Diagnosis And Management Of March 2015

Volume 17, Number 3

Deep Venous Thrombosis In Authors

Kar-mun C. Woo, MD

The Emergency Department

Assistant Professor, Department of Emergency Medicine, Mount Sinai
Beth Israel, New York, NY
Jacob K. Goertz, MD, RDMS
Assistant Professor, Department of Emergency Medicine, Mount Sinai
Abstract Beth Israel, New York, NY

Peer Reviewers
Although the clinical presentations of deep venous thrombosis are Steven A. Godwin, MD, FACEP
notoriously subtle and nonspecific, risk stratification tools such Professor and Chair, Department of Emergency Medicine; Assistant
Dean for Simulation Education; University of Florida College of
as the Wells clinical model have improved the efficiency of the Medicine-Jacksonville, Jacksonville, FL
diagnostic evaluation. The emergency clinician may be guided Christopher R. Tainter, MD, RDMS
down several pathways, including D-dimer assays and/or ul- Department of Emergency Medicine, Department of Anesthesiology,
Division of Critical Care, University of California-San Diego, La Jolla,
trasonography. New oral anticoagulants offer alternatives to the CA
traditional heparins and vitamin K antagonists in the treatment of
CME Objectives
deep venous thrombosis. This review examines the current litera-
Upon completion of this article, you should be able to:
ture, evidence, and guidelines in the diagnosis and management of
1. Risk stratify a patient with suspected first-time DVT for pretest
deep venous thrombosis. It also explores some of the controversies probability of disease.
and developments regarding risk stratification, adjusted D-dimer 2. Choose appropriate diagnostic testing for patients with
thresholds, special populations, isolated distal deep venous throm- suspected DVT.
bosis, upper extremity deep venous thrombosis, outpatient treat- 3. Describe the anticoagulant regimens used to treat patients with a
confirmed diagnosis of DVT.
ment, and the new oral anticoagulants.
Prior to beginning this activity, see “Physician CME Information” on the
back page.

Editor-In-Chief
Andy Jagoda, MD, FACEP
Professor and Chair, Department of
Emergency Medicine, Icahn School
of Medicine at Mount Sinai, Medical
Director, Mount Sinai Hospital, New
York, NY
Associate Editor-In-Chief
Kaushal Shah, MD, FACEP
Associate Professor, Department of
Emergency Medicine, Icahn School
of Medicine at Mount Sinai, New
York, NY
Editorial Board
William J. Brady, MD
Professor of Emergency Medicine
and Medicine, Chair, Medical
Emergency Response Committee,
Medical Director, Emergency
Management, University of Virginia
Medical Center, Charlottesville, VA
Calvin A. Brown III, MD
Director of Physician Compliance,
Credentialing and Urgent Care
Services, Department of Emergency
Medicine, Brigham and Women's
Hospital, Boston, MA
Mark Clark, MD
Assistant Professor of Emergency
Medicine, Program Director,
Emergency Medicine Residency,
Mount Sinai Saint Luke's, Mount
Sinai Roosevelt, New York, NY
Peter DeBlieux, MD James Damilini, PharmD, BCPS
Professor of Clinical Medicine,
Interim Public Hospital Director
of Emergency Medicine Services,
Louisiana State University Health
Science Center, New Orleans, LA
Nicholas Genes, MD, PhD
Assistant Professor, Department of
Emergency Medicine, Icahn School
of Medicine at Mount Sinai, New
York, NY
Michael A. Gibbs, MD, FACEP
Professor and Chair, Department
of Emergency Medicine, Carolinas
Medical Center, University of North
Carolina School of Medicine, Chapel Michael S. Radeos, MD, MPH
Hill, NC
Steven A. Godwin, MD, FACEP
Professor and Chair, Department
of Emergency Medicine, Assistant
Dean, Simulation Education,
University of Florida COM-
Jacksonville, Jacksonville, FL
Gregory L. Henry, MD, FACEP
Clinical Professor, Department of
Emergency Medicine, University
of Michigan Medical School; CEO,
Medical Practice Risk Assessment,
Inc., Ann Arbor, MI
John M. Howell, MD, FACEP
Clinical Professor of Emergency
Medicine, George Washington
University, Washington, DC; Director Alfred Sacchetti, MD, FACEP
of Academic Affairs, Best Practices,
Inc, Inova Fairfax Hospital, Falls
Church, VA
Shkelzen Hoxhaj, MD, MPH, MBA
Chief of Emergency Medicine, Baylor Robert Schiller, MD
College of Medicine, Houston, TX
Eric Legome, MD
Chief of Emergency Medicine,
King’s County Hospital; Professor of
Clinical Emergency Medicine, SUNY
Downstate College of Medicine,
Brooklyn, NY
Keith A. Marill, MD
Research Faculty, Department of
Emergency Medicine, University
of Pittsburgh Medical Center,
Pittsburgh, PA
Charles V. Pollack Jr., MA, MD,
FACEP
Professor and Chair, Department of
Emergency Medicine, Pennsylvania
Hospital, Perelman School of
Medicine, University of Pennsylvania,
Philadelphia, PA
Assistant Professor of Emergency
Medicine, Weill Medical College
of Cornell University, New York;
Research Director, Department of
Emergency Medicine, New York
Hospital Queens, Flushing, NY
Ali S. Raja, MD, MBA, MPH
Vice-Chair, Emergency Medicine,
Massachusetts General Hospital,
Boston, MA
Robert L. Rogers, MD, FACEP,
FAAEM, FACP
Assistant Professor of Emergency
Medicine, The University of
Maryland School of Medicine,
Baltimore, MD
Assistant Clinical Professor,
Department of Emergency Medicine, Associate Professor of Emergency
Thomas Jefferson University,
Philadelphia, PA
Chair, Department of Family Medicine,
Beth Israel Medical Center; Senior
Faculty, Family Medicine and
Community Health, Icahn School of
Medicine at Mount Sinai, New York, NY
Scott Silvers, MD, FACEP
Chair, Department of Emergency
Medicine, Mayo Clinic, Jacksonville, FL
Corey M. Slovis, MD, FACP, FACEP
Professor and Chair, Department
of Emergency Medicine, Vanderbilt
University Medical Center, Nashville, TN Michael Guthrie, MD
Stephen H. Thomas, MD, MPH
George Kaiser Family Foundation
Professor & Chair, Department of
Emergency Medicine, University of
Oklahoma School of Community
Medicine, Tulsa, OK
David M. Walker, MD, FACEP, FAAP
Director, Pediatric Emergency
Services, Division Chief, Pediatric
Emergency Medicine, Elmhurst
Hospital Center, New York, NY
Ron M. Walls, MD
Professor and Chair, Department of
Emergency Medicine, Brigham and
Women's Hospital, Harvard Medical Giorgio Carbone, MD
School, Boston, MA
Critical Care Editors
William A. Knight IV, MD, FACEP
Associate Professor of Emergency
Medicine and Neurosurgery, Medical
Director, EM Midlevel Provider
Program, Associate Medical Director,
Neuroscience ICU, University of
Cincinnati, Cincinnati, OH
Scott D. Weingart, MD, FCCM
Medicine, Director, Division of ED
Critical Care, Icahn School of Medicine Dhanadol
at Mount Sinai, New York, NY
Senior Research Editors
Clinical Pharmacist, Emergency
Room, St. Joseph’s Hospital and
Medical Center, Phoenix, AZ
Joseph D. Toscano, MD
Chairman, Department of Emergency
Medicine, San Ramon Regional
Medical Center, San Ramon, CA
Research Editors
Emergency Medicine Residency,
Icahn School of Medicine at Mount
Sinai, New York, NY
Federica Stella, MD
Emergency Medicine Residency,
Giovani e Paolo Hospital in Venice,
University of Padua, Italy
International Editors
Peter Cameron, MD
Academic Director, The Alfred
Emergency and Trauma Centre,
Monash University, Melbourne,
Australia
Chief, Department of Emergency
Medicine Ospedale Gradenigo,
Torino, Italy
Amin Antoine Kazzi, MD, FAAEM
Associate Professor and Vice Chair,
Department of Emergency Medicine,
University of California, Irvine;
American University, Beirut, Lebanon
Hugo Peralta, MD
Chair of Emergency Services,
Hospital Italiano, Buenos Aires,
Argentina
Rojanasarntikul, MD
Attending Physician, Emergency
Medicine, King Chulalongkorn
Memorial Hospital, Thai Red Cross,
Thailand; Faculty of Medicine,
Chulalongkorn University, Thailand
Suzanne Y.G. Peeters, MD
Emergency Medicine Residency
Director, Haga Teaching Hospital,
The Hague, The Netherlands
Case Presentations
Your first patient of the shift is a 42-year-old woman
who details a family history of protein S deficiency and
presents with new-onset atraumatic left calf swelling and
pain. The proximal compression ultrasound is normal, but
your patient is understandably still nervous, and your
own suspicion for DVT remains high as well. You wonder
about the next best steps to ensure a safe discharge.
As your shift gets progressively busier, your ultra-
sound technician informs you that he is backed up with
studies and will take “a while” before he is caught up
again. Unfortunately, your next patient is a 22-year-old
otherwise healthy man sent from a clinic to “rule out
DVT,” with a handwritten script requesting an ultra-
sound. The patient reports unilateral left calf pain after
starting a new exercise regimen and has no other risk
factors or predispositions for DVT. You wonder if you
can safely rule out a DVT in this patient with a D-dimer
instead of the ultrasound.
In the meantime, a patient whom your colleague had
previously signed out to you as admitted for a straight-
forward, proximal DVT of the left lower extremity now
approaches the nurses’ station expressing his frustration
about the extended wait for his inpatient bed upstairs. The
patient no longer wants to stay and asks why he cannot
just follow up with his primary care doctor tomorrow.
You wonder if it would be safe to discharge him with an
outpatient treatment plan.
Introduction
Deep venous thrombosis (DVT) is commonly diag-
nosed and managed in the emergency department
(ED), accounting for about 1 in 1000 patients per year.1
In just the past 2 decades, the evaluation and recogni-
tion of DVT have significantly advanced from the days
of uncertainty about how to address notoriously vague
physical signs and symptoms and performing un-
wieldy venograms. Clinical judgment is supplemented
by the use of standardized risk stratification tools,
primarily the Wells clinical score, which has under-
gone several modifications over time. Uncomfortable
venograms have been replaced with noninvasive
proximal or whole-leg ultrasound. In addition, the use
of D-dimer assays to exclude DVT in appropriately
risk-stratified patients has increased rapidly, reducing
the need for ultrasound in certain patients.
Evidence-based recommendations, including
several clinical practice guidelines, help clinicians
navigate the expanding array of available diagnos-
tic tools and therapeutic options. On the diagnostic
side, ultrasonography is migrating to the bedside
and into the hands of emergency clinicians, while
newer evidence challenges the upper-limit thresholds
of D-dimer interpretation in select populations. On
the therapeutic side, updated guidelines now chal-
lenge the emergency clinician to consider outpatient
Copyright © 2015 EB Medicine. All rights reserved. 2
management of proximal DVT as well as initial
nontreatment of certain types of DVT. Furthermore,
the newest oral formulations of anticoagulants offer
an enticing glimpse of a possible future without daily
injections, dietary restrictions, and frequent interna-
tional normalized ratio (INR) checks. In short, there is
much new ground to cover.
Critical Appraisal Of The Literature
A literature search of PubMed and the Cochrane
Database of Systematic Reviews was carried out us-
ing combinations of the following key search terms:
DVT, D-dimer, Wells score, Wells criteria, venous throm-
boembolism, proximal ultrasound, whole-leg ultrasound,
distal DVT, calf vein DVT, and anticoagulation. Targeted
literature searches were also performed for recurrent
DVT, pregnancy and DVT, upper extremity DVT, inferior
vena cava filter, thrombolysis and DVT, thrombectomy and
DVT, and novel oral anticoagulants.
There are several comprehensive evidence-
based guidelines on the diagnosis and therapy of
acute DVT, most notably the 2012 American College
of Chest Physicians (ACCP) ninth edition recom-
mendations for antithrombotic therapy and preven-
tion of thrombosis.1-4 Throughout this review, level
of evidence and strength of recommendations are
ranked as per the Grading of Recommendations As-
sessment, Development and Evaluation (GRADE)
Working Group.5,6 Definitions for strength of recom-
mendation and levels of evidence are available at:
http://journal.publications.chestnet.org/article.
aspx?articleid=1084215. Other available guidelines to
consider include the American College of Emergency
Physicians (ACEP) clinical policy recommendations
for suspected DVT (which have not been revisited
since 2003),7 the 2013 guidelines provided by the
Institute for Clinical Systems Improvement regarding
venous thromboembolism (VTE) diagnosis and treat-
ment,8 and the 2013 fifth edition of the International
Consensus Statement on Prevention and Treatment of
Venous Thromboembolism.2
The heterogeneity of various available D-dimer
assays and discordant versions of the Wells clini-
cal prediction rule make interpretation of available
data challenging. Many diagnostic and therapeutic
studies have combined or extrapolated analyses of
DVT from pulmonary embolism (PE) cohorts. Spe-
cial subsets of patients, such as those with recurrent
or upper extremity DVT and pregnant patients,
are often excluded from larger analyses, so docu-
mentation of their experience is much less robust.
Finally, randomized controlled trials concerning the
emerging novel oral anticoagulants have primarily
targeted noninferiority. Post marketing experience
and direct comparisons are limited, and recommen-
dations for their use are not formally graded by
most existing guidelines.
www.ebmedicine.net • March 2015
 Pathophysiology And Anatomy
Pathophysiology
Formation of a deep venous thrombus is influenced
by the elements of the Virchow triad: venous stasis,
vascular endothelial injury, and hypercoagulabil-
ity.9 The identified risk factors for DVT affect one or
more of the elements of this triad and are classified
as either intrinsic or extrinsic, and either provoked or
unprovoked. (See Table 1.) Intrinsic causes include
inherited thrombophilias, the most common of
which (in order of decreasing relative risk [RR]) are
antithrombin deficiency, protein C or S deficiency,
and factor V Leiden and prothrombin gene muta-
tions. Acquired intrinsic causes include malignancy,
advanced age (especially > 75 years), antiphospho-
lipid syndrome, nephrotic syndrome, and obesity.10
Provoked DVT is precipitated by extrinsic factors,
most commonly immobilization of > 48 hours’ dura-
tion, recent hospitalization, recent surgery, localized
or generalized trauma, pregnancy, oral contracep-
tives or hormone replacement therapy, air travel (> 8
hours in duration11), transient infectious disease, and
deterioration of general condition or frailty.10
One of the strongest risk factors for develop-
ment of DVT is a history of prior DVT or PE; previ-
ous undiagnosed or silent VTE may account for
unidentified risk factors. Many patients will have ≥
1 intrinsic risk factors with a subsequent acquired or
extrinsic risk factor.10,12,13 Recognition of the pre-
disposing conditions helps in the risk stratification
of patients with suspected DVT and elucidates the
potential etiology in patients with confirmed DVT.
Table 1. Risk Factors For Development Of
Deep Venous Thrombosis
• Hypercoagulability
Antithrombin deficiency
l
Protein C deficiency
l
Protein S deficiency
l
Factor V Leiden
l
Prothrombin gene mutations
l
Antiphospholipid syndrome
l
Homocysteinuria
l
Nephrotic syndrome
l
• Immobilization > 48 hours
• Recent hospitalization or surgery
• Malignancy
• Trauma
• Pregnancy
• Exogenous hormone use
• Deterioration in general condition
• Prior DVT or VTE
Abbreviations: DVT, deep venous thrombosis; VTE, venous thrombo-
embolism.
March 2015 • www.ebmedicine.net
If not appropriately diagnosed and treated,
DVT has the potential for significant morbidity and
mortality, including development of PE, recur-
rent DVT, or postthrombotic syndrome. PE is the
most deadly complication; it is usually caused by
lower extremity DVT and it accounts for a 1-month
mortality rate of 9.4% from first-time DVT. It is
estimated that 40% to 50% of patients with DVT
may harbor a silent PE.14-17 When it is treated,
the mortality from DVT drops to < 6%, although
the 2-year mortality from DVT may be as high as
20%.16,17 Accurate diagnosis is imperative, as the
risk of acute disease must be balanced with the
risks of treatment with anticoagulation.18
Anatomy
The majority (84%) of patients diagnosed with DVT
present with lower extremity thrombosis, with
smaller percentages presenting with isolated distal
(13%) or upper extremity DVT (16%).13 Although
venous drainage flows only in a proximal direc-
tion and anatomic variation is common, it is easier
to conceptualize the anatomy of the deep veins in
the opposite direction, ie proximally to distally.
(See Figure 1, page 4.) The inferior vena cava (IVC)
bifurcates into the iliac veins, dividing into internal
and external branches. When the external iliac pass-
es underneath the inguinal ligament, it is renamed
the common femoral vein. At about 5 to 7 cm infe-
rior to the inguinal ligament, the common femoral
vein bifurcates into the femoral vein and the deep
femoral vein, which courses to the lateral thigh
muscles. The femoral vein (also referred to as the
"superficial" femoral vein, which is a misnomer in
that it is part of the deep venous system) continues
posterior to the femoral artery, coursing inferiorly
and medially along the upper thigh before passing
through the popliteal fossa to become the popliteal
vein. At the popliteal fossa, the popliteal vein lies
lateral to the popliteal artery, before coursing more
posteriorly as it travels distally. At the level of the
popliteal trifurcation, the popliteal vein branches
into the deep veins of the calf, with the anterior tib-
ial vein coursing anteromedially, the posterior tibial
vein posteriorly, and the peroneal vein laterally.
The muscular veins – including the gastrocnemius
and soleus veins – branch off the deep veins of the
calf, with some debate as to whether or not they
should still be considered part of the deep venous
system.19 In contrast to the deep venous system, the
superficial venous system includes the greater sa-
phenous vein, which branches off the femoral vein
inferior to the inguinal ligament, and the lesser
saphenous vein, which commonly branches off the
popliteal vein.
3 Reprints: www.ebmedicine.net/empissues
Differential Diagnosis
The signs and symptoms of DVT overlap with
many other conditions. (See Table 2.) Overall, the
prevalence of confirmed thrombosis in patients
suspected of having DVT is about 19%, with rates
ranging from 5% in patients with low probability to
53% in patients with high probability for DVT.20 As
such, the majority of patients presenting with signs
or symptoms suggestive of DVT will ultimately be
diagnosed with alternate conditions. For instance,
lower extremity swelling may be caused by pre-ex-
isting orthopedic abnormalities or paralysis.21 Table
2 summarizes the findings from a recent outpatient
study of 1028 patients suspected of having DVT,
Figure 1. Schematic Representation Of The
Lower Extremity Veins
Ilac veins
2
3 Emergency Department Evaluation
4
Femoral veins
5 consists of unilateral pain, swelling, edema, ery-
6 of course, the classic Homans sign (demonstrated by
Popliteal vein/
Trifurcation area
7 8 9 obscuring which variables are most helpful in the
Distal veins 10 11 12
only 13.4% of whom were ultimately diagnosed with
DVT.22 As many as half of all patients with a history
of prior DVT may have persistent pain and swelling
in the affected extremity from postthrombotic syn-
drome, further complicating the clinical evaluation.23
Prehospital Care
There are no published guidelines on prehospital
management of DVT. Since the primary acute con-
cern in patients with suspected DVT is progression
to PE, emergency medical services personnel should
be cognizant of the patient with a swollen painful
lower extremity who is or becomes hemodynami-
cally unstable or who develops chest pain, shortness
of breath, or other signs of acute cardiopulmonary
distress. Management should consist of supportive
care and transport to an appropriate facility. The
stable patient with suspected DVT may be transport-
ed by a basic life support ambulance. If the patient
is or becomes unstable and resources are available,
1
transport via an advanced life support provider
should be considered.
Inguinal ligament
History And Physical Examination
The textbook presentation of lower extremity DVT
thema, warmth, tenderness to palpation along the
Adductor canal
deep venous system, secondary dilation of superfi-
Knee joint cavity cial collateral veins or a palpable venous cord, and,
eliciting of calf pain with passive, abrupt dorsiflex-
13 ion of the foot). As every practicing emergency clini-
cian knows, however, actual clinical presentations
of DVT can be insidiously subtle and nonspecific,
clinical diagnosis of DVT.
To address this uncertainty, Goodacre et al per-
formed a meta-analysis of over 50 diagnostic cohorts
with suspected DVT to evaluate test characteristics
of these variables.24 Based on the strongest positive

