinct pro-
the surgical work (75978). These 2 compan- cedure) to establish the separate work zones.
ion codes will travel together for each vessel In the above example, the reporting would
treated with balloon dilation. be as follows: 35476 + 75978 (superior vena
cava), 35476-59 + 75978-59 (right brachio-
cephalic), 35476-59 + 75978-59 (right sub-
Venous Angioplasty lcavian). It is important to note that other
Only one procedure code is established for modifiers may be required by payors to
venous angioplasty (35476). However, this further establish these separate work zones.
code would be reported for each venous Our recommendation is offered because we
vessel treated. The venous vessels zones of find it most consistent with coding guide-
treatment are established from venous bifur- lines and payor acceptance across the coun-
cation to bifurcation. It is not reported for try. Also, note that the first set of codes is not
each size of balloon utilized for dilation. It modified (ie, superior vena cava). Only the
is also not reported for each dilation at the following codes that duplicate the initial set
same disease site or each dilation within the are modified.
same vessel zone, but at different disease
sites within a vessel.
For example, if a physician dilates a Venous Stent
right brachiocephalic vein (35476 + 75978) The code established for venous stent is the
and a right subclavian vein (35476 + 75978), same that is used for arterial-based stent
each is reported separately because of the procedures (37205initial vessel treated by
bifurcation from the brachiocephalic to the stent deployment, 37206additional vessel
subclavian with the internal jugular. In fact, treated by stent deployment). Intravascular
if a venous dilation was also performed in the stent is reported in the same manner as ve-

nous angioplasty. Each vessel zone treated and right subclavian as separate work zones
by intravascular stent is reported. It is impor- for venous stent deployment.
tant to note that a physician can only report
the initial vessel treated by stent deployment Modifier Adjustments
(37205) once during an interventional pro- As discussed earlier, when a code is dupli-
cedure. Each vessel after the initial vessel cated but reported correctly, most payors will
treated by stent deployment is considered an require a modifier appended (ie, modifier 59
additional vessel treated (37206). distinct procedure) to establish the separate
The venous vessels zones of treatment work zones. Our recommendation is offered
are established from venous bifurcation to because we find it most consistent with cod-
bifurcation. It is not reported for each size ing guidelines and payor acceptance across
(ie, diameter or length) of stent placed in the the country. It would be appropriate to con-
venous vessel. Nor does it reflect any differ- sult each of your payor guidelines for modi-
ence in the deployment methods (ie, balloon fier submission requirements.
expandable or self-expanding). It is also not It is important to note that each of the
reported for the number of stents placed in other phases of procedure capture are not in-
the same vessel zone. It is reported once for cluded in this interventional technique and
each vessel treated by stent deployment. One should be reported separately. Add the cath-
initial (37205) vessel treated and the remain- eter placement to the interventional zone as
ing vessels treated with stent deployment well as any diagnostic imaging performed to
past separate bifurcations are the additional assess disease prior to intervention in the ve-
vessels (37206). The additional vessel treated nous vessels.
(37206) can be reported multiple times but
only once within a vessel.
There is a supervision and interpreta- Venous Angioplasty Followed by
tion imaging component that accompanies Stent Deployment
the surgical work (75960). These 2 com-
panion codes will travel together for each When combining multiple interventional
vessel treated with stent deployment (37205 techniques like venous angioplasty followed
+ 75960 initial vessel, 37206 + 75960 addi- by stent deployment, documentation is nec-
tional vessel, 37206 + 75960 additional ves- essary to establish each interventional pro-
sel, etc). cedure as its own separate work component.
For example, if a physician places a stent For instance, if the balloon angioplasty yield-
in the right brachiocephalic vein (37205 + ed a suboptimal recoiled residual stenosis, a
75960) and a right subclavian vein (37206 + predilation and postdilation percentage as-
75960), each is reported separately because sessment of the disease would reveal the rea-
of the bifurcation from the brachiocephalic son for further intervention with the stent.
to the subclavian with the internal jugular. Also, if a balloon angioplasty resulted in a
In fact, if a venous stent was also performed vessel dissection, then the multiple inter-
in the superior vena cava (37206 + 75960), ventional techniques are established by the
it would be established as its own zone of result of the initial procedure. In each of the
treatment and added as a separate work com- balloon results (ie, residual stenosis, dissec-
ponent. Thus, a physician would report the tion) both interventions are reported. It is not
superior vena cava, right brachiocephalic, appropriate to modify and angioplasty from