14

Table 2. Differential Diagnosis For

Suspected Deep Venous Thrombosis22
Diagnosis Percentage
Legend: 1, external iliac vein; 2, common femoral vein; 3, greater
saphenous vein; 4, deep femoral vein; 5, femoral vein; 6, popliteal Muscle strain / tear / hematoma 16.0%
vein; 7, anterior tibial confluent segment; 8, posterior tibial confluent Deep venous thrombosis 13.4%
segment; 9, peroneal confluent segment; 10, anterior tibial veins; 11,
Chronic venous insufficiency 12.6%
posterior tibial veins; 12, peroneal veins; 13, gastrocnemius muscle
veins; 14, soleus muscle veins. Cellulitis 10.9%
Superficial thrombophlebitis 9.4%
Reprinted with permission from G. Palareti and S. Schellong. Isolated Lymphedema 4.8%
distal deep vein thrombosis: what we know and what we are doing.
Baker (popliteal) cyst 3.9%
Journal of Thrombosis and Haemostasis. © 2011 International Soci-
ety on Thrombosis and Haemostasis. Other 29%

Copyright © 2015 EB Medicine. All rights reserved. 4 www.ebmedicine.net • March 2015

likelihood ratios (LR+) with 95% confidence inter-
vals (CI), the features that proved independently
predictive of DVT included malignancy, history of
previous DVT, recent immobilization, difference
in calf diameter, and recent surgery. In contrast,
Homans sign, calf pain, calf swelling, and obesity
did not carry high enough LR+ to serve as reliable
indicators of DVT. (See Table 3.) No single feature
had a low enough negative likelihood ratio (LR-) to
reliably exclude DVT, but absence of calf swelling
(LR- 0.67; 95% CI, 0.58-0.78) and lack of difference
in calf diameter (LR- 0.57; 95% CI, 0.44-0.72) may
indicate a slight reduction in the likelihood of DVT.
Not surprisingly, it is more effective to look at a
combination of the clinical features, risk factors, and
alternate diagnoses than to focus on isolated vari-
ables.24 For instance, a high Wells score incorporat-
ing a scripted combination of the above factors (see
Table 4) yielded a much more predictive LR+ of 5.2
(95% CI, 4.0-6.0) while a low Wells score rendered
the disease less likely, with an LR- of 0.25 (95% CI,
0.21-0.29). When physicians dichotomized their clini-
cal judgment to high versus low, high-probability
estimations resulted in an LR+ of 6.2 (95% CI, 1.0-
40.0) while low-probability estimations were more
fine-tuned at an LR- of 0.18 (95% CI, 0.13-0.26).
When the clinical suspicion for DVT is high, the
emergency clinician should simultaneously assess
for the complications of DVT. In addition to PE, DVT
with massive clot burden can cause a severe form of
venous obstruction known as phlegmasia cerulean
dolens, which is characterized by sudden-onset
swelling, bluish discoloration, and even loss of arte-
rial pulses that can lead to gangrene and loss of the
limb. Prompt consultation for thrombolysis or surgi-
cal thrombectomy is indicated in this situation. More
common is concurrent or long-term postthrombotic
syndrome, which affects up to one-half of patients
Table 3. Diagnostic Value Of Clinical
Features Of Deep Venous Thrombosis24
Clinical Feature LR+ 95% CI
Highly Predictive
Malignancy 2.71 2.16-3.39
Previous DVT 2.25 1.57-3.23
Recent immobilization 1.98 1.70-2.30
Difference in calf diameter 1.80 1.48-2.19
Recent surgery 1.76 1.40-2.20
Less Predictive
Homans sign 1.40 1.18-1.66
Calf pain 1.08 0.96-1.20
Calf swelling 1.45 1.25-1.69
Obesity 0.85 0.59-1.23
Abbreviations: DVT, deep venous thrombosis; LR+, positive likelihood
ratio; CI, confidence interval.
March 2015 • www.ebmedicine.net
(especially those with recurrent ipsilateral DVT) and
causes symptoms of venous insufficiency, persistent
pain and swelling, skin discoloration, varicose veins,
and, in severe cases, nonhealing lower extremity
ulcerations.23
Risk Stratification
Since its inception in 1995, the Wells clinical
model has been the most often-cited prediction
rule for stratifying pretest probability of sus-
pected first-time lower extremity DVT. Wells et
al used signs and symptoms, risk factors, and the
potential for alternative diagnoses to prospec-
tively score outpatients referred for suspected
DVT as having either low (< 0 points), moderate
(1-2 points), or high (≥ 3 points) pretest prob-
ability. These risk groups corresponded to a DVT
prevalence of 5%, 33%, and 85% on venography,
respectively (P < .001), while demonstrating high
interobserver reliability (kappa = 0.85) using
their original 12-point clinical model.25
Table 4. Wells Clinical Model For Predicting
The Pretest Probability Of Deep Venous
Thrombosis
Clinical Characteristic Score
Active cancer (patient receiving treatment for cancer 1
within the previous 6 mo or currently receiving palliative
treatment)
Paralysis, paresis, or recent plaster immobilization of the 1
lower extremities
Recently bedridden for 3 days or more, or major surgery 1
within previous 12 wk requiring general or regional
anesthesia
Localized tenderness along the distribution of the deep 1
venous system
Entire leg swollen 1
Calf swelling at least 3 cm larger than that on the 1
asymptomatic side (measured 10 cm below the tibial
tuberosity)
Pitting edema confined to the symptomatic leg 1
Collateral superficial veins (nonvaricose) 1
*Previously documented deep venous thrombosis 1
Alternative diagnosis at least as likely as deep venous -2
thrombosis
Three-tiered scoring method20: High ≥ 3; moderate = 1 to 2; low < 0.
Two-tiered scoring method27: Likely ≥ 2; unlikely ≤1.
* Later added to modified versions of Wells scoring;27,20 absent from
original scoring.25,26
From The New England Journal of Medicine. Philip S. Wells, David R.
Anderson, Marc Rodger, et al. Evaluation of D-dimer in the diagnosis
of suspected deep-vein thrombosis. Volume 349, pages 1227-1235.
Copyright © 2003 Massachusetts Medical Society. Reprinted with
permission from Massachusetts Medical Society.
5 Reprints: www.ebmedicine.net/empissues
 In 1997, using a simplified 9-point version of
this model, Wells et al prospectively demonstrated
the safety of a more focused application of serial
ultrasounds tailored to pretest probability by elimi-
nating the traditional repeat ultrasound at 1-week
follow-up in patients with low pretest probability of
DVT.26 Differences in the prevalence of DVT in the
low, moderate, and high pretest probability groups
were maintained at 3%, 17%, and 75%, respectively
(P < .00001), while interobserver reliability between
study nurses and physicians remained high
(kappa = 0.75).
Yet another modification of the Wells clinical
model accounts for inclusion of patients with previ-
ously documented DVT by adding a point to the
clinical score for this history, while also differentiat-
ing pretest probability into "likely" versus "unlikely"
categories.27 In this 2003 iteration of the Wells
criteria, Wells et al randomized 1096 outpatients
with suspected DVT to either initial ultrasound or
D-dimer testing, and the results demonstrated a safe
reduction in ultrasounds in unlikely patients via
the alternative use of D-dimer. Patients who were
prospectively scored as clinically likely to have DVT
demonstrated a 28% rate of DVT on 3-month follow-
up, compared with a 5.5% rate in the clinically
unlikely category.
In 2006, Wells et al revisited their earlier 3-tiered
clinical model and similarly modified it with the
additional point for prior history of DVT.20 (See
Table 4, page 5.) A systematic review of 14 prospec-
tive cohorts that included 8239 outpatients with
suspected DVT revealed pooled prevalences of 5%,
17%, and 53% for DVT in the low, moderate, and
high pretest probability categories, respectively. The
3-tiered stratification of pretest probability forms the
suggested basis for tailoring the diagnostic work-up
toward a patient with suspected DVT in multiple
current guidelines.1,2,7,8 While available guidelines
agree that a negative D-dimer is sufficient initial
testing in low pretest probability groups, the ACCP
additionally advocates for the use of initial D-dimer
testing in moderate pretest probability patients
under specific conditions. (See Clinical Pathway
for the Diagnostic Evaluation of a Suspected First
Lower Extremity DVT, page 14).
Despite how widely the Wells clinical score is
cited in the literature, its comparative performance
against clinician gestalt remains unclear. Although
dedicated studies are limited, Goodacre et al sug-
gested similar performance between unstructured
clinician assessments and standardized Wells scor-
ing.24 A more recent single-center prospective cohort
with a high prevalence of DVT (47%) reported an LR+
of 6.82 for empirical estimation of high probability by
vascular specialists versus an LR+ of 2.23 for a likely
score via Wells criteria applied by emergency physi-
cians.28 This suggests that, with the appropriate expe-
rience, empirical estimation may be as predictive as, if
Copyright © 2015 EB Medicine. All rights reserved. 6
not more predictive than, Wells scoring. Nonetheless,
its explicit standardization and ease of application
continue to make the Wells scoring tool ideal for both
ongoing research and clinical practice.
Additional research evaluates the interobserver
variability of the Wells clinical model, as well as
its applicability both to broader and more specific
populations. External calculations of interobserver
variability using the score have ranged from a kappa
of 0.58 between resident and senior vascular medicine
specialists29 to 0.74 between emergency physicians
and nurse practitioners using 3-tiered Wells scoring,30
to 0.836 between emergency physicians and vascular
specialists using 2-tiered Wells scoring.28
In 2005, Oudega et al suggested that the Wells
criteria may not perform as reliably in the primary
care setting,31 although the earlier version of Wells
scoring used in this analysis did not account for prior
documented DVT. Subgroup analysis of a combined
meta-analysis of 10,002 patients has uncovered ad-
ditional insights with regard to patients with prior
history of DVT and active malignancy.32 In 941 pa-
tients with suspected recurrent DVT, correction from
the original Wells criteria to the modified version (by
adding a point for prior history of DVT) improved the
false-negative rate of DVT from 2.5% (95% CI, 1.2-5.4)
to 1.0% (95% CI, 0.6-1.6). However, subgroup analysis
of 834 patients with active cancer revealed that the
combination of an unlikely score and a negative D-
dimer resulted in a higher miss rate of 2.2% (95% CI,
0.5-8.6), suggesting that imaging may have stronger
benefit in this subset of patients than risk stratification
and laboratory testing alone.
Diagnostic Studies
D-dimer
The D-dimer is considered a nonspecific but sen-
sitive biomarker for VTE. Because they may be
elevated in a number of nonthrombotic conditions
(eg, recent surgery, infection, trauma, malignancy,
pregnancy, stroke, and advanced age), D-dimers are
more helpful in excluding the diagnosis of VTE with
a negative result in the appropriately risk-stratified
patient than in confirming VTE when positive.
The various methods of D-dimer testing result
in different sensitivities and specificities for VTE that
should be viewed as institution-specific, based on the
particular type of assay used. Microplate enzyme-
linked immunosorbent assays (ELISAs) and newer-
generation quantitative latex agglutination (or immu-
noturbidimetric) assays are considered highly sensitive
for DVT (93%-94% sensitive), whereas whole blood D-
dimer assays such as SimpliRED® are only moderately
sensitive (at 83%), although they offer a more rapid
turnaround time and demonstrate a higher specificity
(approximately 71%).1,33-35
Given the current state of numerous coexisting
www.ebmedicine.net • March 2015
D-dimer assays and techniques across laboratories,
the astute clinician should be familiar with local
laboratory practices and maintain awareness of tech-
nique-specific units of measurement (eg, fibrinogen
equivalent units [FEU] vs D-dimer units [DDU]) and
laboratory-specific variable magnitudes of measure-
ment that are not always easily converted from one
to another.36
Laboratory variability notwithstanding, for the
purposes of excluding lower extremity DVT, a nega-
tive D-dimer result should be interpreted within
the context of the test method's degree of sensitivity
(moderate vs high) and the patient's clinical pretest
probability of DVT.1 (See Table 3, page 5; and Clini-
cal Pathway For The Diagnostic Evaluation Of A
Suspected First Lower Extremity DVT, page 14.) A
negative result using a moderate-sensitivity assay
may be sufficient to safely rule out DVT without
further testing in a patient with low pretest clinical
probability; however, a moderate-risk patient would
need at least a negative high-sensitivity D-dimer in
order to complete the workup without imaging. In
patients with moderate to high pretest probability of
DVT, a negative D-dimer result, in combination with
an initial negative proximal compression ultrasound,
may safely eliminate the need for follow-up serial
or initial whole-leg ultrasound. There is insufficient
evidence to support the safety of using only D-dimer
to rule out DVT in high-risk patients. The practice is
inefficient (fewer than 15% of high-risk outpatients
will yield a negative result37), and relying on D-
dimer results without imaging in high-risk patients
is not recommended.1
Adjusted D-dimer Thresholds
Although D-dimer testing is generally interpreted
in a binary manner, using the traditional cut-off (ie,
> 500 mcg/L FEU or < 500 mcg/L FEU) to yield
a positive or negative result, more recent studies
advocate adjusting the D-dimer threshold in se-
lect patient groups to improve specificity without
sacrificing sensitivity. Newer literature suggests that
future strategies may allow for routine use of higher
D-dimer cut-offs in patients with predefined low
pretest probability and for elderly patients. Consid-
erations for D-dimer in recurrent DVT, DVT during
pregnancy, and upper extremity DVT are discussed
in the ”Special Populations“ section (page 12).
In 2004, a retrospective study demonstrated that
D-dimer test specificity could be improved, while
maintaining equally high negative predictive values,
by using a higher D-dimer cut-off in patients with
low pretest probability of DVT.38 A 2013 follow-up
prospective randomized multicenter controlled trial
stratified and studied the safety of using a higher
threshold of < 1 mcg/mL FEU for outpatients with
low pretest probability of suspected first-time DVT,
compared with the standard threshold of < 0.5 mcg/
mL.37 Both testing strategies resulted in similar rates
March 2015 • www.ebmedicine.net
of VTE at 3-month follow-up in patients with low
pretest probability (0.0% vs 0.3%), while also reduc-
ing the number of ultrasounds in this group at an
absolute rate of 21%. The safety of this approach and
its ability to decrease the initial number of ultra-
sounds was also externally validated in a post hoc
analysis of a population with a much higher DVT
prevalence (22% vs 7%).39
Because D-dimer levels increase with age,
traditional cut-off values have less utility in elderly
patients for excluding VTE (eg, 500 mcg/L FEU).40,41
In response, there has been growing interest in ap-
plying age-adjusted cut-offs for patients aged > 50
years, determined by multiplying the patient’s age
in years times 10. For example, a patient who is 80
years old would be allowed an age-adjusted cut-off
of up to 800 mcg/L FEU.
A 2013 meta-analysis of 12,497 patients using
age-adjusted D-dimer cut-offs for suspected VTE
(either PE or DVT) demonstrated higher specificities
using age-adjusted cut-offs while still maintaining
a sensitivity > 97% in all age groups.42 The use of
age-adjusted D-dimer cut-offs for DVT was adapted
from the use of a retrospectively validated strategy
with PEs.43 A DVT-specific meta-analysis combining
cohorts with various D-dimer assays and versions
of the Wells criteria found a higher exclusion rate of
DVT in 1672 patients with nonhigh probability when
using an age-adjusted cut-off (51%) versus the exclu-
sion yield with the standard 500 mcg/L FEU cut-off
(42%), while still maintaining similar false-negative
rates in the age-adjusted (0.8%) and standard (0.7%)
groups.44 This has also been retrospectively validat-
ed in the primary care and outpatient DVT referral
settings in patients with low clinical probability.45,46
More recently, a prospective validation study of
3346 patients using age-adjusted D-dimer thresholds
to exclude suspected PE demonstrated a false-
negative rate of 0.3%,47 lending further credibility
to the movement toward age-adjusted thresholds. If
the DVT evidence continues to mirror the PE trend,
a similar prospective validation for DVT can be
expected in the near future.
Imaging Studies
Although rarely utilized in recent years due to the ad-
vances in ultrasonography, contrast venography has,
historically, been the gold standard for the diagnosis
of lower extremity DVT. After cannulating and inject-
ing radiopaque dye into the distal venous system, the
diagnosis is made by visualization of filling defects,
abrupt termination of the dye column, nonfilling of
the venous system, or diversion of flow.48 However,
this method has many limitations, including the need
for invasive testing, failure to cannulate an appropri-
ate vein in the dorsal foot, inadequate visualization of
veins, high interobserver variability of the results, and
even postprocedure thrombosis.21 Patients with an ad-
7 Reprints: www.ebmedicine.net/empissues
equate, negative study demonstrate a 1.3% (95% CI, and isolated distal deep vein thrombosis may be
0.2-4.4) risk of developing DVT within 3 months.1,21 revealed. Especially important in an ED population,
Despite these shortcomings, contrast venography a negative whole-leg ultrasound may obviate the
established the < 2% risk threshold standard against need for a repeat study in appropriately stratified
which modern forms of diagnostic testing are com- patients.55-57 However, evaluation of the distal veins
pared, and it remains a follow-up test of choice when is significantly less accurate than the proximal veins,
initial imaging is nondiagnostic.49 Current modalities with a pooled sensitivity of 63.5% (95% CI, 59.8-67.0)
for imaging (duplex ultrasonography, computed to- and specificity of 93.8% (95% CI, 93.1-94.4).50 Other
mography [CT] venography, and magnetic resonance disadvantages of whole-leg ultrasound include the
imaging [MRI]) offer less invasive and more readily increased complexity of the examination, increased
available alternatives. time to perform the examination, an increase in the
proportion of technically inadequate examinations,
Ultrasound and the potential for overtreatment of isolated distal
Venous duplex ultrasonography of the proximal deep vein thrombi that may not pose significant risk
deep veins is the most common means of evaluating for propagation.58
for the presence of DVT.49,50 Duplex ultrasonogra- Many studies that compared serial proximal
phy combines 2-dimensional grayscale (B-mode) im- compression ultrasound with whole-leg ultrasound
aging with Doppler color flow or spectral imaging. (while withholding anticoagulation after initial
Imaging of the deep venous system is performed negative studies) have demonstrated that both strat-
from the proximal common femoral vein through egies have a similar incidence of VTE on 3-month
the popliteal trifurcation.51 DVT is confirmed by follow-up (0.9%-2.0% for proximal compression
a combination of the following: abnormal venous ultrasound vs 1.2% for whole-leg ultrasound).57,59
compression on B-mode ultrasound (normally, the
vein should fully collapse with compression); the
presence of an echogenic thrombus within the lu- Figure 2. Duplex Ultrasound Image Of Deep
men of the vein; abnormal Doppler color flow or Venous Thrombosis
visualization of filling defects with color Doppler; or
abnormal spectral Doppler flow.50,52 (See Figure 2.)
Meta-analysis of duplex ultrasonography compared
with venography demonstrates a sensitivity of 96.5%
(95% CI, 95.1-97.6) and specificity of 94.0% (95% CI,
92.8-95.1) for diagnosis of proximal DVT by duplex
ultrasound.50 Limitations of duplex ultrasonogra-
phy include limited evaluation of DVT proximal to
the common femoral vein (ie, iliac veins), inability
to visualize the femoral vein within the adductor
canal, and, with traditional proximal duplex ultra-
sonography, nonimaging of the calf veins.49 Because
the risk of propagation of an undiagnosed distal
calf vein thrombus could be as high as 25%, serial
ultrasonography at 1 week has been traditionally
recommended for all patients.53,54 However, later
literature found only a 1.3% yield of propagation on
repeat ultrasound (95% CI, 0.97-1.9).50 Utilization of
a risk stratification scheme may eliminate the need
for the follow-up ultrasound. In a low-risk popula-
tion (DVT unlikely), Wells et al found that only 1.4%
(95% CI, 0.4-3.8) had DVT on follow-up after a single
proximal ultrasound, calling into question the need
to obtain repeat ultrasounds in patients who are risk
stratified with unlikely probability of DVT.27,50
Whole-leg duplex compression ultrasonogra-
phy can also be performed as an extension of the
standard proximal duplex compression ultrasound. To view a full-color version of this image, go to:
Because both the proximal and calf veins are studied www.ebmedicine.net/USDuplexDVT
with whole-leg ultrasonography, alternative diagno- A noncompressible echogenic thrombus with color Doppler filling de-
ses (such as superficial thrombophlebitis or muscu- fect within the lumen is demonstrated in the femoral vein at the level
loskeletal pathology) are more readily recognized of the saphenofemoral junction.
Image courtesy of Jacob Goertz, MD, RDMS.