a stent deployment. They are independent components (venous angioplasty + stent

procedures that are commonly not bundled deployment).
by payors. Any postdilation performed within a
For example, a proximal right subcla- stent at the same session as the deployment
vian vein 90% stenosis was dilated with an is considered part of the initial stent de-
angioplasty balloon (35476 + 75978). On ployment. Obviously, if balloon dilation is
follow-up imaging a 50% residual stenosis performed within in-stent stenosis then the
was identified. After several attempts with angioplasty can be reported as its own work
multiple dilations at the same site within this component. It is often noted that there is
vessel (no additional codes), a 50% residual no suboptimal result to a stent deployment
stenosis remained. This was followed by a when a balloon is needed to further expand
stent deployment (37205 + 75960). Both pro- a stent.
cedures are reported with their companion A physician can only report both work
supervision and interpretation components. components (angioplasty + stent) when there
No modifiers are necessary to separate these is a suboptimal result of angioplasty and the
individual work components. desired outcome was not accomplished. The
Along with the predilation and postdi- dissection or residual stenosis (best reported
lation percentages, it is also good documen- in pre- and postpercentage) reveals the dis-
tation principles to establish the location tinct work components and meets the sepa-
of the disease within a vessel as proximal, rate guidelines for reporting angioplasty and
midlevel, or distal. These documentation stent.
principles assist in revealing the complexity
of treatment in the venous system without
the visual reference of a set of venography Inferior Vena Cava Filter Placement
images. When placing an IVC filter all 3 phases
It is important to note that each of the (placement, imaging, and intervention)
other phases of procedure capture are not in- can be reported. Venous access (ie, either
cluded in this interventional technique and jugular access or common femoral vein ac-
should be reported separately. Add the cath- cess) is accomplished with placement of
eter placement to the interventional zone as the diagnostic catheter to the IVC (36010)
well as any diagnostic imaging performed to for venography imaging. Remember, sizing
assess disease prior to intervention in the ve- of the vena cava and mapping the location
nous vessels. of the renal veins are not considered part of
diagnostic imaging. Thus, a physician must
Predilation and/or Postdilation of Stent establish a diagnostic assessment of throm-
Deployment bus or false passages in the IVC. A normal
Any predilation performed to be able to IVC is acceptable as long as the documen-
place a stent at a venous stenosis is consid- tation reflects no thrombus disease present
ered part of the stent deployment and not or duplicated IVCs. When a diagnostic IVC
a separate work component. Remember, study is reported, a modifier 59 (distinct pro-
a venous angioplasty followed by a docu- cedure) is necessary to establish it as its own
mented suboptimal result (ie, residual ste- work component (75825-59).
nosis or dissection) and stent deployment Once the imaging is established and
is considered as their own separate work the physician deploys the vena cava filter,

2 work components are established to re- thrombectomy, a separate work component

port the procedure (37620 + 75940). Fol- can be established (37188). All fluoroscopic
low-up venography after the placement of guidance to assist in the thrombectomy is
the filter is considered part of the assurance not reported separately.
of accurate deployment and not reported It is noted that these are not the pro-
as a separate work component. Also, the cedures utilized for dialysis/fistula throm-
reporting of a permanent or temporary re- bectomy interventions. Also, there is no
movable filter does not reflect any different supervision and interpretation component
work component. attached as a component to this procedure.
Inferior Vena Cava Filter (IVC) Endovenous Ablation

Endovenous ablation of incompetent veins
Removal in the extremity is divided into 2 interven-
When an existing IVC filter is no longer nec- tional techniques (ie, radiofrequency and
essary to capture venous thrombosis, some laser). Each are further divided into first ves-
of the 3 phases (placement, imaging, and in- sel treated and second with all other vessels
tervention) may be reported. Venous access treated in an extremity. All imaging guid-
(ie, usually jugular access) is accomplished ance to treat the incompetent veins is not
with placement of the filter capture device to considered a separate work component.
the IVC (36010). Remember, if a diagnostic
venogram is not accomplished but contrast is radiofrequency Ablation
injected to locate the position of the filter, no Radiofrequency endovenous ablation first
diagnostic imaging is reported. vessel (36475), second, and subsequent veins
Once the location is established and treated in the same extremity through sepa-
the physician retrieves the vena cava filter, rate access sites (36476) reflects the further
2 work components are established to report division within this interventional tech-
the procedure (37203 + 75961, considered nique.
foreign body retrieval). Follow-up venogra-
phy after the removal of the filter is not con- laser Ablation
sidered a separate work component. Laser endovenous ablation first vessel
(36478), second, and subsequent veins treat-
ed in the same extremity through separate
Venous Thrombectomy access sites (36479) reflects the further divi-
Venous thrombectomy includes any infusion within this interventional technique.
sion/injection of thrombolytic agents during Also, there is no supervision and inter-
the mechanical removal of thrombus in the pretation component attached as a compo-
venous system. It also is not delineated by nent to this procedure.
vessel as with angioplasty and stent deploy-
ment. Any mechanical thrombectomy dur-
ing one date of service is all included in one Conclusion
procedure work component (37187). When
the calendar date changes to a subsequent Venous physician work components are
day, and a mechanical device is utilized for complex. Most venous vascular procedure

reporting can be divided into the 3-phase physician must have a high-quality support
approach outlined in this document (cath- system (ie, certified medical coders) that
eter placement, diagnostic imaging, and ve- understands the guidelines for each payor.
nous intervention). The above is an attempt Without a high-quality support system, a
to guide a physician though the maze of complex venous procedure involving mul-
complex procedure reporting requirements, tiple work components may reflect payments
as well as offer suggestions for documenta- that do not reveal all the procedures per-
tion quality. In no way is it possible to list formed by the physician.
all scenarios for venous procedures. Each

Index
Figures are indicated by f
and tables by t following the page number.
abdominal cancer, 207 anticoagulation, 6775. See also heparin

abdominal pain, 210 ACCP guidelines (2008), 5859, 6061
abdominal trauma, 206 anticoagulant drugs, 1315
access site thrombosis, 126 approved agents for VTE, 7t
activated partial thromboplastin time (aPTT), 6, for axillosubclavian venous thrombosis, 168
71, 73 complications, 89
activated protein C (APC), 22, 207 direct thrombin inhibitors, 7374
alteplase, 8081 indirect factor Xa inhibitors, 7273
ambulation, early, 6162 for mesenteric venous thrombosis, 21516
ambulatory venous pressure (AVP), 105 natural anticoagulants, 22
amediplase (prourokinase), 81 new anticoagulants, 7475, 74f
American College of Chest Physicians (ACCP) oral agents, 6770, 68f
DVT treatment guidelines (2008), 5763 parenteral anticoagulants, 7074
Amplatz Thrombectomy Device, 14243, 142f postprocedure, 100
angiography prolonged, in cancer patients, 9
magnetic resonance (MRA), 166, 167f, 213 for recurrent VTE prevention, 1, 68
mesenteric, 21314 systemic, for DVT, 6
pulmonary, 51 antiphospholipid antibodies, 2078
AngioJet rheolytic thrombectomy catheter antiplatelet agents, 168
(MEDRAD) , 90, 91f, 9899, 218 antithrombin (AT), 22
AngioJet Xpeedior thrombectomy device antithrombin III deficiency, 207
(MEDRAD), 14547, 146f Antithrombotic Therapy for Venous
Angiomax (bivalirudin), 73 Thromboembolic Disease (ACCP), 57
angioplasty, venous, 23032. See also balloon APC (activated protein C), 22, 207
angioplasty apixaban, 14
235

236 Index
aPTT (Activated partial thromboplastin time), 6, bivalirudin (Angiomax), 15, 7374
71, 73 bleeding, as anticoagulant complication, 89
argatroban, 14, 15, 73 blood coagulation tests, 166
Arrow-Trerotola thrombectomy device, 143 blood gas analysis, 4849
arterial thrombosis, 27 bowel necrosis, 216
arteriography, spiral CT, 5051 bradyarrhythmias, 14647
Aspirex PE Catheter (Straub Medical), 147, 147f brain natriuretic peptide (BNP), 26, 26t
ASVT. See axillosubclavian venous thrombosis Budd-Chiari syndrome, 19697
(ASVT)
AT (antithrombin), 22
ATTRACT trial (Acute Venous Thrombosis: calf pain, 46
Thrombus Removal with Adjunctive cancer
Catheter-Directed Thrombolysis), 87, abdominal, 207
9293, 112 intrathoracic, 194
AVP (ambulatory venous pressure), 105 mesenteric venous thrombosis and, 207
axillosubclavian venous thrombosis (ASVT) prolonged anticoagulation in, 9
anticoagulation therapy, 157, 168 VTE risk and, 45
balloon angioplasty for, 16869 catheter-directed thrombolysis (CDT)
blood coagulation tests, 166 ACCP guidelines (2008), 5960
clinical presentation, 165 catheter placement, coding and billing,
diagnosis, 16566 22528
general principles, 156 complications, 96
laboratory tests, 16566 for DVT, 12, 9596
natural history, 157 for iliofemoral DVT, 183
nonsurgical management, 169, 169t indications for, 107, 136
observational care, 15559 for pulmonary embolism, 137
Paget-Schroetter syndrome, 199, 200f, 201 technique, 8486, 85f
pathophysiology, 156, 163, 164f, 165 thrombolytic agents for, 95
primary vs. secondary, 34, 35, 155 Celect IVC filter (Cook), 125, 126
radiologic imaging, 166, 167f cell adhesion molecules, 21
surgical treatment central venous disease, overview, 17980
indications and protocols, 16971 central venous stenosis
selection of approach, 17172 about, 193
SVC filters for, 12021 Budd-Chiari syndrome, 19697
therapeutic options, 15859 dialysis access-related, 201
thoracic outlet decompression, 155, 157, infradiaphragmatic disease, 19596
17276, 173f, 175f intrathoracic disease
thrombolytic therapy, 157, 168 benign, 19495
treatment goals, 166, 168 malignant, 194, 195f
management of, 193201
May-Thurner syndrome, 198, 199f
balloon angioplasty Paget-Schroetter syndrome, 199, 200f, 201
for acute DVT, 2008 ACCP guidelines, renal vein thrombosis, 19798
5960 chest X-ray, 49, 49f
for axillosubclavian venous thrombosis, chronic venous insufficiency, 11
16869 clinical trials
for pulmonary embolism fragmentation, 143 ATTRACT trial, 87, 9293, 112
bariatric surgery, IVC filters in, 119 PIOPED (Prospective Investigation of
bed rest, for acute DVT, 58 Pulmonary Embolism Diagnosis), 47, 50,
billing. See coding and billing 51
Birds Nest Filter (Cook), 122t, 123f PREVENT trial, 8