Copyright © 2015 EB Medicine. All rights reserved. 8 www.ebmedicine.net • March 2015

However, whole-leg ultrasound may diagnose more
DVT than needs to be treated, with isolated distal
DVT accounting for 52.1% of diagnosed DVT in a
meta-analysis of whole-leg ultrasonography.60
When determining the role of ultrasound in
the diagnostic workup, pretest probability for
DVT should be risk stratified so that diagnostic
testing can be tailored accordingly.1,2,8 (See Clini-
cal Pathway For The Diagnostic Evaluation Of A
Suspected First Lower Extremity DVT, page 14.)
This may be accomplished using validated scoring
systems (eg, Wells score) or, alternatively, clinical
gestalt. (See the ”Risk Stratification“ section, page 5.)
For a patient with low or moderate pretest probabil-
ity, proximal ultrasound may be utilized as either
the initial diagnostic test or after prompting by a
positive D-dimer result (Grade 1B). (See the ”D-
dimer“ section, page 6.) If the proximal ultrasound
is negative in a patient with low pretest probability,
DVT is excluded without need for serial proximal
ultrasound (Grade 1B). If the proximal ultrasound is
negative in a patient with moderate pretest prob-
ability, either a concurrent D-dimer test on the index
visit (Grade 1C) if not already performed, or a serial
proximal ultrasound should be arranged (Grade
1B-C). Alternatively, in the patient with moderate
pretest probability, whole-leg instead of proximal
ultrasound can be elected (Grade 1B), particularly
if the patient has more-severe symptoms or may be
unable to return for serial testing. For patients with
high pretest probability, the initial test of choice is
proximal or whole-leg ultrasound (Grade 1B), as the
use of initial D-dimer testing in this population is
not recommended.1
Point-Of-Care Emergency Ultrasound
Focused point-of-care ultrasound has become
commonplace in the practice of emergency medi-
cine, and examination of the lower extremities for
DVT is among the core ultrasound applications for
emergency physicians.61 Unlike a comprehensive
duplex compression ultrasound, focused emergency
ultrasound consists of a 2-point compression-only
examination focusing on the common femoral vein
and the popliteal vein. Using a linear 5 to 10 MHz
probe, the common femoral vein is compressed seri-
ally from proximal to the saphenofemoral junction
distally past the confluence of the deep femoral vein
and femoral vein. The popliteal vein is then com-
pressed from the popliteal trifurcation proximally.62
Studies have demonstrated that point-of-care
lower extremity ultrasound is fast and accurate, and
decreases time to disposition in the ED.62-64 A recent
meta-analysis of ultrasound performed by emergency
physicians demonstrated a pooled sensitivity of 96.1%
(95% CI, 90.6-98.5) and specificity of 96.8% (95% CI,
94.6-98.1) in the evaluation of DVT.63 Potential areas
of concern for point-of-care lower extremity ultra-
March 2015 • www.ebmedicine.net
sound arise due to the inability to visualize of certain
parts of the anatomy, including the femoral vein
along the course of the leg, the calf veins, and, more
proximally, the iliac veins (where the addition of Dop-
pler findings can help identify thrombi). However, in
at least one study, no patients had isolated ("superfi-
cial") femoral vein DVT, with thrombus noted either
proximally in the common femoral vein or distally in
the popliteal vein.65 In addition, the compliance rate
for recommended 5- to 7-day serial studies can be low
in the ED population.66
Alternative Imaging Modalities
Computed Tomography Venography
In a meta-analysis, CT venography (which involves
helical CT imaging of the lower extremities following
venous injection of contrast) demonstrated a pooled
sensitivity of 95.9% (95% CI, 93.0-97.8) and specific-
ity of 95.2% (95% CI, 93.6-96.5) for lower extremity
DVT.67 Most of the included studies compared CT
venography to compression duplex ultrasonography,
reported only proximal DVT, and evaluated for DVT
as a part of a PE workup. CT venography has several
advantages: it is noninvasive, capable of diagnosing
pelvic or iliac vein thrombosis, and it can be per-
formed in conjunction with CT pulmonary angiog-
raphy to increase the sensitivity of evaluation for
VTE.68 CT venography can also be used to evaluate
for DVT in patients with indeterminate or technically
inadequate compression duplex ultrasound stud-
ies or in cases where performance of a compression
duplex ultrasound would be impossible, such as for a
patient with an overlying cast.67 Disadvantages of CT
venography include the risks involved with ionizing
radiation and intravenous contrast administration, as
well as lower sensitivity.68
Magnetic Resonance Imaging
MRI also has occasions for use in the diagnosis
of DVT. Several different techniques have been
documented, including visualization of blood flow
without administration of contrast (time-of-flight or
phase-contrast venography MRI), MRI with intrave-
nous gadolinium contrast, and magnetic resonance
direct thrombus imaging.1 Meta-analysis of MRI
demonstrated a pooled sensitivity of 91.5% (95%
CI, 87.5-94.5), with a higher sensitivity for proximal
DVT than distal DVT (93.9% vs 62.1%), and a pooled
specificity of 94.8% (95% CI, 92.6-96.5).69 Although
it avoids the ionizing radiation of CT venography,
MRI is more costly, more time-consuming, and less
frequently available. However, it remains an alterna-
tive for patients in whom compression duplex ultra-
sound is nondiagnostic and for whom CT or contrast
venography is contraindicated.
Current guidelines recommend against the rou-
tine use of CT venography or MRI in patients with
suspected first (Grade 1C) or recurrent (Grade 1B)
lower extremity DVT.1 If subtle ultrasound findings
9 Reprints: www.ebmedicine.net/empissues
or extensive unexplained swelling with high clinical
suspicion suggest the presence of an iliac vein DVT,
use of CT venography or MRI should be considered
(Grade 1C).8,70
Treatment
Treatment for acute DVT is divided into 3 phases:
(1) initial (0 to ~7 days), (2) long-term (~7 days to
~3 months), and (3) extended (> ~3 months). Most
cases of proximal DVT and some cases of severe
or high-risk distal DVT will be treated with antico-
agulation in the initial and long-term phases, with
adjunct recommendations for early ambulation and
compression therapy to prevent or decrease sequelae
of postthrombotic syndrome. New oral direct inhibi-
tors, which act on factor Xa (eg, rivaroxaban and
apixaban) or thrombin (eg, dabigatran), are adding
to the array of anticoagulant options, and some of-
fer the advantages of monotherapy and once-daily
dosing in the initial phase of anticoagulation. Select
patients may require longer durations of therapy in
the extended phase of anticoagulation; additionally
or alternatively, they may benefit from specialized
interventions ranging from IVC filters to thrombo-
lytics or operative thrombectomies. (See Clinical
Pathway For Treatment Of Lower Extremity DVT,
page 15.)
Initial Anticoagulation
Parenteral anticoagulation has long been the main-
stay of first-line DVT treatment to prevent extension
or recurrence of clots (Grade 1B). Provided there is
no renal insufficiency or other contraindications,
fixed-dose subcutaneous low-molecular-weight
heparin (LMWH) or fondaparinux are preferred
(Grades 2B and 2C, respectively) over unfraction-
ated heparin (UFH),4 offering greater convenience,
adaptability to the outpatient setting, and lower
potential for heparin-induced thrombocytopenia.
Meta-analysis of 23 randomized controlled trials that
compared LMWH to UFH for VTE demonstrated
a favorable decrease in recurrent VTE (odds ratio
[OR], 0.70; 95% CI, 0.57-0.85), major hemorrhage
(OR, 0.58; 95% CI, 0.40-0.83), and death (OR, 0.77;
95% CI, 0.63-0.93) with LMWH.71 Once-per-day dos-
ing of LMWH at the same total daily dose (eg, 2 mg/
kg once daily vs 1 mg/kg twice daily) has shown
no significant increases in bleeding, extension, or
recurrence compared with twice-daily dosing and
is the preferred regimen (Grade 2C).4 Fondaparinux
proved noninferior to LMWH in a randomized dou-
ble-blind trial comparing the 2 regimens.72 However,
there are some caveats: LMWH and fondaparinux
will accumulate in patients with renal insufficiency
(creatinine clearance < 30 mL/min), and neither has
reliable monitoring capabilities or effective reversal
agents (although protamine and recombinant factor
Copyright © 2015 EB Medicine. All rights reserved. 10
VIIa may show partial response).73 Like LMWH and
fondaparinux, rivaroxaban and apixaban are, in part,
renally excreted.
Novel Oral Anticoagulants
The novel oral anticoagulants offer an attrac-
tive and potentially game-changing alternative to
the more traditional parenteral heparins and oral
vitamin K antagonists (VKAs). (For more informa-
tion on new oral anticoagulants, see the October
2013 issue of Emergency Medicine Practice, available
at: www.ebmedicine.net/NOACs) Their use is
largely expanded from prior experiences with these
medications for postoperative DVT prevention and
stroke prevention in atrial fibrillation. Open-label
randomized trials of the direct factor Xa inhibitor,
rivaroxaban, in 3449 patients with DVT74 and 4832
patients with PE75 demonstrated noninferior rates of
recurrent VTE and decreased rates of major bleed-
ing (hazard ratio, 0.54; 95% CI, 037-0.79; P = .002) in
pooled analysis compared with standard LMWH/
VKA therapy,76 and it was the first of the novel oral
anticoagulants to obtain Food and Drug Adminis-
tration (FDA) approval for acute VTE treatment in
November 2012. More recently, in August 2014, the
FDA approved another factor Xa inhibitor, apixaban
(Eliquis®), for initial therapy of acute VTE, following
on the heels of its randomized double-blinded trial
(N = 5395 patients with combined acute VTE) that
demonstrated noninferiority to LMWH/warfarin as
well as significantly decreased rates of major bleed-
ing (RR, 0.31; 95% CI, 0.17-0.55; P < .001).77
Although the risk of bleeding appears lower,
the reversal agents are still in development and the
partial reversal efficacy of prothrombin complex
concentrate (PCC), activated prothrombin complex
concentrate (aPCC), and recombinant human factor
VIIa are being investigated.78
Cost-effectiveness calculations appear to favor
oral anticoagulants over combination LMWH/
VKA therapy after accounting for medication pric-
ing, shorter hospital stays, and costs of medication
administration and INR monitoring.77 Regardless
of the type of medication selected, initial therapy
should be started immediately and continued for a
minimum of 5 days, until maintenance or long-term
anticoagulation reaches therapeutic levels (Grade
1B).4 The ACCP guidelines even suggest consider-
ing empiric initial anticoagulation while awaiting
confirmatory diagnosis in patients with high clinical
suspicion (Grade 2C) or in patients with intermedi-
ate clinical suspicion when delays to diagnosis > 4
hours are anticipated (Grade 2C).4
Long-Term Anticoagulation
Although the long-term (ie, maintenance) phase of
anticoagulation therapy through the first 3 months
may have previously fallen beyond the scope of ED
www.ebmedicine.net • March 2015
care, growing prioritization for outpatient treatment
of acute DVT – as well as return visits for suspected
bleeds – necessitate that the up-to-date emergency
clinician be familiar with these regimens as well.
Therapeutic options for maintenance therapy range
from continuation of LMWH to transitioning to oral
VKAs or a new oral direct inhibitor anticoagulant.
In noncancer patients, VKAs have traditionally
been the oral maintenance anticoagulation agents
of choice. Nonetheless, they require the ability to
establish close INR monitoring and dietary counsel-
ing, are notorious for drug-drug interactions, and
necessitate initial parenteral therapy to bridge the
transition to therapeutic INR levels between 2 and 3.
To avoid unnecessary and prolonged hospital stays,
guidelines recommend commencing VKA therapy
on the first day of diagnosis4 and continuing paren-
tal therapy for a minimum of 5 days until the INR is
therapeutic for 24 hours (Grade 1B). There is some
debate as to the initial loading dose (5 mg or 10 mg
orally).79,80 Oral VKAs have long-standing proven
methods of anticoagulation reversal with vitamin K,
fresh frozen plasma, and, for life-threatening bleeds,
PCC.
Of the new oral anticoagulants, rivaroxaban and
apixaban offer the added benefit of monotherapy
from the time of initial anticoagulation; they are
also associated with a lower risk of bleeding than
the standard LMWH/VKA combination.15,74,75,77
As an alternative, dabigatran, a direct thrombin
inhibitor, reflects similar rates of recurrent VTE,
major bleeding, and death compared with VKAs,
although it requires initial bridging with parenteral
anticoagulation.81,82 A fourth new oral anticoagulant,
the factor Xa inhibitor edoxaban, is currently under
FDA review for long-term therapy following initial
parenteral therapy. It also demonstrated noninferior-
ity to VKAs in randomized, double-blind testing (N
= 8292) and, additionally, it reflected decreased rates
of clinically relevant bleeding (hazard ratio, 0.81;
95% CI, 0.71-0.94; P = .004).83 However, as previ-
ously discussed, unlike VKAs, the new oral antico-
agulants currently lack a dedicated reversal agent in
the setting of acute bleeding, and their varying rates
of renal excretion limit their use in patients with
decreased creatinine clearance.84
Patients with active cancer carry a higher risk
of recurrent VTE, bleeding, and variable responses
to traditional VKA therapy. Systematic review of
randomized trials has shown LMWH to be superior
to VKA therapy in this population, with a 6.7% rate
of recurrent VTE in the LMWH arm, compared with
12.9% in the VKA group (RR, 0.53; 95% CI, 0.36-0.76; P
= .0007).85 Subgroup analysis combining the DVT and
PE trials for rivaroxaban (n = 597 patients) suggests
a net clinical benefit with rivaroxaban (hazard ratio,
0.60; 95% CI, 0.36-0.99) after accounting for trends
toward decreased recurrent VTE (hazard ratio, 0.69;
95% CI, 0.36-1.33) and major bleeds (hazard ratio,
March 2015 • www.ebmedicine.net
0.53; 95% CI, 0.23-1.23),76 but the overall experience
with this new class of medications in this subset of
patients is still limited. In the meantime, daily LMWH
injections for at least 6 months remain the general
first-line recommendation for long-term and extend-
ed-term treatment in patients with active cancer.4,79
Extended Anticoagulation
For most uncomplicated acute DVT in noncancer
patients, 3 months of treatment are recommended
(Grade 1B).4 Extension of therapy beyond this dura-
tion, generally using the same anticoagulant agent
used for long-term therapy, may depend on presence
or absence of provoking factors (eg, recent surgery),
known hypercoagulability, prior history of DVT,
active malignancy, bleeding risk, and, more recently,
the use of D-dimer assays to predict future VTE
recurrence.86 To be fully aware of the risks of bleed-
ing, potential reversal agents (or lack thereof), and
treatment alternatives, emergency clinicians should
familiarize themselves with both the traditional and
the new anticoagulants.
Inferior Vena Cava Filters
An IVC filter is a permanent or retrievable (optional)
vascular filter designed to mechanically trap large
emboli and prevent massive PE in the approximately
4% of patients with DVT who have contraindications
to anticoagulation or who have recurrent PE while
they are therapeutically anticoagulated. Placement
of an IVC filter might also be appropriate for DVT
patients who are pregnant or have cancer. An IVC
filter may also be considered prophylactically in
high-risk patients with significant trauma or spinal
surgery.87 Current guidelines recommend use of an
IVC filter in patients with acute proximal DVT, as
well as patients for whom there are contraindica-
tions for anticoagulation (Grade 1B).2,4,8 The Society
of Interventional Radiology makes further recom-
mendations regarding the choice of retrievable ver-
sus permanent filters based on the patient’s expected
duration of elevated risk for PE or contraindication
for anticoagulation.88
No randomized controlled studies compare IVC
filter placement to anticoagulation, and no trials
compare the various filter designs.87,89 The Preven-
tion du Risque d'Embolie Pulmonaire par Interrup-
tion Cave (PREPIC) study found that, in comparison
with anticoagulation alone, placement of an IVC
filter in combination with anticoagulation decreased
the rate of PE (15% vs 9%; RR, 0.37; 95% CI, 0.17-
0.79), increased the rate of recurrent DVT (RR, 1.52;
95% CI, 1.02-2.27), and had no effect on develop-
ment of postthrombotic syndrome (RR, 0.87; 95% CI,
0.66-1.13) or death (RR, 0.97; 95% CI, 0.74-1.28).90 A
subsequent Cochrane review (a major component
of which was the PREPIC study) reached similar
conclusions.87 Complications of IVC filter placement
include DVT (as there is an elevated risk of throm-
11 Reprints: www.ebmedicine.net/empissues
bogenicity with the presence of any intravascular
foreign body), filter migration or embolization, and
IVC stenosis or perforation, with complication rates
increasing in relation to prolonged duration of the
placement procedure.89 Given the risk of DVT, the
ACCP guidelines recommend the use of a con-
ventional course of anticoagulant medication for
patients in whom an IVC filter has been placed if the
risk of bleeding has resolved (Grade 2B).4
Thrombolysis And Thrombectomy
Treatment modalities that remove the thrombus have
the potential to immediately increase the patency of
the vein and decrease the short-term and long-term
effects of DVT, including vascular compromise, PE,
and postthrombotic syndrome. The available modali-
ties include systemic thrombolysis, catheter-directed
thrombolysis, interventional mechanical clot removal,
or surgical thrombectomy. Early thrombus removal
strategies are preferred in patients with limb-threat-
ening venous ischemia due to iliofemoral DVT (ie,
phlegmasia cerulean dolens) (Grade 1A).4,70
Treating lower extremity venous thrombosis
with thrombolysis (as opposed to anticoagulation)
results in active lysis of the clot, with disadvan-
tages primarily related to the increased risk of
bleeding.91 Thrombolysis can be performed using
either systemic thrombolytic therapy or localized
catheter-directed therapy, with or without mechani-
cal thrombus fragmentation or aspiration (phar-
macomechanical thrombolysis). Catheter-directed
thrombolysis achieves clot lysis more rapidly and
with lower doses of thrombolytic agents than
systemic thrombolysis, and pharmacomechanical
thrombolysis can, potentially, further shorten the
procedure and reduce the necessary dose of throm-
bolytic medication.4,70
A Cochrane review demonstrated significantly
more complete clot lysis with thrombolysis (sys-
temic or localized) than with anticoagulation at 1
month (RR, 4.91; 95% CI, 1.66-14.53) and > 6 months
(RR, 2.37; 95% CI, 1.48-3.80), as well as decreased
rates of postthrombotic syndrome (RR, 0.64; 95% CI,
0.52-0.79). Patients treated with thrombolysis had an
increased risk of bleeding (RR, 2.23; 95% CI, 1.41-
3.52) but no change in occurrence of PE (RR, 1.00;
95% CI. 0.33-3.05) or recurrent DVT (RR, 1.41; 95%
CI, 0.37-5.40). There were also no changes in early
mortality at 1 month (RR, 0.76; 95% CI, 0.32-1.89) or
intermediate mortality within 6 years (RR, 1.12; 95%
CI, 0.319-1.89).91 There is a paucity of studies directly
comparing systemic to catheter-directed throm-
bolysis. In one study, patients treated with catheter-
directed thrombolysis had better preserved valvular
competence (44% vs 15%;
P = .049), less deep venous reflux (44% vs 81%;
P = .03), and a trend toward improved venous
patency in comparison with patients treated with
Copyright © 2015 EB Medicine. All rights reserved. 12
systemic thrombolysis.92 When comparing catheter-
directed thrombolysis to anticoagulation alone and
systemic thrombolysis to anticoagulation alone,
benefits and risks are similar.91
Current guidelines recommend the use of anti-
coagulant therapy over both systemic and catheter-
directed thrombolysis (Grade 2C) and advise that
thrombolysis should be considered only in patients
meeting all of the following criteria: iliofemoral
DVT, < 14 days since onset of symptoms, good func-
tional status, life expectancy of ≥ 1 year, low risk of
bleeding, and no other contraindications to throm-
bolytic therapy.4
Surgical Management
Historically, prior to development of anticoagulant
medications, surgical thrombectomy was the treat-
ment for iliofemoral DVT; now it is reserved for pa-
tients who have contraindications to thrombolysis or
for whom other treatment modalities have failed.93
Current recommendations (Grade 2C) state that
operative thrombectomy should be considered only
if all of the following criteria are present: iliofemo-
ral DVT, < 7 days since onset of symptoms, good
functional status, life expectancy of ≥ 1 year, and
readily available resources and expertise.4,70 Surgical
thrombectomy involves a venotomy and removal of
the clot. Embolization is prevented by placement of
a proximal occluding balloon catheter and the use of
positive pressure ventilation to reverse blood flow
in the IVC.94 Benefits of surgical thrombectomy for
iliofemoral DVT include increased venous patency
(75%-80%, compared with 35% for anticoagulation
alone), preservation of venous valve competency,
and fewer postthrombotic complications.93,94 Risks
include blood loss, embolization of thrombus, and
development of postprocedure DVT due to endothe-
lial disruption.93 Due to additional postthrombec-
tomy risk of DVT, it is recommended that patients
receive anticoagulation postoperatively if there are
no contraindications.4
Special Populations
Recurrent Deep Venous Thrombosis
The diagnosis of recurrent DVT is complicated by
a number of factors. Patients with prior DVT are
prone to postthrombotic syndrome of the affected
leg, which can be difficult to distinguish from a
new thrombotic episode. Although the Wells clini-
cal model was modified to allow for inclusion of
patients with prior DVT, many earlier cohort stud-
ies excluded patients with a documented history of
DVT. Ultrasonography may pose technical dif-
ficulties in interpretation of chronic versus acute
thrombus in an ipsilateral extremity that still bears
residual abnormalities, requiring direct comparison
with the exact location and venous diameters of
www.ebmedicine.net • March 2015
prior thrombi during compression.
A handful of studies have evaluated the util-
ity of D-dimer in this more complicated subset of
patients. Rathburn et al prospectively evaluated
300 consecutive patients with suspected recurrent
DVT, conducting sensitive D-dimer tests on all the
patients.95 They found a 0.75% rate of confirmed
thromboembolism on 3-month follow-up (95% CI,
0.02-4.09), which increased to 1.5% (95% CI, 0.18-
53%) when including an initially indeterminate re-
sult in a patient whose repeat acute thrombosis was
confirmed just outside the 3-month window. Bates
et al reported a 98% negative predictive value (95%
CI, 96-99) for a negative D-dimer result from their
prospective multicenter cohort study.96
Noting these diagnostic limitations, coupled
with early but promising results from unstratified
D-dimer testing, the ACCP guidelines suggest 1 of
3 diagnostic workup options for this population, as
appropriate1: (1) a single highly sensitive D-dimer
(especially if the prior ultrasound is not available
for comparison), followed by a proximal compres-
sion ultrasound if the D-dimer is positive; (2) serial
proximal compression ultrasounds on day 1 and day
7 (±1); or (3) negative initial proximal compression
ultrasound with a concurrent initial D-dimer (of ei-
ther moderate or high sensitivity), with no need for
follow-up testing if the D-dimer is negative. Patients
whose ultrasounds are abnormal but equivocal for
acute thrombus (particularly if there is no prior
ultrasound for direct comparison) are advised to
undergo either a more rapid follow-up ultrasound
prior to the 1-week ultrasound or a formal venog-
raphy for confirmation. During serial testing, it is
considered safe to withhold anticoagulation.1,96
The next progression in the diagnosis of acute
recurrent DVT is to incorporate risk stratification
into this diagnostic algorithm. Previously, a small
prospective study of 105 patients used the combina-
tion of dichotomized, modified Wells scoring and
D-dimer testing to evaluate the safety of this man-
agement approach. Although the subset of patients
unlikely to have DVT with a negative D-dimer
accounted for only 15% of evaluated patients in this
population with a high (45%) prevalence of DVT,
none of these 6 patients had a DVT on 3-month
follow-up.97 More recently, subgroup analysis of
a combined meta-analysis of 10,002 patients with
suspected DVT yielded 941 patients with a prior
confirmed history of DVT.32 Using the original Wells
criteria in this subset risked a 2.5% rate of DVT in
patients with low pretest probability and a negative
D-dimer (95% CI, 1.2-5.4). However, using the modi-
fied version of scoring, whereby an additional point
is added for previous DVT, the combination of a
low-risk score (< 1) with a negative D-dimer resulted
in a more acceptable miss rate of 1.0% (95% CI,
0.6-1.6). This suggests that using the modified Wells
scoring and a negative D-dimer is likely safer in rul-
March 2015 • www.ebmedicine.net
ing out recurrent acute DVT than using an unstrati-
fied approach. The data are still not as robust as for
diagnosis of first-time DVT, and larger prospective
validations specific to this population and the use of
risk stratification are needed.
Isolated Deep Distal Vein Thrombosis
The clinical significance and treatment of isolated
deep distal vein thrombosis (IDDVT) remains con-
troversial. IDDVT can occur anywhere in the distal
veins of the calf, including the deep calf veins (pero-
neal, posterior tibial, and anterior tibial veins) and
the muscular veins (soleus and gastrocnemius veins),
although there is some debate as to whether or not
the calf muscle veins should be considered part of the
deep venous system.19
Most DVT starts in the calf; the historic rate of
extension to the proximal veins – generally occur-
ring within the first 2 weeks – is up to 25%. Recent
studies have documented lower rates of propaga-
tion to the proximal veins (8%-16%), with a 1.6% to
3.4% risk of PE.19,53,98 In thrombosis isolated to the
muscular veins, only 3% propagate to the proximal
veins.99 Furthermore, up to 90% resolve spontane-
ously without treatment.19,53,98 In a review of stud-
ies diagnosing calf vein DVT by either venography
or whole-leg ultrasound, the overall prevalence of
DVT ranged between 14% and 37%, 23% to 59% of
whom were diagnosed with IDDVT.19 Patients with
IDDVT are more likely to be younger, have transient
provoking risk factors (eg, hospitalization, recent
surgery, trauma, or travel), and, therefore, have a
decreased rate of DVT recurrence, compared with
patients who have proximal DVT (2.6% vs 8.4%).19,98
Treatment of IDDVT must balance these lower
risks of propagation, embolization, and recurrence
with the reduced benefits of treatment and the risks
of anticoagulation.100 The goal of diagnosis in this
population is to identify the subset that would actu-
ally benefit from therapy,4 in contrast with patients
who should be observed via serial ultrasounds for
propagation of thrombus to the proximal system.
There is a paucity of studies regarding this treatment
paradigm. A 2012 meta-analysis identified 8 studies,
with a total of 454 patients, and found a decrease
in PE (OR, 0.12; 95% CI, 0.02-0.77; P = .03) and
thrombus propagation (OR, 0.29; 95% CI, 0.14-0.62;
P = .04) with anticoagulation.101 A 2012 review also
acknowledged the heterogeneity of the literature but
noted that unprovoked IDDVT (including a history
of malignancy or thrombophilia) has been demon-
strated to be more likely to propagate than provoked
IDDVT (such as that occurring after trauma or
surgery); that there was no benefit to extending an-
ticoagulation from 6 weeks to 12 weeks in low-risk
patients with transient risk factors, and that the risk
of major bleeding after treatment with anticoagu-
lants ranged from 0% to 6%.98
13 Reprints: www.ebmedicine.net/empissues
 Clinical
Clinical Pathway
Pathway ForFor The Diagnostic
Emergency Evaluation
Department Of A Suspected
Management Of Multiple
First Lower ExtremityShocks
Deep Venous Thrombosis1

Suspected first lower extremity DVT

Determine clinical pretest probability

(Grade 2B)

Moderate
Low probability High probability
probability

D-dimera POSITIVEc Proximal US Whole-leg USb

• Low probability: moderate- or (Grade 1B) (Grade 1B)
high-sensitivity test
• Moderate probability: high-
sensitivity test
(Grade 1B)

NEGATIVE NEGATIVE POSITIVE NEGATIVE POSITIVE

No DVT YES Was patient previously Treat for DVT No DVTd

(Grade 1B) stratified as low (Grade 1B) (Grade 1B)
probability?

NO

Isolated distal
Proximal DVT
DVT
Serial proximal D-dimer
USe • Moderate probabil-
• Moderate prob- POSITIVE ity: moderate- or
ability (Grade high-sensitivity test High risk of propa-
Treat for DVT gation or severe
1B-C) (Grade 1C)
(Grade 1B) symptoms?
• High probability • High probability:
(Grade 1B) high-sensitivity test (See "Ultrasound"
NO section, page 8)
(Grade 1B)
POSITIVE NEGATIVE
NEGATIVE YES

Treat for DVT No DVTc

Serial US Treat for DVT
(Grade 1B) (Grade 1B)
(Grade 2C) (Grade 2B)

a
Preferred initial strategy for patients with low probability (Grade 2B) and for select patients with moderate probability (see "D-dimer" section, page 6).
(Grade 2C)
b
Whole-leg US may be preferred in patients unable to return for serial testing and for those with severe symptoms or at high risk for propagation of distal
DVT.
c
Alternative acceptable strategy if moderate risk would be to proceed to whole-leg US. (Grade 1B)
d
Extensive unexplained swelling in patient with high probability should prompt consideration of isolated iliac DVT.
e
Alternative acceptable strategy if high risk would be to proceed to whole-leg US. (Grade 1B)
Abbreviations: DVT, deep venous thrombosis; US, ultrasound.
Definitions for strength of recommendation and levels of evidence are available at: http://journal.publications.chestnet.org/article.aspx?articleid=1084215.

Copyright © 2015 EB Medicine. All rights reserved. 14 www.ebmedicine.net • March 2015

 Clinical Pathway
Clinical Pathway For Treatment
For Emergency Of Lower
Department ExtremityOf Multiple
Management
Deep Venous Thrombosis
Shocks 4,8

Treatment indicated for

lower extremity DVT

Complicated DVT or contraindications

YES
to anticoagulation?