Index 237
coagulation cascade, 67, 68f, 74 cancer and risk of, 45

coagulation factors, 1920 complications, 1, 1034
coding and billing, 22534 diagnosis, 3738, 4547, 106
diagnostic venogram imaging, 22829 early thrombus removal rationale, 1034
venous catheter placement, 22528 endovascular interventions, 95101
advancement to vena cava (superior or epidemiology, 3334
inferior), 226 future medical treatments, 1315
application of modifiers, 228 hypercoagulable states in, 34, 35t
selective advancement to multiple imaging modalities, 48
branches within vascular family, 22728 incidence, 1
selective advancement to multiple vena indications for intervention, 1067
cava branches, 227 inflammatory response, 106
selective advancement to vena cava invasive therapies, 11
branch, 22627 lower limb, 3334, 35 (See also iliofemoral
venous access, 226 deep venous thrombosis)
venous interventions, 23034 management, 2, 3840, 89
endovenous ablation, 233 natural history, 3638
IVC filter placement, 23233 nonpharmacological treatment, 11
IVC filter removal, 233 pathophysiology, 3436, 1046
laser ablation, 233 percutaneous mechanical thrombectomy,
radiofrequency ablation, 233 8993, 97100
venous angioplasty, 230 postprocedure anticoagulation, 100
venous angioplasty followed by stent prevention, 3839
deployment, 23132 proximal propagation of, 3637
venous stent, 23031 pulmonary embolism management, 3940
venous thrombectomy, 233 risk factors, 2425, 34, 36t, 46
compression stockings, 11, 6162 therapeutic algorithm, 8384, 83f
compression ultrasonography, 48 thoracic outlet decompression, 39
computed tomography (CT) thrombolysis for, 1112, 40
angiography, 1 catheter-directed, 1112, 8486, 85f,
contrast-enhanced, 166 9596
multidetector, 21213, 213f pharmacological, 7987
spiral, pulmonary arteriography, 5051 ultrasound-accelerated, 9697, 97f,
venography, indirect, 51 101f
Consortium on Outcomes Research and upper extremity (See axillosubclavian venous
Education for Thoracic Outlet Syndrome thrombosis (ASVT)
(CORE-TOS), 159 Doppler flow ultrasonography, 212
contact activation system, 2223, 67 drotrecogin alfa, 75
contrast venography, 48, 166 DU-176b (direct factor Xa inhibitor), 75
Crockett syndrome. See May-Thurner syndrome duplex imaging, 16566
CT. See computed tomography (CT) DVT. See deep venous thrombosis (DVT)
dabigatran, 14, 75 early ambulation with compression, 6162

dalteparin (Fragmin), 61, 72 echocardiography, 136
D-dimers effort thrombosis. See axillosubclavian venous
assays, 46, 5253 thrombosis (ASVT)
as VTE biomarker, 2526, 26t EKOS EkoSonic system, 96, 97f
deep venous thrombosis (DVT) EKOS EndoWave system, 91, 92f, 9697, 97f
about, 34 electrocardiogram (ECG), 4849, 49f
ACCP treatment guidelines (2008), 5763 embolectomy, surgical, 137

238 Index
embolectomy devices, 140t heparin
Amplatz Thrombectomy Device, 14243, for axillosubclavian venous thrombosis, 168
142f low-molecular-weight (LMWH)
Arrow-Trerotola device, 143 ACCP guidelines (2008), 60
Aspirex PE Catheter, 147, 147f contraindications, 72
Greenfield device, 141, 141f dosage, 72
hydrodynamic thrombectomy devices, for DVT, 38
14547 efficacy vs. unfractionated heparin, 7172
suction embolectomy, 145 pharmacodynamics, 71
thrombus fragmentation catheters and prolonged use in cancer patients, 9
balloons, 14344, 144f for VTE anticoagulation, 6, 1314
endothelium for mesenteric venous thrombosis, 21516
hemostasis and, 22f, 2324 postthrombotic syndrome and, 8283
injury and VTE risk, 4445 unfractionated (UFH), 13
endovenous ablation, 233 ACCP guidelines (2008), 60
enoxaparin (Lovenox), 72 anticoagulant effects, 7071
aPTT monitoring, 6, 71
dosage, 7071, 71t, 72t
factor II (thrombin), 2021 for DVT, 6, 38
factor V Leiden mutation, 207 indications and contraindications, 71
factor VIII, 25, 26t heparin-induced thrombocytopenia (HIT), 9,
factor IXa inhibitors, 75 1415
factor Xa inhibitors, 14, 7273, 75 Homans sign, 46
fibrinolysis, 2223, 7980, 81f hormone therapy, 207
flovagatran, 75 Hydrolyzer device, 145
fondaparinux hypercoagulable states
ACCP guidelines, 60 in DVT, 34, 35t
dosage, 73, 73t mesenteric venous thrombosis and, 2078
for DVT prophylaxis, 14 VTE risk and, 45
pharmacokinetics and efficacy, 7273 warfarin therapy risks in, 78
Fragmin (dalteparin), 61, 72
idraparinux, 14, 75
G2 X IVC filter (CR Bard), 123, 123f, 124f, 125 iliac vein compression syndrome, 18283, 198. See
gastrointestinal bleeding, 210 also May-Thurner syndrome
Glidewire guidewire (Terumo), 99 iliocaval occlusive disease, 19495
Greenfield embolectomy device (Boston iliofemoral deep venous thrombosis, 18189
Scientific), 141, 141f about, 181
Greenfield IVC filter (Boston Scientific), 127, ACCP guidelines (2008), 5960, 182
128f, 129 catheter-directed thrombolysis, 183, 18586
Greenfield Titanium IVC filter (Boston Scientific), diagnosis, 106
123f early thrombus removal, rationale, 1034
Gnther Tulip IVC filter (Cook), 123f, 124f, 125, endovascular management, 18288
126 iliac vein compression, 18283, 198
indications for intervention, 1067
interventions and clinical outcomes, 183, 184t
Hamptons hump, 49, 49f medical therapy, 18182
hematemesis, 210 natural history vs. distal DVT, 104
hemodialysis, access-related venous stenosis in, pathophysiology, 1045
201 pharmacomechanical thrombectomy, 186,
hemostasis, 22f, 2324 187f, 188

Index 239
postthrombotic morbidity, 11 magnetic resonance imaging (MRI), 213

surgical thrombectomy, 10810, 182 malignancies. See cancer
treatment algorithm, 113f, 183, 185f May-Thurner syndrome, 179, 182, 183, 198, 199f
immobility, and VTE risk, 44 mesenteric venous thrombosis (MVT)
inferior vena cava (IVC) filters, 11732 about, 205
about, 11718 bowel necrosis in, 216
available devices, 122t, 123f clinical presentation, 21011
complications, 12627, 128t129t, 129, conditions associated with, 2067, 206t
130f diagnostic workup
contraindications, 11920 Doppler flow ultrasonography, 212
implantation magnetic resonance angiography, 213
coding and billing, 23233 magnetic resonance imaging, 213
techniques, 12126 multidetector computed tomography,
incomplete opening of, 127, 128f 21213, 213f
indications, 117, 11820 radiography, 21112, 212f
intravenous ultrasonography for placement, selective mesenteric angiography, 21314
126 etiology, 2068
migration of, 127 laboratory tests, 211
prophylactic use, 119 management
with pulmonary embolectomy, 148 algorithm, 214, 215f
for pulmonary embolism prevention, 131 anticoagulation, 21516
removal endovascular interventions, 21719
coding and billing, 233 exploratory laparotomy, 21617
procedures, 124f, 12526 intra-arterial vasodilator infusion, 21819
retrievable, 118, 120, 12930, 13132 principles, 21415
in trauma patients, 13031 supportive measures, 219
inferior vena cava (IVC) occlusion, 19596, 196f surgical thrombectomy, 21617
Innohep (tinzaparin), 72 outcomes, 21920
international normalized ratio (INR), 6 pathophysiology, 20810, 208f
microparticles (MPs), 21
monteplase, 82
lanoteplase, 82 MRA (magnetic resonance angiography), 166,
laparotomy, exploratory, 21617 167f, 213
laser ablation, coding and billing, 233 MRI (magnetic resonance imaging), 213
lepirudin (Refludan), 1415, 73 MVT. See mesenteric venous thrombosis (MVT)
liver transplantation, 197 myeloproliferative disorders, 207
Lovenox (enoxaparin), 72
Lowenbergs sign, 46
lower extremity, DVT of. See also iliofemoral deep NAPc2 (anticoagulant), 74
venous thrombosis non-small cell lung cancer, 194
management, 3839
pathogenesis, 3334, 35
low-molecular-weight heparin (LMWH). See obesity, morbid , IVC filters in, 119
under heparin odiparcil, 75
LY-517717 (direct factor Xa inhibitor), 75 Opt Ease IVC filter (Cordis), 122t, 125, 125f
Option IVC filter (Angiotech), 122t
otamixaban, 75
magnetic resonance angiography (MRA), 166,
167f, 213
magnetic resonance direct thrombus imaging, Paget-Schroetter syndrome. See axillosubclavian
5152 venous thrombosis (ASVT)