NO

Start initial anticoagulation

• Parenteral: LMWHa SC or UFH IV or
fondaparinuxa SC (Grade 1B)
or
• NOAC: rivaroxaban PO or apixaban
PO (Ungradedb)

Outpatient treatment appropriate?

• Adequate outpatient follow-up
NO • Adequate home circumstances
• No significant comorbidities
• Normal renal function

YES

Transition to long-term outpatient anticoagulation (Grade 1B)

• Parenteral bridge (5 days): LMWH or fondaparinux (Grade 1B) and initiation of oral VKAc
until INR 2-3 (Grade 1B) or dabigatran (Ungradedb)
or
• Continuation of NOAC: rivaroxaban or apixaban (Ungradedb) Other treatments (see text)
or • IVC filter (Grade 1B)
• Continuation of parenteral therapy in select populations (eg, active cancer) (Grade 2B) • Catheter directed thrombolysis
(Grade 2C)
• Systemic thrombolysis (Grade 2C)
• Surgical thrombectomy (Grade 2C)
YES
Inpatient treatment Treatment complications or failure?

NO

Outpatient treatment
• Long-term anticoagulation for a minimum of 3 months
(Grade 1B)
• Consideration of extended anticoagulation (> 3 months) in
select patients (eg, active cancer or second unprovoked
DVT) (Grade 1B)
• Adjuvant measures (eg, compression stockings (Grade
2B), early ambulation (Grade 2C))

Abbreviations: ACCP, American College of Chest Physicians; DVT, deep venous thrombosis; INR, international normalized ratio; IV, intravenous; IVC,
inferior vena cava; LMWH, low-molecular-weight heparin; NOAC, novel oral anticoagulant; PO, by mouth; SC, subcutaneous; UFH, unfractionated heparin;
VKA, vitamin K antagonist.
a
Preferred initial strategy. (Grade 2C)
b
No ACCP grade of recommendation available at time of publication of guidelines. Studies have demonstrated noninferiority compared with standard treat-
ment. LMWH (Grade 2C) or VKA antagonist (Grade 2B) suggested over NOAC.
c
Preferred regimen (Grade 2C).
Definitions for strength of recommendation and levels of evidence are available at: http://journal.publications.chestnet.org/article.aspx?articleid=1084215.