240 Index
PAI-1 (plasminogen activation inhibitor), 15, 23 prourokinase (amediplase), 81
pamiteplase, 81, 82 PRT054021 (direct factor Xa inhibitor), 75
pancreatic cancer, 207 P-selectin, 26, 26t
papaverine, 216, 21819 P-selectin glycoprotein ligand 1 (PSGL-1), 15,
PE. See pulmonary embolism (PE) 2122
pegmusirudin, 75 P-selectin inhibitors, 15
peptic ulcer disease, 156 PTS. See postthrombotic syndrome (PTS)
percutaneous mechanical thrombectomy (PMT) pulmonary angiography, 51
ACCP guidelines (2003), 60 pulmonary arteriography, spiral CT, 5051
for acute DVT, 8993 pulmonary embolectomy, percutaneous, 13548
adjunctive procedures, 9192, 100 catheter-directed thrombolysis, 137, 148
devices for, 9091, 91f, 92f, 98100, 99f complications, 14748
efficacy, 1213, 9798 devices, 140t, 14147 (See also embolectomy
for iliofemoral DVT, 186, 187f, 188 devices)
indications for, 107 indications for, 13536
for mesenteric venous thrombosis, 218 initiation of therapy, 148
outcomes, 9293 intravenous thrombolytic therapy, 13637,
strategies, 9092 148
pharmacological thrombolysis, 7987 IVC filter placement with, 148
pharmacomechanical thrombolysis. See suction embolectomy with guide catheter/
percutaneous mechanical thrombectomy sheath, 145
(PMT) surgical embolectomy, 13738 (See also
phlegmasia cerulea dolens, 103 thrombectomy, surgical)
pigtail catheters, 144, 144f technique, 13841, 139f
PIOPED (Prospective Investigation of Pulmonary pulmonary embolism (PE)
Embolism Diagnosis) trial, 47, 50, 51 anticoagulant therapy, 3940, 148
plasmin, 22 clinical diagnosis, 4647, 47f
plasminogen activation, 2223, 7980 diagnostic tests, 4851
plasminogen activation inhibitor (PAI-1), 15, 23 differential risk vs. DVT, 2425
platelet activation, 1920 as DVT complication, 1
platelet adhesion, 1920 experimental, 24
PMT. See percutaneous mechanical future medical treatments, 1315
thrombectomy (PMT) incidence, 135
pneumatic compression, intermittent, 62 indications for interventions, 13536
polycythemia vera, 207 invasive therapies, 11
portal vein thrombosis, 19697 IVC filters preventing, 131
postthrombotic syndrome (PTS) mortality, 33
ambulatory venous pressure and, 105
anticoagulation complications, 12
as DVT complication, 1, 155 radiofrequency ablation, coding and billing, 233
iliofemoral DVT and, 183 radiography, and mesenteric venous thrombosis,
incidence, 5 21112
morbidity, 58 RB006 (factor IXa inhibitor), 75
prevention of, 39, 8283 Refludan (lepirudin), 1415, 73
power-pulse spray technique, 9899 renal vein thrombosis, 19798
PREVENT trial, 8 Retavase (tissue plasminogen activator), 22, 80, 95
Prospective Investigation of Pulmonary Embolism reteplase (rPA), 80, 81, 82t
Diagnosis (PIOPED) trial, 47, 50, 51 rivaroxaban, 14, 75
protein C deficiency, 207
protein S deficiency, 207
prothrombin gene mutations, 207 Simon Nitinol (CR Bard), 122t

Index 241
soft tissue sarcoma, retroperitoneal, 194 contraindications, 84

spiral computed tomography, pulmonary defined, 80
arteriography, 5051 for DVT, 40, 83f
SR123781A (indirect factor Xa inhibitor), 75 indications, 8284
SSR12517E (indirect factor Xa inhibitor), 75 intravenous, for pulmonary embolism, 13637
SSR126517 (anticoagulant), 14 for mesenteric venous thrombosis, 21718
staphylokinase, 82 monitoring, 86
streptokinase, 8081, 82t outcomes, 8687
subclavian vein occlusion, 201 pharmacological, 7987
suction embolectomy, 145 pharmacomechanical (See percutaneous
superior vena cava (SVC) mechanical thrombectomy (PMT)
filters, 12021 physiology, 7980
obstruction, 19495, 195f postprocedure anticoagulation, 100
rationale for, 1112
techniques, 8990
TAT (thrombin-antithrombin) complex, 25, 26, 26t thrombolytic agents, 8082, 82t
tenecteplase, 81, 82, 95 ultrasound-accelerated, 9697, 98f, 101f
TFPI (tissue factor pathway inhibitor), 22 thrombolytic therapy. See thrombolysis
thoracic outlet decompression thrombomodulin, 21
for axillosubclavian venous thrombosis, 155, thrombophilias, inherited, 25, 25t
157 thrombosis, arterial, 27. See also venous
for DVT, 39 thrombosis
technique, 17276, 173f, 175f thymoma, 194
thoracic outlet syndrome. See also axillosubclavian Tifacogin, 74
venous thrombosis (ASVT) tinzaparin (Innohep), 72
DVT and decompression of, 39 tissue factor, 1920
overview, 153 tissue factor pathway, 67
venous, pathophysiology, 163, 164f, 165 tissue factor pathway inhibitor (TFPI), 22
thrombectomy, surgical. See also percutaneous tissue plasminogen activator (tPA), 22, 80, 95
mechanical thrombectomy (PMT); transjugular intrahepatic portosystemic shunts
pulmonary embolectomy, percutaneous (TIPS), 197
ACCP guidelines (2008), 59 Trap Ease IVC filter (Cordis), 123f
complications and results, 110 trauma, abdominal, 206
evidence-based recommendations, 11112 Trellis-8 infusion system (Covidien), 90, 91f,
for iliofemoral DVT, 18182 99100, 99f
for mesenteric venous thrombosis, 21617 troponin, 26, 26t
new developments, 11011
rationale, 1034
technique, 10810 ultrasonography
venous, 13, 59, 1078 compression, 48
thrombin (factor II), 2021 Doppler flow, 212
thrombin activatable fibrinolysis inhibitor (TAFI), intravenous, for IVC filter placement, 126
80 ultrasound-accelerated thrombolysis, 9697, 98f,
thrombin inhibitors, 1415, 75 101f
thrombin-antithrombin complex (TAT), 25, 26, 26t unfractionated heparin (UFH). See under heparin
thrombocytopenia upper extremity deep vein thrombosis. See
essential, 207 axillosubclavian venous thrombosis
heparin-induced, 9, 1415 (ASVT)
thrombolysis. See also catheter-directed urokinase, 8081, 82t, 95
thrombolysis urokinase-type plasminogen activator (uPA), 22, 80
for axillosubclavian venous thrombosis, 168

242 Index
valvular incompetence, distal, 105 deep venous thrombosis; mesenteric
vasodilators, 21819 venous thrombosis (MVT)
VenaTech LP IVC filter (Braun), 123f commonalities with arterial thrombosis, 27
venography endothelium and hemostasis, 22f, 2324
coding and billing, 22829 fibrinolysis, 2223
contrast, 48, 166 inflammation and, 2122
indirect CT, 51 natural anticoagulants in, 22
venous angioplasty, coding and billing, 23032 pathophysiology, 1924, 46
venous catheter placement, coding and billing, pathways, 1921, 20f
22528 ventilation-perfusion (V/Q) scan, 1, 4950, 50f
venous disorders, acute, 12 Virchows triad, 36t, 44
venous hypertension, ambulatory, 11 vitamin K antagonists (VKAs). See also warfarin
venous insufficiency, chronic, 11 as anticoagulants, 6770
venous interventions, coding and billing, 23034 dalteparin vs., for VTE, 61
venous stasis, 35 for DVT prophylaxis, 38
venous stenosis, central. See central venous von Willebrand factor (vWF), 20
stenosis V/Q (ventilation-perfusion) scan, 1, 4950, 50f
venous stent, coding and billing, 23031 VTE. See venous thromboembolism (VTE)
venous thrombectomy, coding and billing, 233
venous thromboembolism (VTE)
about, 4344 warfarin (Coumadin)
approved anticoagulant agents, 7t contraindications, 70
biomarkers, 2527, 26t dosage, 70, 71t
clinical diagnosis, 4547 food interactions, 68, 69f, 70
early anticoagulation, 6061 with heparin, 70
future medical treatments, 1315 in hypercoagulable states, 78
incidence, 5, 43 indications for, 6768
inherited thrombophilias, 25, 25t length of therapy, 78
IVC filters for, 11819 as oral anticoagulant, 6770
objective testing, 4748 for recurrent VTE prevention, 68
pathophysiology, 4445 in unprovoked VTE, 8
recurrent, prevention of, 1, 68
risk factors, 25, 4445, 53
therapy, overview, 56 ximelagatran, 73
unprovoked, warfarin therapy in, 8
venous thrombosis (VT). See also axillosubclavian
venous thrombosis (ASVT); iliofemoral YM-150 (direct factor Xa inhibitor), 75