March 2015 • www.ebmedicine.net 15 Reprints: www.ebmedicine.net/empissues

 Current guidelines have mixed recommenda-
tions regarding treatment of IDDVT. The Interna-
tional Consensus Statement recommends 3 months
of anticoagulation for IDDVT,2 while the ACCP
guidelines caution that routine treatment of distal
calf veins is not warranted, as the risk of bleeding
from anticoagulation can outweigh the relatively
low risk of propagation or extension into a clinically
significant VTE. As such, only serial imaging of the
lower extremities for the first 2 weeks is recom-
mended in patients with provoked IDDVT without
severe symptoms or risk factors for extension (Grade
2B). Anticoagulation should be reserved only for
patients with severe symptoms or with risk factors
for extension (Grade 2C). These risk factors include
positive D-dimer, extensive thrombosis (eg, > 5 cm
in length, > 7 mm maximum diameter, or involving
multiple veins), lack of reversible provoking factor,
active cancer, history of VTE, and inpatient status.4 If
anticoagulation therapy is chosen, the ACCP guide-
lines currently recommend the full 3-month duration
of therapy for IDDVT; however, as discussed above,
shorter durations may be equally effective in select
populations.
Deep Venous Thrombosis In Pregnancy
Pregnancy is a known risk factor for VTE and VTE is
a leading cause of maternal mortality,102 yet preg-
nant patients have routinely been excluded from
studies of suspected DVT (including the develop-
ment of Wells clinical score). Further complicating
the presentation, pregnant patients may exhibit very
different characteristics from the general population.
As pregnancy progresses, patients are more likely to
develop leg swelling and discomfort independent of
DVT. D-dimer levels also rise with advancing gesta-
tional age, prompting proposals to consider validat-
ing higher testing thresholds for this population.103
In addition, there are anatomic variations in preg-
nancy-related DVT. More than 80% of pregnancy-re-
lated DVT involves the left lower extremity, and 17%
may be isolated to the iliac veins and, therefore, eas-
ily missed by proximal compression ultrasounds.104
Further compounding the diagnostic difficulties, the
overall prevalence of confirmed DVT in pregnant
patients with suspected disease appears generally
lower (8%-9%) than in nonpregnant cohorts,105-107
even though the risk of VTE in pregnant patients is
4-fold higher than in the general population.108
Accounting for these limitations and consider-
ations, the American Congress of Obstetricians and
Gynecologists (ACOG) and the ACCP recommend
ultrasonography as the initial diagnostic test of
choice for pregnant patients with suspected lower
extremity DVT.1,109 If this test is negative, the ACOG
recommends serial ultrasound or additional imag-
ing, and they do not endorse the use of D-dimer
testing in the pregnant patient. In contrast, the
Copyright © 2015 EB Medicine. All rights reserved. 16
ACCP guidelines allow for either a follow-up serial
proximal ultrasound performed on a more frequent
basis (tested on days 3 and 7) or a sensitive D-dimer
test conducted at the time of presentation. A posi-
tive D-dimer on the initial visit should prompt a
stronger emphasis on the importance of follow-up
ultrasounds on days 3 and 7. Involvement of the
entire leg, buttocks, or back should trigger suspicion
of iliac vein thrombosis and prompt follow-up with
the requisite imaging when standard proximal serial
compression ultrasounds are negative.1
In the meantime, a clinical prediction rule for
pregnant patients is in the process of development
and validation, though sample sizes remain lim-
ited.105-107 The first study to address this challenge
prospectively evaluated 194 pregnant patients with
suspected DVT at multiple sites and determined the
following 3 clinical variables to be highly predic-
tive105: (1) symptoms in the left leg; (2) calf circum-
ference difference > 2 cm; and (3) first-trimester
pregnancy. Each of the 17 patients (8.8%) diagnosed
with DVT in this cohort had at least 1 of these 3
clinical variables. With increasing presence of these
variables, DVT prevalence rose from 0% (95% CI,
0-4.2) to 16.4% (95% CI, 10.5-24.7). Two subsequent
multicenter post hoc analyses of pregnant patients
appear to externally validate these findings,106,107
with pooled statistics from all 3 studies revealing a
DVT prevalence of 0.8% (95% CI, 0.2-2.7) in patients
who lacked all 3 clinical features.107 The authors
of the study caution that, with an upper CI limit >
2% for pooled DVT prevalence, this prediction rule
should not be used without adjunct D-dimer or
ultrasonographic testing.106 Prospective validation in
larger populations of pregnant patients remains the
next challenging step in order to evaluate whether
the current standard of initial imaging may similarly
evolve into a more selective strategy, as has progres-
sively become the case in nonpregnant patients.
Use Of Anticoagulant Agents In Pregnancy
Heparins, both unfractionated heparin and LMWH,
do not cross the placenta and are considered safe for
use in pregnancy, whereas VKAs cross the placenta,
are potentially teratogenic, and can cause bleeding in
the fetus. Also, during pregnancy, there is a decreased
anticoagulant response to unfractionated heparin, as
well as the potential for heparin-induced thrombocy-
topenia. Therefore, according to both the ACCP and
ACOG recommendations, the treatment of choice for
DVT in pregnancy is LMWH (Grade 1B), with adjust-
ment of dosing to account for the pharmacokinetics
of pregnancy. Anticoagulation should be continued
through the first 6 weeks of the postpartum period
and for a minimum of 3 months (Grade 2C), although
studies supporting this recommendation tend to be
retrospective and observational in nature.4,109 Use
of LMWH is supported over VKAs for treatment of
VTE in all trimesters of pregnancy (Grade 1A in first
www.ebmedicine.net • March 2015
trimester, Grade 1B in second and third trimesters)
and during imminent delivery (Grade 1A).4 There are
limited data regarding the use of fondaparinux and
no data regarding the use of the novel oral anticoagu-
lants in pregnancy. Based on this current evidence,
the ACOG and ACCP guidelines recommend limit-
ing the use of fondaparinux to patients with severe
reactions to heparin (Grade 2C) and avoiding the use
of the novel oral anticoagulants during pregnancy
(Grade 1C).102,109
Upper Extremity Deep Venous Thrombosis
Upper extremity DVT tends to have a different etiol-
ogy, and it is much less common than lower extremity
DVT, occurring in 10% to 15% of documented cases
of VTE, with an incidence of approximately 0.4 to 1
per 10,000 persons per year.13,110 Complications from
upper extremity DVT are also less frequent, with a
lower risk of pulmonary embolism (4%-6% for upper
extremity DVT vs 15%-32% for lower extremity DVT),
fewer recurrences (2%-5% vs 19%), and lower rates of
postthrombotic syndrome (5% vs 29%).110,111
Anatomically, the deep venous system of the
upper extremity, from proximal to distal, consists of
the subclavian vein, which passes underneath the
clavicle and exits the thoracic outlet under the first
rib, becoming the axillary vein. The axillary vein then
passes underneath the teres major muscle, becom-
ing the brachial vein, which then courses inferiorly
alongside the brachial artery and the median nerve
medial to the humerus. Numerous anastomoses with
the superficial venous system (the basilic and cephalic
veins) occur before the veins enter the cubital fossa,
becoming the superficial veins of the forearm.
Twenty percent of upper extremity DVT is of
primary etiology, such as from venous thoracic outlet
obstruction as a result of anatomic abnormalities or
effort-induced repetitive microstress of the vessels
(Paget-Schroetter syndrome). The remaining 80% has
a secondary etiology, such as indwelling catheters or
pacemaker wires, surgery, or trauma.110 Older age
(average age of 38 years for upper extremity DVT vs
49 for lower extremity DVT), obesity, oral contracep-
tive use, thrombophilias (18% for upper extremity vs
39% for lower extremity), and malignancy are less
common etiologies of upper extremity DVT than
lower extremity DVT.110,111
Like lower extremity DVT, the presentation of
upper extremity DVT can be subtle; pain, swelling,
edema, paresthesias, discoloration, and weakness are
common signs and symptoms.110 However, unlike the
well-established diagnostic paradigm for the evalua-
tion of lower extremity DVT, there is a paucity of data
regarding the evaluation of upper extremity DVT.1
With ultrasonography emerging as the preferred
imaging test,110 meta-analysis has demonstrated an
overall sensitivity of 97% (95% CI, 90-100) and speci-
ficity of 96% (95% CI, 87-100), with no improvement
in accuracy upon adding color Doppler.112 Compres-
March 2015 • www.ebmedicine.net
sion ultrasonography is limited by the inability to
visualize or compress the proximal central deep veins
due to the overlying bony structures.110
Risk Stratification
A 2008 study developed and validated a clinical pre-
diction score for upper extremity DVT.113 The score
adds 1 point each for pain, edema, and the presence
of foreign venous material (eg, catheter or pacemak-
er), and subtracts 1 point if an alternative diagnosis is
more likely. A score of 0 or -1 is considered low risk,
although upper extremity DVT may be diagnosed in
13% of patients. A score of 1 is considered intermedi-
ate risk, with a 38% prevalence of upper extremity
DVT, while a score of ≥ 2 is considered high risk,
with a 69% prevalence of upper extremity DVT in the
validation study.113 The use of D-dimer testing has not
been well studied in the evaluation of upper extrem-
ity DVT, although one study of 52 patients reported a
sensitivity of 100% (95% CI, 78-100) and specificity of
14% (95% CI, 4-29).114 CT venography and MRI have
not been well studied in the evaluation of upper ex-
tremity DVT, but may have a role in the diagnosis of
the proximal upper extremity deep veins that are not
amenable to compression ultrasonography.1,115
Diagnosis
The current ACCP guidelines recommend initial
evaluation for upper extremity DVT with compres-
sion duplex ultrasonography (Grade 2C) followed
by testing with a moderate or highly sensitive
D-dimer; serial ultrasound; or imaging with venog-
raphy, CT, or MRI if there is a high clinical suspicion
and the initial ultrasound is negative (Grade 2C).
Upper extremity DVT can be excluded in patients
with a negative compression duplex ultrasound
and subsequent negative D-dimer, CT, MRI, or
venography (Grade 1C). Patients in whom the initial
compression duplex ultrasound was negative, with
a subsequent positive D-dimer, and those with a
suboptimal compression duplex ultrasound should
undergo venography (Grade 2B), unless there is a
compelling alternative diagnosis (Grade 2C).
A multicenter trial of 406 patients suspected of
having upper extremity DVT validated the use of a
diagnostic algorithm consisting of sequential appli-
cation of clinical risk stratification, D-dimer testing,
and compression ultrasonography. Using the clinical
decision rule derived and validated by Constans,113
patients were stratified as either unlikely (score of
≤ 1) or likely (score of ≥ 2) to have upper extrem-
ity DVT. Patients unlikely to have DVT underwent
initial testing with D-dimer, did not undergo further
testing if the test was negative, and proved a 0% rate
of upper extremity DVT on 3-month follow-up (95%
CI, 0.0-4.2). Patients who were likely to have DVT
underwent compression ultrasonography, followed
by combination with D-dimer testing if the ultra-
17 Reprints: www.ebmedicine.net/empissues
sound was negative. The combination of a negative
ultrasound and negative D-dimer resulted in a 1.2%
rate of upper extremity DVT at 3 months (95% CI,
0.0-6.5). The authors concluded that this diagnostic
paradigm was safe and effective in excluding upper
extremity DVT, with an overall miss rate of 0.4%
(95% CI, 0.0-2.2).116
Treatment
Since the goals of upper extremity DVT treatment
are the same as for lower extremity DVT – namely,
prevention of DVT propagation, PE, and recurrence,
as well as prevention of postthrombotic syndrome
– the treatment strategies have largely been extrapo-
lated from those for lower extremity disease. Strate-
gies include anticoagulation, thrombolysis, surgical
thrombectomy, superior vena cava filter placement,
and removal of any indwelling venous catheters.110
The current ACCP guidelines recommend initiation
of anticoagulation (Grade 1B) and continuation of
anticoagulation for at least 3 months (Grade 2B).
Risk Management Pitfalls For Deep Venous Thrombosis
(Continued on page 19)
1. “The patient was high-risk, but had a negative
D-dimer.”
Sole use of D-dimer tests in high-risk patients
is not recommended, as there is insufficient
evidence supporting their safety when they are
not used in combination with ultrasonography.
The majority of high-risk patients may have
a positive D-dimer for alternative reasons, so
starting with this test is also generally low-yield.
2. “Radiology confirmed DVT of the calf, so I
anticoagulated as I normally do.”
For routine, provoked, isolated distal DVT
without severe symptoms or risk factors, the risk
of bleeding posed by anticoagulation treatment
likely outweighs the risk of propagation of distal
thrombi into the proximal venous system. Thus,
current ACCP guidelines recommend serial
ultrasounds instead of routine anticoagulation
in uncomplicated distal DVT. However, this
decision may be discussed with patients to
determine their degree of comfort with this plan.
3. “Although cancer is a risk factor for DVT, his
Wells clinical score identified him as unlikely
to have DVT and he had a negative D-dimer.”
Based on subgroup analysis of patients with
active cancer, an unlikely Wells clinical score and
negative D-dimer resulted in a > 2% risk of acute
DVT. Thus, the combination does not appear
to be safe for use in this particular subset of
patients who harbor a higher risk for DVT.
Copyright © 2015 EB Medicine. All rights reserved. 18
LMWH is preferred over UFH (Grade 2C).4 Data
from 4 observational studies with a total of 209
patients demonstrate a 1.9% rate of recurrence after
treatment, with no cases of PE.110 Indwelling venous
catheters should not be removed as long as the need
for them persists and the catheters remain functional
(Grade 2C); however, anticoagulation should be con-
tinued for as long as the catheter remains in place
(Grade 2C).4 In a study of 74 patients with catheter-
related upper extremity DVT treated with LMWH
(dalteparin), no patients had catheter failure or
extension of DVT resulting from leaving the catheter
in place.117 Catheter-directed or systemic throm-
bolysis may be performed for patients with severely
symptomatic massive upper extremity DVT and a
low risk of bleeding (Grade 2C), but anticoagulation
is preferred over thrombolysis (Grade 2C). Catheter-
directed or surgical interventions should be reserved
for patients with persistent symptoms and failure of
anticoagulation or thrombolysis.4,110 Patients found
to have venous thoracic outlet obstruction may need
4. “I told the patient to follow up for a repeat ul-
trasound, but he took a long flight the follow-
ing day anyway. What else could I have done?”
After a negative proximal ultrasound, a patient
with moderate to high pretest probability of
disease still needs additional testing to safely
rule out DVT. While this has traditionally taken
the form of a serial ultrasound at 1 week, if the
patient is unable or unwilling to follow up, a
negative concurrent D-dimer test on the initial
ED visit or a negative whole-leg ultrasound
may obviate the need for follow-up testing.
Alternatively, a discussion with the patient
about the potential benefit versus harm of
empiric anticoagulation could be considered,
given his impending travel.
5. “The patient did not want to be admitted and
was afraid of needles for self-injecting LMWH,
so I decided to start him on dabigatran.”
The new oral anticoagulants have the
advantages of better compliance, a more reliable
efficacy of anticoagulation, less bleeding, and do
not require monitoring compared to treatment
with a VKA. However, dabigatran and edoxaban
both require a parenteral anticoagulation bridge.
Rivaroxaban and apixaban offer the benefit of
monotherapy during the initial anticoagulation
period and are more suitable choices in this
scenario.
www.ebmedicine.net • March 2015
to undergo surgical decompression, which, gener-
ally, has good outcomes.110
Controversies And Cutting Edge
As discussed throughout this review, new strategies
in the management of DVT are numerous, with still
more anticipated in the near future.
The Wells clinical score has undergone several
rounds of modifications over the years, improving its
ease of use and expanding its applicability to patients
with prior DVT. An even further modification with
stronger emphasis on the higher risk of DVT in active
malignancy may further fine tune this commonly
used risk stratification tool. Despite its standardized
scoring rubric, its interobserver variability is less
consistent in external studies and it may not necessar-
ily improve upon clinical gestalt for more experienced
providers. Higher D-dimer thresholds are being
tested in select populations, including the low-risk
group and the elderly. The use of D-dimers in the
Risk Management Pitfalls For Deep Venous Thrombosis
(Continued from page 18)
6. “The ultrasound was negative. How was
I supposed to know the patient still needed
another ultrasound a week later?”
A single initial proximal ultrasound in a patient
with moderate to high pretest probability of
DVT does not rule out the possibility of distal
thrombosis in the calf veins. Particularly in
the first 2 weeks, distal DVT may carry up
to a 25% risk of propagation to the proximal
venous system where it becomes more clinically
significant and poses the threat of PE.
7. “The patient had a prior DVT but was still
low-risk and had a negative D-dimer. I didn’t
know there were multiple versions of the Wells
clinical score.”
Be sure to add an additional point to the Wells
clinical score for history of previous DVT, as the
earlier versions excluded patients with suspected
recurrent DVT. Without this modification, an
unlikely score with a negative D-dimer can still
result in > 2% risk of acute DVT.
8. “The patient had cancer and did not want to be
admitted, so I offered LMWH and an oral VKA
as an outpatient regimen.”
Patients with cancer do not respond as well to
oral VKA therapy, which results in higher rates
of recurrent VTE compared with LMWH. First-
line recommended therapy for cancer patients
with acute DVT consists of daily LMWH
injections for a minimum of 6 months.
March 2015 • www.ebmedicine.net
workup for suspected DVT that is recurrent, occurs
during pregnancy, or affects the upper extremities is
actively debated, with new clinical prediction rules
being tested and validated in these special popula-
tions.
The use of ultrasonography has become com-
monplace with use of routine serial proximal ul-
trasounds to exclude propagation of potentially