Venous Thromboembolic Disease

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Venous Thromboembolic Disease

Uploaded by

Copyright:

Available Formats

LD2 VTD Nearly Final Pages.

indd 8 5/13/11 10:27 AM

LD2 VTD Nearly Final Pages.indd 1 5/13/11 10:27 AM

The Contemporary Endovascular Management series is designed to be a focused,

Volume 1 Chronic Venous Insufficiency

Volume 2 Venous Thromboembolic Disease

Volume 3 Dialysis Access: Creation and Maintenance

Volume 4 Dialysis Access: Special Cases and Complications

Please visit www.cardiotextpublishing.com for more information about this series.

LD2 VTD Nearly Final Pages.indd 2 5/13/11 10:27 AM

Mark G. Davies MD, PhD, MBA

Daynene Vykoukal PhD assistant editor

LD2 VTD Nearly Final Pages.indd 3 5/13/11 10:27 AM

Any updates to this book may be found at:

Cover design by Brad Norr Design

Library of Congress Control Number: 2011924295

LD2 VTD Nearly Final Pages.indd 4 5/13/11 10:27 AM

Part 1: Venous Thromboembolic Disease

1. Overview of Therapy for Venous Thromboembolic Disease 5

LD2 VTD Nearly Final Pages.indd 5 5/13/11 10:27 AM

5. Treatment of Acute Deep Venous Thrombosis:

Part 2: Thoracic Outlet Syndrome

13. Primary Axillosubclavian Venous Thrombosis: Observational Care 155

Part 3: Central Venous Disease

15. Iliofemoral Deep Venous Thrombosis 181

LD2 VTD Nearly Final Pages.indd 6 5/13/11 10:27 AM

16. Management of Central Venous Stenosis 193

Part 4: Practice Management

18. Current Venous Coding and Billing 225

LD2 VTD Nearly Final Pages.indd 7 5/13/11 10:27 AM

Alan B. Lumsden MD Chairman, Department of Cardiovascular Surgery; Medical Director,

LD2 VTD Nearly Final Pages.indd 9 5/13/11 10:27 AM

Hosam F. El Sayed MD, RVT Department of Cardiovascular Surgery, Methodist DeBakey

Mark H. Meissner MD Professor of Surgery, University of Washington School of

Erin H. Murphy Division of Vascular and Endovascular Surgery, Department of

LD2 VTD Nearly Final Pages.indd 10 5/13/11 10:27 AM

Elina Quiroga MD Division of Vascular Surgery, University of Washington, Seattle,

Eduardo Ramacciotti MD, PhD Section of Vascular Surgery, Department of Surgery,

Michael J. Reardon MD Department of Cardiovascular Surgery, Methodist DeBakey

Saher Sabri MD Assistant Professor of Radiology, Division of Vascular and Interventional

Claudio J. Schnholz MD Heart & Vascular Center, Interventional Radiology, Medical

Robert W. Thompson MD Professor of Surgery, Vascular Surgery, Radiology and Cell

Thomas W. Wakefield MD S. Martin Lindenauer Professor of Surgery, Section Head of

LD2 VTD Final Pages.indd 11 5/19/11 12:17 PM

LD2 VTD Nearly Final Pages.indd 13 5/13/11 10:27 AM

LD2 VTD Nearly Final Pages.indd 14 5/13/11 10:27 AM

LD2 VTD Nearly Final Pages.indd 15 5/13/11 10:27 AM

A cute venous disorders comprise the

LD2 VTD Nearly Final Pages.indd 1 5/13/11 10:27 AM

LD2 VTD Nearly Final Pages.indd 2 5/13/11 10:27 AM

LD2 VTD Nearly Final Pages.indd 3 5/13/11 10:27 AM

LD2 VTD Nearly Final Pages.indd 4 5/13/11 10:27 AM

V enous thomboembolism (VTE), a

LD2 VTD Nearly Final Pages.indd 5 5/13/11 10:27 AM

LD2 VTD Nearly Final Pages.indd 6 5/13/11 10:27 AM

Table 1.1. FDA-ApprovedAnticoagulantAgentsforTreatmentofVTE

UFH IV 80 U/kg initial bolus, followed Adjusted to achieve aPTT 1.5

UFH SQ 333 IU/kg SQ initially, No. However, reserved

Dalteparin SQ 100 mg/kg b.i.d. or No

Tinzaparin SQ 175 IU/kg q.d. No

Fondaparinux SQ 7.5 mg q.d. No

farin-induced skin necrosis, usually on the with shorter administration (3 months) of

LD2 VTD Nearly Final Pages.indd 7 5/13/11 10:27 AM

LD2 VTD Nearly Final Pages.indd 8 5/13/11 10:27 AM