undiagnosed distal DVT. However, patients at low
risk for DVT likely do not require serial imaging,
nor do all patients ultimately require imaging of the
whole leg. Controversy still surrounds the definition,
significance, risk of extension, treatment, and even the
need to diagnose isolated distal DVT. Selective serial
observation may be superior to routine treatment of
isolated distal DVT, and shorter courses of anticoagu-
lation may be as effective as standard therapy.
Current recommendations now advocate for
outpatient management of acute proximal DVT,
when possible.4,8 Although there is uncertainty
regarding the reversal strategy for the new oral
9. “The D-dimer was negative, though I don’t
know what kind of test my hospital uses.”
Knowing the sensitivity of the D-dimer test
used in one’s institution allows for better
interpretation of the result within the context
of a patient’s pretest probability of DVT. Some
laboratories still use semiquantitative tests,
which are faster and more specific, but they
perform at a lower sensitivity. While a patient
with low pretest probability for DVT can be
ruled out with a D-dimer of either moderate
or high sensitivity, only a highly sensitive test
is sufficient to rule out DVT in a patient with
moderate pretest probability of DVT without the
use of imaging.
10. “I had high suspicion, but the patient had 2
negative serial compression ultrasounds, so I
assumed the workup was complete.”
A patient with high pretest probability of DVT
and negative serial ultrasounds may still need to
undergo imaging of the pelvic veins, especially
if the swelling is extensive and includes the
thigh and buttocks. Of note, pregnant patients
with DVT have a higher prevalence (17%) of
isolated iliac vein thrombosis.
19 Reprints: www.ebmedicine.net/empissues
anticoagulants, they appear to pose a lower risk of
bleeding than the standard LMWH/VKA combina-
tion therapy. As experience with these medications
grows, a more comprehensive incorporation of the
new oral anticoagulants into treatment guidelines
for DVT can be anticipated in the coming years.
Disposition
Admission for parenteral anticoagulation was previ-
ously the expected disposition for most patients
newly diagnosed with acute DVT, but newer guide-
lines open the doors – and, in fact, show preference
– for outpatient treatment, when possible (Grade
1B).4,8 Assuming the patient has normal renal func-
tion, the appropriateness of an outpatient strategy
is strongly dependent on the patient’s reliability for
follow-up and associated social factors, including
potential ability to either learn how to self-inject
LMWH or to arrange for services to administer these
medications and to conduct laboratory testing.
The evidence behind the safety of outpatient
treatment is derived from systematic reviews of stud-
ies that randomized patients to either home treatment
(with LMWH) or inpatient treatment (with either
LMWH or UFH).118 Patients treated at home had a
lower recurrence of VTE (RR, 0.61; 95% CI, 0.42-.90)
and trended toward lower mortality (RR, 0.72; 95%
CI, 0.45-1.15) and fewer major bleeds (RR, 0.67-1.36).
Significant limitations to these studies, however,
include high rates of exclusion as well as the fact that
the majority of inpatients were treated with UFH.
Only 1 small study (N = 201) used LMWH in its in-
Cost-Effective Strategies
• Not every patient needs an ultrasound on initial
diagnostic workup. Risk stratification and a D-
dimer can preclude the need for ultrasound in
low-risk patients, and, if the D-dimer is highly
sensitive, in moderate-risk patients without ac-
tive cancer.
• Point-of-care, focused emergency ultrasound is
accurate in diagnosing proximal lower extremity
DVT and decreases length of stay in the ED.
• Not every patient needs a repeat proximal
ultrasound. Risk stratification into the low
pretest probability group can obviate the need
for further testing. For patients identified as
having moderate to high pretest probability, a
repeat ultrasound may be avoided if a concur-
rent D-dimer on the initial visit is negative or if
a whole-leg (instead of a proximal ultrasound) is
performed and results are negative.
• Consider outpatient treatment of acute DVT if
the patient has good follow-up, an adequate
home situation, and normal renal function.
Copyright © 2015 EB Medicine. All rights reserved. 20
patient arm, reporting no increase in combined rates
of VTE, PE, or major bleeding (3% vs 3.9%), while
effectively halving the overall costs from an economic
standpoint.119 Although the quality of evidence is
only moderate, the estimated savings of $500 to $2500
per patient are attractive enough, in the absence of ev-
idence of increased complications, to warrant a strong
recommendation from the ACCP for outpatient treat-
ment.4 Deliberation of disposition options should, of
course, incorporate individual patient preferences.120
If the patient’s home conditions and likelihood
of reliable follow-up are inadequate or the patient’s
renal function is limited, inpatient admission is indi-
cated for treatment of DVT. Patients with inadequate
creatinine clearance or a higher risk of bleeding, or
patients scheduled for procedures that could require
rapid reversal of anticoagulation, usually benefit
from parenteral UFH, rather than LMWH. Com-
plicated DVT may require alternatives to standard
anticoagulation, such as IVC filters, or intravenous
direct thrombin inhibitors in the case of heparin-
induced thrombocytopenia. Extensive clot burden
may, rarely, precipitate phlegmasia cerulean dolens,
which can require immediate operative thrombec-
tomy or thrombolysis.
Summary
DVT is a common disease with potential for mortal-
ity from PE and morbidity ranging from the com-
mon postthrombotic syndrome to rare loss of limb.
Although the clinical signs and symptoms of DVT
can be subtle, the diagnostic evaluation can benefit
from risk stratification of pretest probability, fol-
lowed by a systematic workup that includes ultra-
sonography and/or D-dimer testing. If there are
no contraindications, treatment of confirmed DVT
consists of a combination of initial and long-term
therapy for anticoagulation, the options for which
include UFH, LMWH, VKAs, and, more recently,
the novel oral anticoagulant agents. Future develop-
ments in the management of this disease include
fine-tuning risk stratification, particularly for certain
subsets of patients (ie, those with active cancer and
pregnant patients) and, pending postmarketing data
and availability of a reversal strategy, expansion of
use of the new oral anticoagulants as the treatment
of DVT increasingly shifts to the outpatient setting.
Case Conclusions
For the 42-year-old woman with a family history of protein
S deficiency, you initially recommended a repeat serial ul-
trasound at 1 week to rule out the possibility of distal clots
that could migrate proximally. Because her symptoms were
mild, there was likely no need for interim anticoagulation
between studies. However, the patient remained anxious,
so you offered the possibility of a same-day D-dimer test
www.ebmedicine.net • March 2015
instead. The test came back negative, and she was reassured
to follow up with her primary care doctor.
You decided to order a D-dimer test on the 22-year-
old man sent from the clinic. Although you knew that
your lab used SimpliRED®, you risk stratified the
patient into the low pretest probability group using the
Wells clinical score, and you were relieved to know that
a single negative test result – even if only moderately
sensitive – is sufficient to rule out DVT in an otherwise-
healthy patient.
For your third patient who wanted to leave before
his bed became available, you discussed the case with his
on-call primary doctor, and then discussed it with the
patient, outlining the benefits and risks (including the
risk of bleeding) of outpatient treatment with LMWH and
a VKA as compared to one of the new oral anticoagulants
(eg, rivaroxaban or apixaban). The patient, in conjunction
with his primary doctor, elected to be treated using a new
oral anticoagulant agent instead of learning how to self-
inject LMWH, and he committing to multiple follow-up
INR checks. You decided to select either rivaroxaban or
apixaban, since neither require a parenteral bridge during
the initial anticoagulation period.
References
Evidence-based medicine requires a critical ap-
praisal of the literature based upon study methodol-
ogy and number of subjects. Not all references are
equally robust. The findings of a large, prospective,
random­ized, and blinded trial should carry more
weight than a case report.
To help the reader judge the strength of each
reference, pertinent information about the study will
be included in bold type following the ref­erence,
where available. In addition, the most informative
references cited in this paper, as determined by the
authors, will be noted by an asterisk (*) next to the
number of the reference.
1.* Bates SM, Jaeschke R, Stevens SM, et al. Diagnosis of DVT:
antithrombotic therapy and prevention of thrombosis, 9th
ed: American College Of Chest Physicians evidence-based
clinical practice guidelines. Chest. 2012;141(2 Suppl):e351S-
e418S. (Practice guidelines)
2. Nicolaides AN, Fareed J, Kakkar AK, et al. Prevention and
treatment of venous thromboembolism -- International
Consensus Statement. Int Angiol. 2013;32(2):111-260. (Practice
guidelines)
3. Guyatt GH, Akl EA, Crowther M, et al. Executive summary:
antithrombotic therapy and prevention of thrombosis, 9th
ed: American College Of Chest Physicians evidence-based
clinical practice guidelines. Chest. 2012;141(2 Suppl):7S-47S.
(Practice guidelines)
4.* Kearon C, Akl EA, Comerota AJ, et al. Antithrombotic thera-
py for VTE disease: antithrombotic therapy and prevention
of thrombosis, 9th ed: American College Of Chest Physicians
evidence-based clinical practice guidelines. Chest. 2012;141(2
Suppl):e419S-e494S. (Practice guidelines)
5. Guyatt G, Gutterman D, Baumann MH, et al. Grading
strength of recommendations and quality of evidence in
clinical guidelines: report from an American College of Chest
March 2015 • www.ebmedicine.net
Physicians task force. Chest. 2006;129(1):174-181. (Review
article)
6. Guyatt GH, Norris SL, Schulman S, et al. Methodology for
the development of antithrombotic therapy and preven-
tion of thrombosis guidelines: antithrombotic therapy and
prevention of thrombosis, 9th ed: American College Of Chest
Physicians evidence-based clinical practice guidelines. Chest.
2012;141(2 Suppl):53S-70S. (Practice guidelines)
7. American College of Emergency Physicians. Clinical policy:
critical issues in the evaluation and management of adult
patients presenting with suspected lower-extremity deep
venous thrombosis. Ann Emerg Med. 2003;42(1):124-135.
(Practice guidelines)
8. Dupras D, Bluhm J, Felty C, et al. Institute for Clinical
Systems Improvement. Venous thromboembolism diagnosis
and treatment. Available at: https://www.icsi.org/guide-
lines__more/catalog_guidelines_and_more/catalog_guide-
lines/catalog_cardiovascular_guidelines/vte_treatment/.
Accessed February 8, 2015.
9. Bagot CN, Arya R. Virchow and his triad: a question of at-
tribution. Br J Haematol. 2008;143(2):180-190. (Review)
10. Lijfering WM, Rosendaal FR, Cannegieter SC. Risk factors
for venous thrombosis - current understanding from an
epidemiological point of view. Br J Haematol. 2010;149(6):824-
833. (Review)
11. Philbrick JT, Shumate R, Siadaty MS, et al. Air travel and
venous thromboembolism: a systematic review. J Gen Intern
Med. 2007;22(1):107-114. (Systematic review)
12. Goldhaber SZ. Risk factors for venous thromboembolism. J
Am Coll Cardiol. 2010;56(1):1-7. (Review)
13. Spencer FA, Emery C, Lessard D, et al. The Worcester venous
thromboembolism study: a population-based study of the
clinical epidemiology of venous thromboembolism. J Gen
Intern Med. 2006;21(7):722-727. (Retrospective observational
study; 587 patients)
14. Meignan M, Rosso J, Gauthier H, et al. Systematic lung scans
reveal a high frequency of silent pulmonary embolism in
patients with proximal deep venous thrombosis. Arch Intern
Med. 2000;160(2):159-164. (Prospective observational study;
622 patients)
15. Castellucci LA, Cameron C, Le Gal G, et al. Clinical and
safety outcomes associated with treatment of acute venous
thromboembolism: a systematic review and meta-analysis.
JAMA. 2014;312(11):1122-1135. (Systematic review)
16. Seinturier C, Bosson JL, Colonna M, et al. Site and clinical
outcome of deep vein thrombosis of the lower limbs: an epi-
demiological study. J Thromb Haemost. 2005;3(7):1362-1367.
(Prospective; 1913 patients)
17. White RH. The epidemiology of venous thromboembolism.
Circulation. 2003;107(23 Suppl 1):I4-I18. (Review)
18. Carrier M, Le Gal G, Wells PS, et al. Systematic review:
case-fatality rates of recurrent venous thromboembolism and
major bleeding events among patients treated for venous
thromboembolism. Ann Intern Med. 2010;152(9):578-589.
(Systematic review)
19.* Palareti G, Schellong S. Isolated distal deep vein thrombosis:
what we know and what we are doing. J Thromb Haemost.
2012;10(1):11-19. (Review article)
20.* Wells PS, Owen C, Doucette S, et al. Does this patient have
deep vein thrombosis? JAMA. 2006;295(2):199-207. (System-
atic review; 8239 patients)
21. Hull R, Hirsh J, Sackett DL, et al. Clinical validity of a nega-
tive venogram in patients with clinically suspected venous
thrombosis. Circulation. 1981;64(3):622-625. (Prospective; 160
patients)
22. ten Cate-Hoek AJ, van der Velde EF, Toll DB, et al. Common
alternative diagnoses in general practice when deep venous
thrombosis is excluded. Neth J Med. 2012;70(3):130-135. (Pro-
spective; 1028 patients)
23. Kahn SR. Frequency and determinants of the postthrombotic
21 Reprints: www.ebmedicine.net/empissues
 syndrome after venous thromboembolism. Curr Opin Pulm
Med. 2006;12(5):299-303. (Review article)
24. Goodacre S, Sutton AJ, Sampson FC. Meta-analysis: the
value of clinical assessment in the diagnosis of deep venous
thrombosis. Ann Intern Med. 2005;143(2):129-139. (Meta-
analysis)
25. Wells PS, Hirsh J, Anderson DR, et al. Accuracy of
clinical assessment of deep-vein thrombosis. Lancet.
1995;345(8961):1326-1330. (Prospective; 529 patients)
26. Wells PS, Anderson DR, Bormanis J, et al. Value of assess-
ment of pretest probability of deep-vein thrombosis in clini-
cal management. Lancet. 1997;350(9094):1795-1798. (Prospec-
tive; 593 patients)
27.* Wells PS, Anderson DR, Rodger M, et al. Evaluation of D-
dimer in the diagnosis of suspected deep-vein thrombosis. N
Engl J Med. 2003;349(13):1227-1235. (Prospective randomized
controlled; 1096 patients)
28. Wang B, Lin Y, Pan FS, et al. Comparison of empirical esti-
mate of clinical pretest probability with the Wells score for
diagnosis of deep vein thrombosis. Blood Coagul Fibrinolysis.
2013;24(1):76-81. (Prospective; 155 patients [calculation of
interobserver variability] and 405 patients [comparison of
clinical judgements vs Wells score])
29. Bigaroni A, Perrier A, Bounameaux H. Is clinical probability
assessment of deep vein thrombosis by a score really stan-
dardized? Thromb Haemost. 2000;83(5):788-789. (Prospective;
165 patients)
30. Dewar C, Corretge M. Interrater reliability of the Wells score
as part of the assessment of DVT in the emergency depart-
ment: agreement between consultant and nurse practitioner.
Emerg Med J. 2008;25(7):407-410. (Prospective; 100 patients)
31. Oudega R, Hoes AW, Moons KG. The Wells rule does not
adequately rule out deep venous thrombosis in primary care
patients. Ann Intern Med. 2005;143(2):100-107. (Prospective;
1295 patients)
32.* Geersing GJ, Zuithoff NP, Kearon C, et al. Exclusion of deep
vein thrombosis using the Wells rule in clinically important
subgroups: individual patient data meta-analysis. BMJ.
2014;348:g1340. (Meta-analysis; 10,002 patients)
33. Di Nisio M, Squizzato A, Rutjes AW, et al. Diagnostic ac-
curacy of D-dimer test for exclusion of venous thromboem-
bolism: a systematic review. J Thromb Haemost. 2007;5(2):296-
304. (Meta-analysis)
34. Bockenstedt P. D-dimer in venous thromboembolism. N Engl
J Med. 2003;349(13):1203-1204. (Review article)
35. Bates SM. D-dimer assays in diagnosis and management of
thrombotic and bleeding disorders. Semin Thromb Hemost.
2012;38(7):673-682. (Review article)
36. Olson JD, Cunningham MT, Higgins RA, et al. D-di-
mer: simple test, tough problems. Arch Pathol Lab Med.
2013;137(8):1030-1038. (Review article)
37. Linkins LA, Bates SM, Lang E, et al. Selective D-dimer
testing for diagnosis of a first suspected episode of deep
venous thrombosis: a randomized trial. Ann Intern Med.
2013;158(2):93-100. (Randomized, controlled; 1723 patients)
38. Linkins LA, Bates SM, Ginsberg JS, et al. Use of differ-
ent D-dimer levels to exclude venous thromboembolism
depending on clinical pretest probability. J Thromb Haemost.
2004;2(8):1256-1260. (Post hoc analysis; 571 patients)
39. van der Hulle T, Tan M, den Exter PL, et al. Selective D-di-
mer testing for the diagnosis of acute deep vein thrombosis:
a validation study. J Thromb Haemost. 2013;11(12):2184-2186.
(Retrospective; 389 patients)
40. Harper PL, Theakston E, Ahmed J, et al. D-dimer concentra-
tion increases with age reducing the clinical value of the
D-dimer assay in the elderly. Intern Med J. 2007;37(9):607-613.
(Retrospective; 6631 patients)
41. Righini M, Nendaz M, Le Gal G, et al. Influence of age on
the cost-effectiveness of diagnostic strategies for suspected
pulmonary embolism. J Thromb Haemost. 2007;5(9):1869-1877.
Copyright © 2015 EB Medicine. All rights reserved. 22
(Post hoc analysis of prospective trials; 1721 patients)
42. Schouten HJ, Geersing GJ, Koek HL, et al. Diagnostic accu-
racy of conventional or age adjusted D-dimer cut-off values
in older patients with suspected venous thromboembolism:
systematic review and meta-analysis. BMJ. 2013;346:f2492.
(Meta-analysis; 12,497 patients)
43. Douma RA, le Gal G, Sohne M, et al. Potential of an age
adjusted D-dimer cut-off value to improve the exclusion of
pulmonary embolism in older patients: a retrospective analy-
sis of three large cohorts. BMJ. 2010;340:c1475. (Retrospec-
tive; 5132 patients)
44. Douma RA, Tan M, Schutgens RE, et al. Using an age-
dependent D-dimer cut-off value increases the number of
older patients in whom deep vein thrombosis can be safely
excluded. Haematologica. 2012;97(10):1507-1513. (Retrospec-
tive; 2818 patients)
45. Schouten HJ, Koek HL, Oudega R, et al. Validation of two
age dependent D-dimer cut-off values for exclusion of deep
vein thrombosis in suspected elderly patients in primary
care: retrospective, cross sectional, diagnostic analysis. BMJ.
2012;344:e2985. (Retrospective; 1374 patients)
46. Hamblin AD, Cairns K, Keeling DM. The use of age-depen-
dent D-dimer cut-off values to exclude deep vein throm-
bosis. Reply to “Using an age-dependent D-dimer cut-off
value increases the number of older patients in whom deep
vein thrombosis can be safely excluded”. Haematologica.
2012;97(10):1507-13. (Retrospective; 6599 patients)
47. Righini M, Van Es J, Den Exter PL, et al. Age-adjusted D-
dimer cutoff levels to rule out pulmonary embolism: the
ADJUST-PE study. JAMA. 2014;311(11):1117-1124. (Prospec-
tive; 3346 patients)
48. Rabinov K, Paulin S. Roentgen diagnosis of venous throm-
bosis in the leg. Archives of Surgery. 1972;104(2):134-144.
(Review)
49. Kearon C, Julian JA, Newman TE, et al. Noninvasive
diagnosis of deep venous thrombosis. McMaster Diagnos-
tic Imaging Practice Guidelines Initiative. Ann Intern Med.
1998;128(8):663-677. (Systematic review)
50.* Goodacre S, Sampson F, Thomas S, et al. Systematic review
and meta-analysis of the diagnostic accuracy of ultrasonog-
raphy for deep vein thrombosis. BMC Med Imaging. 2005;5:6.
(Systematic review and meta-analysis; 100 cohort studies;
10,323 patients)
51. American Institute of Ultrasound in Medicine. AIUM
practice guideline for the performance of peripheral venous
ultrasound examinations. J Ultrasound Med. 2011;30(1):143-
150. (Practice guidelines)
52. Lensing AW, Prandoni P, Brandjes D, et al. Detection of
deep-vein thrombosis by real-time B-mode ultrasonography.
N Engl J Med. 1989;320(6):342-345. (Prospective consecutive
controlled study; 220 patients)
53. Kearon C. Natural history of venous thromboembolism.
Circulation. 2003;107(23 Suppl 1):I22-I30. (Review)
54. Philbrick JT, Becker DM. Calf deep venous thrombosis. A
wolf in sheep’s clothing? Arch Intern Med. 1988;148(10):2131-
2138. (Systematic review; 20 studies)
55. Elias A, Mallard L, Elias M, et al. A single complete ultra-
sound investigation of the venous network for the diagnostic
management of patients with a clinically suspected first epi-
sode of deep venous thrombosis of the lower limbs. Thromb
Haemost. 2003;89(2):221-227. (Prospective cohort study, 623
patients)
56. Schellong SM, Schwarz T, Halbritter K, et al. Complete
compression ultrasonography of the leg veins as a single test
for the diagnosis of deep vein thrombosis. Thromb Haemost.
2003;89(2):228-234. (Prospective observational cohort study;
1646 patients)
57. Gibson NS, Schellong SM, Kheir DY, et al. Safety and sensi-
tivity of two ultrasound strategies in patients with clinically
suspected deep venous thrombosis: a prospective manage-
www.ebmedicine.net • March 2015
 ment study. J Thromb Haemost. 2009;7(12):2035-2041. (Obser-
vational & randomized controlled study; 1002 patients)
58. Horner D, Hogg K, Body R, et al. Single whole-leg compres-
sion ultrasound for exclusion of deep vein thrombosis in
symptomatic ambulatory patients: a prospective observa-
tional cohort study. Br J Haematology. 2014;164(3):422-430.
(Prospective observational cohort study; 212 patients)
59. Bernardi E, Camporese G, Buller HR, et al. Serial 2-point ul-
trasonography plus D-dimer vs whole-leg color-coded Dop-
pler ultrasonography for diagnosing suspected symptomatic
deep vein thrombosis: a randomized controlled trial. JAMA.
2008;300(14):1653-1659. (Randomized controlled trial, 2465
patients)
60. Johnson SA, Stevens SM, Woller SC, et al. Risk of deep vein
thrombosis following a single negative whole-leg compres-
sion ultrasound: a systematic review and meta-analysis.
JAMA. 2010;303(5):438-445. (Systematic review and meta-
analysis)
61. American College of Emergency Physicians policy state-
ment: emergency ultrasound guidelines. Ann Emerg Med.
2009;53(4):550-570. (Clinical guidelines)
62. Frazee BW, Snoey ER, Levitt A. Emergency department
compression ultrasound to diagnose proximal deep vein
thrombosis. J Emerg Med. 2001;20(2):107-112. (Prospective
observational study, 76 patients)
63.* Pomero F, Dentali F, Borretta V, et al. Accuracy of emergency
physician-performed ultrasonography in the diagnosis of
deep-vein thrombosis: a systematic review and meta-anal-
ysis. Thromb Haemost. 2013;109(1):137-145. (Meta-analysis;
2379 patients)
64. Theodoro D, Blaivas M, Duggal S, et al. Real-time B-mode
ultrasound in the ED saves time in the diagnosis of deep
vein thrombosis (DVT). Am J Emerg Med. 2004;22(3):197-
200. (Prospective observational convenience study; 156
patients)
65. Cogo A, Lensing AW, Prandoni P, et al. Distribution
of thrombosis in patients with symptomatic deep vein
thrombosis. Implications for simplifying the diagnostic
process with compression ultrasound. Arch Intern Med.
1993;153(24):2777-2780. (Prospective cohort study; 562 pa-
tients)
66. McIlrath ST, Blaivas M, Lyon M. Patient follow-up after neg-
ative lower extremity bedside ultrasound for deep venous
thrombosis in the ED. Am J Emerg Med. 2006;24(3):325-328.
(Prospective observational study; 159 patients)
67. Thomas SM, Goodacre SW, Sampson FC, et al. Diagnostic
value of CT for deep vein thrombosis: results of a systematic
review and meta-analysis. Clin Radiol. 2008;63(3):299-304.
(Systematic review and meta-analysis; 13 studies; 1196
patients)
68. Goodman LR, Stein PD, Matta F, et al. CT venography and
compression sonography are diagnostically equivalent: data
from PIOPED II. AJR Am J Roentgenol. 2007;189(5):1071-1076.
(Prospective randomized controlled study; 711 patients)
69. Sampson FC, Goodacre SW, Thomas SM, et al. The accuracy
of MRI in diagnosis of suspected deep vein thrombosis: sys-
tematic review and meta-analysis. Eur Radiol. 2007;17(1):175-
181. (Systematic review and meta-analysis; 14 studies; 701
patients)
70. Meissner MH, Gloviczki P, Comerota AJ, et al. Early throm-
bus removal strategies for acute deep venous thrombosis:
clinical practice guidelines of the Society for Vascular
Surgery and the American Venous Forum. J Vasc Surg.
2012;55(5):1449-1462. (Clinical guidelines; systematic re-
view)
71. Erkens PM, Prins MH. Fixed dose subcutaneous low mo-
lecular weight heparins versus adjusted dose unfractionated
heparin for venous thromboembolism. Cochrane Database
Syst Rev. 2010(9):CD001100. (Meta-analysis; 9587 patients)
72. Buller HR, Davidson BL, Decousus H, et al. Fondaparinux
March 2015 • www.ebmedicine.net
or enoxaparin for the initial treatment of symptomatic deep
venous thrombosis: a randomized trial. Ann Intern Med.
2004;140(11):867-873. (Randomized double-blind trial; 2205
patients)
73. Garcia DA, Baglin TP, Weitz JI, et al. Parenteral anti-
coagulants: antithrombotic therapy and prevention of
thrombosis, 9th ed: American College Of Chest Physicians
evidence-based clinical practice guidelines. Chest. 2012;141(2
Suppl):e24S-e43S. (Review article)
74. EINSTEIN Investigators, Bauersachs R, Berkowitz SD, et al.
Oral rivaroxaban for symptomatic venous thromboembo-
lism. N Engl J Med. 2010;363(26):2499-2510. (Randomized
controlled trials; 3449 patients [acute treatment], 1196
patients [continued treatment])
75. EINSTEIN-PE Investigators, Buller HR, Prins MH, et al. Oral
rivaroxaban for the treatment of symptomatic pulmonary
embolism. N Engl J Med. 2012;366(14):1287-1297. (Random-
ized controlled trial; 4832 patients)
76. Prins MH, Lensing AW, Bauersachs R, et al. Oral rivaroxa-
ban versus standard therapy for the treatment of symp-
tomatic venous thromboembolism: a pooled analysis of
the EINSTEIN-DVT and PE randomized studies. Thromb J.
2013;11(1):21. (Pooled analysis of 2 randomized trials; 8282
patients)
77. Lefebvre P, Coleman CI, Bookhart BK, et al. Cost-effec-
tiveness of rivaroxaban compared with enoxaparin plus a
vitamin K antagonist for the treatment of venous thrombo-
embolism. J Med Econ. 2014;17(1):52-64. (Post hoc analysis)
78. Lee FM, Chan AK, Lau KK, et al. Reversal of new, factor-
specific oral anticoagulants by rFVIIa, prothrombin complex
concentrate and activated prothrombin complex concen-
trate: a review of animal and human studies. Thromb Res.
2014;133(5):705-713. (Systematic review)
79. Garcia P, Ruiz W, Loza Munarriz C. Warfarin initiation
nomograms for venous thromboembolism. Cochrane Database
Syst Rev. 2013;7:CD007699. (Systematic review; 494 patients)
80. Wells PS, Forgie MA, Rodger MA. Treatment of venous
thromboembolism. JAMA. 2014;311(7):717-728. (Review
article)
81 Schulman S, Kearon C, Kakkar AK, et al. Dabigatran versus
warfarin in the treatment of acute venous thromboembolism.
N Engl J Med. 2009;361(24):2342-2352. (Randomized double-
blind; 2539 patients)
82 Schulman S, Kakkar AK, Goldhaber SZ, et al. Treatment of
acute venous thromboembolism with dabigatran or warfarin
and pooled analysis. Circulation. 2014;129(7):764-772. (Ran-
domized, double-blind; 2589 patients)
83. Hokusai-VTE Investigators: Büller HR, Décousus H, Grosso
MA, et al. Edoxaban versus warfarin for the treatment of
symptomatic venous thromboembolism. N Engl J Med.
2013;369(15):1406-1415. (Randomized controlled trial; 4921
DVT and 3319 PE patients)
84. Zahir H, Brown KS, Vandell AG, et al. Edoxaban effects
on bleeding following punch biopsy and reversal by a
4-factor prothrombin complex concentrate. Circulation.
2015;131(1):82-90. (Double-blind randomized placebo-
controlled 2-way crossover study; 110 subjects)
85. Louzada ML, Majeed H, Wells PS. Efficacy of low-molecular-
weight-heparin versus vitamin K antagonists for long term
treatment of cancer-associated venous thromboembolism in
adults: a systematic review of randomized controlled trials.
Thromb Res. 2009;123(6):837-844. (Systematic review; 1158
patients)
86. Cosmi B, Legnani C, Tosetto A, et al. Usefulness of repeated
D-dimer testing after stopping anticoagulation for a first
episode of unprovoked venous thromboembolism: the
PROLONG II prospective study. Blood. 2010;115(3):481-488.
(Prospective; 355 patients)
87. Young T, Tang H, Hughes R. Vena caval filters for the pre-
vention of pulmonary embolism. Cochrane Database Syst Rev.
23 Reprints: www.ebmedicine.net/empissues
 2010. Feb 17;(2):CD006212. (Systematic review and meta-
analysis; 2 studies; 529 patients)
88. Kaufman JA, Kinney TB, Streiff MB, et al. Guidelines for
the use of retrievable and convertible vena cava filters:
report from the Society of Interventional Radiology mul-
tidisciplinary consensus conference. J Vasc Interv Radiol.
2006;17(3):449-459. (Clinical guidelines)
89. Angel LF, Tapson V, Galgon RE, et al. Systematic review of
the use of retrievable inferior vena cava filters. J Vasc Interv
Radiol. 2011;22(11):1522-1530. (Systematic review and meta-
analysis; 6834 patients)
90. The PREPIC Study Group. Eight-year follow-up of pa-
tients with permanent vena cava filters in the prevention of
pulmonary embolism: the PREPIC (Prevention du Risque
d’Embolie Pulmonaire par Interruption Cave) randomized
study. Circulation. 2005;112(3):416-422. (Prospective random-
ized study; 400 patients)
91. Watson L, Broderick C, Armon M. Thrombolysis for acute
deep vein thrombosis. Cochrane Database Syst Rev. 2014. Jan
23;1:CD002783. (Systematic review & meta-analysis; 17
studies; 1103 patients)
92. Laiho MK, Oinonen A, Sugano N, et al. Preservation of
venous valve function after catheter-directed and sys-
temic thrombolysis for deep venous thrombosis. Eur J Vasc
Endovasc Surg. 2004;28(4):391-396. (Retrospective study; 60
patients)
93. Augustinos P, Ouriel K. Invasive approaches to treatment
of venous thromboembolism. Circulation. 2004;110(9 Suppl
1):I27-I34. (Review)
94. Comerota AJ. The current role of operative venous
thrombectomy in deep vein thrombosis. Semin Vasc Surg.
2012;25(1):2-12. (Review)
95. Rathbun SW, Whitsett TL, Raskob GE. Negative D-dimer
result to exclude recurrent deep venous thrombosis: a
management trial. Ann Intern Med. 2004;141(11):839-845.
(Prospective; 300 patients)
96. Bates SM, Kearon C, Kahn SR, et al. A negative D-dimer
excludes recurrent deep vein thrombosis: results of a multi-
centre management trial [abstract]. Blood. 2007;110:214a.
(Prospective; 504 patients)
97. Aguilar C, del Villar V. Combined D-dimer and clinical
probability are useful for exclusion of recurrent deep venous
thrombosis. Am J Hematol. 2007;82(1):41-44. (Prospective; 105
patients)
98. Masuda EM, Kistner RL, Musikasinthorn C, et al. The
controversy of managing calf vein thrombosis. J Vasc Surg.
2012;55(2):550-561. (Systematic review)
99. Macdonald PS, Kahn SR, Miller N, et al. Short-term natural
history of isolated gastrocnemius and soleal vein thrombosis.
J Vasc Surg. 2003;37(3):523-527. (Prospective observational
cohort study; 185 patients)
100. Tainter CR, Huang AW, Strayer RJ. Fatal pulmonary embo-
lization after negative serial ultrasounds. J Emerg Med. 2014.
(Case report)
101. De Martino RR, Wallaert JB, Rossi AP, et al. A meta-analysis
of anticoagulation for calf deep venous thrombosis. J Vasc
Surg. 2012;56(1):228-237. (Systematic review)
102. Bates SM, Greer IA, Middeldorp S, et al. VTE, thrombophilia,
antithrombotic therapy, and pregnancy: antithrombotic
therapy and prevention of thrombosis, 9th ed: American
College of Chest Physicians evidence-based clinical practice
guidelines. Chest. 2012;141(2 Suppl):e691S-e736S. (Practice
guidelines)
103. Chan WS, Lee A, Spencer FA, et al. D-dimer testing in
pregnant patients: towards determining the next ‘level’ in
the diagnosis of deep vein thrombosis. J Thromb Haemost.
2010;8(5):1004-1011. (Prospective; 249 patients)
104. Chan WS, Spencer FA, Ginsberg JS. Anatomic distribution of
deep vein thrombosis in pregnancy. CMAJ. 2010;182(7):657-
660. (Systematic review; 124 patients)
Copyright © 2015 EB Medicine. All rights reserved. 24
105. Chan WS, Lee A, Spencer FA, et al. Predicting deep venous
thrombosis in pregnancy: out in “LEFt” field? Ann Intern
Med. 2009;151(2):85-92. (Prospective; 194 patients)
106. Righini M, Jobic C, Boehlen F, et al. Predicting deep venous
thrombosis in pregnancy: external validation of the LEFT
clinical prediction rule. Haematologica. 2013;98(4):545-548.
(Post hoc analysis of a prospective cohort; 157 patients)
107. Le Moigne E, Genty C, Meunier J, et al. Validation of the
LEFt score, a newly proposed diagnostic tool for deep vein
thrombosis in pregnant women. Thromb Res. 2014;134(3):664-
667. (Post hoc analysis of a prospective cohort; 96 patients)
108. Marik PE, Plante LA. Venous thromboembolic disease and
pregnancy. N Engl J Med. 2008;359(19):2025-2033. (Review
article)
109. James A. Practice bulletin no. 123: thromboembolism in
pregnancy. Obstet Gynecol. 2011;118(3):718-729. (Practice
guidelines)
110.* Kucher N. Deep-vein thrombosis of the upper extremities.
NEJM. 2011;364(9):861-869. (Review)
111. Lechner D, Wiener C, Weltermann A, et al. Comparison
between idiopathic deep vein thrombosis of the upper and
lower extremity regarding risk factors and recurrence. J
Thromb Haemost. 2008;6(8):1269-1274. (Prospective observa-
tional cohort study; 891 patients)
112. Di Nisio M, Van Sluis GL, Bossuyt PM, et al. Accuracy of
diagnostic tests for clinically suspected upper extremity
deep vein thrombosis: a systematic review. J Thromb Haemost.
2010;8(4):684-692. (Systematic review, 17 studies, 793 pa-
tients)
113. Constans J, Salmi LR, Sevestre-Pietri MA, et al. A clinical
prediction score for upper extremity deep venous thrombo-
sis. J Thromb Haemost. 2008;99(1):202-207. (Prospective, 354
patients)
114. Merminod T, Pellicciotta S, Bounameaux H. Limited useful-
ness of D-dimer in suspected deep vein thrombosis of the
upper extremities. Blood Coagul Fibrinolysis. 2006;17(3):225-
226. (Prospective observational study; 52 patients)
115. Desjardins B, Rybicki FJ, Kim HS, et al. ACR Appropri-
ateness Criteria® Suspected upper extremity deep vein
thrombosis. J Am Coll Radiol. 2012;9(9):613-619. (Clinical
guidelines)
116. Kleinjan A, Di Nisio M, Beyer-Westendorf J, et al. Safety and
feasibility of a diagnostic algorithm combining clinical prob-
ability, D-dimer testing, and ultrasonography for suspected
upper extremity deep venous thrombosis: a prospective
management study. Ann Intern Med. 2014;160(7):451-457.
(Prospective observational cohort study; 406 patients)
117. Kovacs MJ, Kahn SR, Rodger M, et al. A pilot study of central
venous catheter survival in cancer patients using low-
molecular-weight heparin (dalteparin) and warfarin without
catheter removal for the treatment of upper extremity deep
vein thrombosis (The Catheter Study). J Thromb Haemost.
2007;5(8):1650-1653. (Prospective observational cohort
study; 74 patients)
118. Othieno R, Abu Affan M, Okpo E. Home versus in-patient
treatment for deep vein thrombosis. Cochrane Database Syst
Rev. 2007(3):CD003076. (Systematic review; 1708 patients)
119. Boccalon H, Elias A, Chale JJ, et al. Clinical outcome and cost
of hospital vs home treatment of proximal deep vein throm-
bosis with a low-molecular-weight heparin: the Vascular
Midi-Pyrenees study. Arch Intern Med. 2000;160(12):1769-
1773. (Randomized controlled trial; 201 patients)
120. MacLean S, Mulla S, Akl EA, et al. Patient values and
preferences in decision making for antithrombotic therapy:
a systematic review: antithrombotic therapy and prevention
of thrombosis, 9th ed: American College Of Chest Physicians
evidence-based clinical practice guidelines. Chest. 2012;141(2
Suppl):e1S-e23S. (Systematic review)
www.ebmedicine.net • March 2015
CME Questions 5. Which of the following anticoagulants is safe
to give to a patient with renal insufficiency?
a. UFH
Take This Test Online! b. Warfarin
c. LMWH
Current subscribers receive CME credit abso- d. Dabigatran
lutely free by completing the following test. Each e. A and B
issue includes 4 AMA PRA Category 1 CreditsTM,
4 ACEP Category I credits, 4 AAFP Prescribed 6. Which of the following anticoagulants has
Take This Test Online!
credits, and 4 AOA Category 2A or 2B credits. proven methods of reversal?
Monthly online testing is now available for cur- a. Warfarin
rent and archived issues. To receive your free b. LMWH
CME credits for this issue, scan the QR code c. Fondaparinux
below with your smartphone or visit d. Rivaroxaban
www.ebmedicine.net/E0315.
7. Which of the following anticoagulants is con-
sidered first-line long-term therapy for patients
with active cancer?
a. UFH
b. LMWH
c. Fondaparinux
d. Oral VKA
1. Which of the following veins is part of the
superficial venous system, rather than the deep 8. In a patient with low pretest probability of
venous system? DVT, which of the following is NOT an appro-
a. Superficial femoral vein priate initial diagnostic test?
b. Greater saphenous vein a. Moderate-sensitivity D-dimer
c. Anterior tibial vein b. High-sensitivity D-dimer
d. Peroneal vein c. Proximal compression ultrasound
d. Whole-leg ultrasound
2. Which of the following features most increases
the likelihood of having an acute DVT? 9. In a patient with moderate pretest probability
a. Homans sign of DVT, which of the following is NOT an ap-
b. Calf swelling propriate initial diagnostic test?
c. High Wells score a. Moderate-sensitivity D-dimer
d. History of malignancy b. High-sensitivity D-dimer
c. Proximal compression ultrasound
d. Whole-leg ultrasound
3. Which of the following risk factors was a later
addition to the original Wells clinical score? 10. In a patient with high pretest probability of
a. Active malignancy DVT, which of the following is appropriate as
b. Prior history of DVT an initial diagnostic test?
c. Age > 75 years a. High-sensitivity D-dimer
d. Alternate diagnosis at least as likely as DVT b. Proximal compression ultrasound
c. Whole-leg ultrasound
4. In addition to the popliteal vein, which of the d. B or C
following veins is evaluated during focused e. A, B, or C
point-of-care emergency ultrasound evaluation
for DVT?
a. Posterior tibial vein
b. External iliac vein
c. Common femoral vein
d. Saphenous vein

March 2015 • www.ebmedicine.net 25 Reprints: www.ebmedicine.net/empissues

 Coming Soon In
Emergency Medicine Practice
Emergency Department
Evaluation And
Management Of Patients
With Upper Gastrointestinal
Bleeding
AUTHORS:
MATTHEW DELANEY, MD
Assistant Professor, Assistant Medical Director,
Department of Emergency Medicine, University of
Alabama at Birmingham, Birmingham, AL
CHRISTOPHER J. GREENE, MD
Department of Emergency Medicine, University of
Alabama at Birmingham, Birmingham, AL
Upper gastrointestinal bleeding in the emergency
department can occur from a wide variety
of conditions with a similarly wide range of
disease severity. While the initial evaluation and
stabilization is standard for nearly all causes of
bleeding, beyond these initial steps, it is crucial to
distinguish between bleeding from a variceal or
nonvariceal source. Treatments such as antibiotics
and somatostatin analogues may benefit patients
with variceal bleeding, while therapies such as
proton pump inhibitors have limited utility in
this subset of patients but may benefit those
bleeding from nonvariceal sources. Patients with
ongoing bleeding and hemodynamic instability
may benefit from emergent endoscopy. There
are several risk stratification scoring systems for
patients with upper gastrointestinal bleeding;
however, to date, there is limited evidence to
identify low-risk patients who are suitable
candidates for outpatient treatment.
Copyright © 2015 EB Medicine. All rights reserved. 26
Emergency Department
Management Of The Alcohol
Withdrawal Syndrome
AUTHORS:
GREG S. SWARTZENTRUBER, MD
Medical Toxicology Fellow, Division of Medical
Toxicology, Department of Emergency Medicine,
University of Pittsburgh Medical Center,
Pittsburgh, PA
JOSEPH H. YANTA, MD
Medical Toxicology Fellow, Division of Medical
Toxicology, Department of Emergency Medicine,
University of Pittsburgh Medical Center,
Pittsburgh, PA
Alcoholism is a prevalent medical and psychiatric
disease and, consequently, alcohol withdrawal
syndrome is encountered frequently in emergency
departments. Uncomplicated alcohol withdrawal,
or alcohol withdrawal tremor, is the most
common and least severe manifestation of
alcohol withdrawal syndrome; it can commonly
be managed on an outpatient basis with oral
benzodiazepines. Alcohol withdrawal seizure and
alcoholic hallucinosis are the first manifestations
of so-called complicated alcohol withdrawal.
They generally signify the need for inpatient
alcohol detoxification and, often, the use of
intravenous benzodiazepines. Delirium tremens
is the most severe and life-threatening form of
alcohol withdrawal. The key diagnostic criteria for
delirium tremens are an alteration in awareness or
attention (delirium) and tremor. Patients commonly
manifest hyperadrenergic signs and symptoms
that necessitate intensive care unit admission,
intravenous benzodiazepines, and, frequently,
adjunctive pharmacotherapy. An aggressive front-
loading approach with benzodiazepines is proposed
and the management of benzodiazepine-resistant
disease is addressed.
www.ebmedicine.net • March 2015
 Special Savings: The
Ultimate
LLSA Prep
The 2012-2015 Lifelong Learning
And Self-Assessment Study Guides
Receive FREE article reprints,* CME, and
more when you order yours today!
DESCRIPTION AMOUNT
The 2015 Lifelong Learning And Self-Assessment Study Guide $199 $50
$149 Off
The 2014 Lifelong Learning and Self-Assessment Study Guide $199 $50
$149 Off
The 2013 Lifelong Learning And Self-Assessment Study Guide $199 $50
$149 Off
The 2012 Lifelong Learning And Self-Assessment Study Guide $199 $50
$149 Off
• Full reprints of the original articles* FREE
• 35 AMA PRA Category 1 CreditsTM or 35 ACEP Category I CME Credits. FREE
• A handy summary of key points so you get the “must-know” information for each article. INCLUDED
• An in-depth discussion of each article to clarify and elaborate on the key points. INCLUDED
Includes • Sample questions to help you quiz yourself on your knowledge of the material. INCLUDED
FULL Article • Answers and explanations to the sample questions that drive home the main points. INCLUDED
Reprints* • A critical discussion and critique of the article that answers the question, INCLUDED
“What does this article really tell us?”
• 100% money-back guarantee: If, for any reason, you are not completely satisfied, INCLUDED
*Due to copyright restrictions, the tables simply call us to receive a full and immediate refund. No questions asked.
from one article in the 2014 LLSA and two
articles in the 2013 LLSA are not included.

2 E A S Y 1. Go online to:
www.ebmedicine.net/NHLBA
WAYS TO 2. Call 1-800-249-5770 or
ORDER 678-366-7933

Use Promotion Code: NHLBA at checkout to secure your discount

March 2015 • www.ebmedicine.net 27 Reprints: www.ebmedicine.net/empissues

 Physician CME Information
Date of Original Release: March 1, 2015. Date of most recent review: February 10,
2015. Termination date: March 1, 2018.
Accreditation: EB Medicine is accredited by the Accreditation Council for Continuing
Medical Education (ACCME) to provide continuing medical education for physicians.
This activity has been planned and implemented in accordance with the Essential
Areas and Policies of the ACCME.
Credit Designation: EB Medicine designates this enduring material for a maximum of 4
AMA PRA Category 1 Credits™. Physicians should claim only the credit commensurate
with the extent of their participation in the activity.
ACEP Accreditation: Emergency Medicine Practice is approved by the American College of
Emergency Physicians for 48 hours of ACEP Category I credit per annual subscription.
AAFP Accreditation: This Medical Journal activity, Emergency Medicine Practice, has
been reviewed and is acceptable for up to 48 Prescribed credits by the American
Academy of Family Physicians per year. AAFP accreditation begins July 31, 2014. Term
of approval is for one year from this date. Each issue is approved for 4 Prescribed
credits. Credit may be claimed for one year from the date of each issue. Physicians
should claim only the credit commensurate with the extent of their participation in the
activity.
AOA Accreditation: Emergency Medicine Practice is eligible for up to 48 American
Osteopathic Association Category 2-A or 2-B credit hours per year.

Emergency Needs Assessment: The need for this educational activity was determined by a survey
of medical staff, including the editorial board of this publication; review of morbidity
and mortality data from the CDC, AHA, NCHS, and ACEP; and evaluation of prior
Medicine Practice activities for emergency physicians.
Target Audience: This enduring material is designed for emergency medicine

Has Gone Mobile! physicians, physician assistants, nurse practitioners, and residents.
Goals: Upon completion of this activity, you should be able to: (1) demonstrate
medical decision-making based on the strongest clinical evidence; (2) cost-effectively
diagnose and treat the most critical presentations; and (3) describe the most common
You can now view all Emergency medicolegal pitfalls for each topic covered.
Medicine Practice content on your iPhone Discussion of Investigational Information: As part of the journal, faculty may be
presenting investigational information about pharmaceutical products that is outside
or Android smartphone. Simply visit Food and Drug Administration–approved labeling. Information presented as part of
this activity is intended solely as continuing medical education and is not intended to
www.ebmedicine.net from your mobile promote off-label use of any pharmaceutical product.
device, and you’ll automatically be Faculty Disclosure: It is the policy of EB Medicine to ensure objectivity, balance,
independence, transparency, and scientific rigor in all CME-sponsored educational
directed to our mobile site. activities. All faculty participating in the planning or implementation of a sponsored
activity are expected to disclose to the audience any relevant financial relationships
and to assist in resolving any conflict of interest that may arise from the relationship.
On our mobile site, you can: In compliance with all ACCME Essentials, Standards, and Guidelines, all faculty for
this CME activity were asked to complete a full disclosure statement. The information
• View all issues of Emergency Medicine Practice received is as follows: Dr. Woo, Dr. Goertz, Dr. Tainter, Dr. Jagoda, Dr. Damilini, Dr.
Toscano, and their related parties report no significant financial interest or other
since inception relationship with the manufacturer(s) of any commercial product(s) discussed
in this educational presentation. Dr. Godwin reports his paid advisory board
• Take CME tests for all Emergency Medicine Practice membership at Otsuka Holdings, Co., Ltd. and Bio-Signal Group Corp.
issues published within the last 3 years – that’s Commercial Support: This issue of Emergency Medicine Practice did not receive any
over 100 AMA Category 1 CreditsTM! commercial support.
Earning Credit: Two Convenient Methods: (1) Go online to www.ebmedicine.net/CME
• View your CME records, including scores, dates of and click on the title of the article. (2) Mail or fax the CME Answer And Evaluation Form
completion, and certificates (included with your June and December issues) to EB Medicine.
Hardware/Software Requirements: You will need a Macintosh or PC to access the online
• And more! archived articles and CME testing.
Additional Policies: For additional policies, including our statement of conflict of interest,
source of funding, statement of informed consent, and statement of human and animal
Check out our mobile site, and give us your rights, visit http://www.ebmedicine.net/policies.
feedback! Simply click the link at the bottom of the
mobile site to complete a short survey to tell us
what features you’d like us to add or change.

CEO & Publisher: Stephanie Williford Senior Business Analyst: Robin Williford Director of Editorial: Dorothy Whisenhunt
Content Editor & CME Director: Erica Carver Content Editor: Jenique Meekins Office Manager: Kiana Collier Member Services Representative: Paige Banks
Director of Business Development: Susan Woodard Account Manager: Cory Shrider Marketing Coordinator: Katherine Johnson

Direct all inquiries to: Subscription Information

EB Medicine Full annual subscription: $349 (includes 12 monthly evidence-based print
Phone: 1-800-249-5770 or 1-678-366-7933 issues; 48 AMA PRA Category 1 CreditsTM, 48 ACEP Category I credits, 48
Fax: 1-770-500-1316 AAFP Prescribed credits, and 48 AOA Category 2A or 2B CME credits; and
full online access to searchable archives and additional CME). Call 1-800-
5550 Triangle Parkway, Suite 150
249-5770 or go to www.ebmedicine.net/subscribe to subscribe.
Norcross, GA 30092
E-mail: ebm@ebmedicine.net Individual issues: $39 (includes 4 CME credits). Call 1-800-249-5770 or
Website: www.ebmedicine.net go to www.ebmedicine.net/EMPissues to order.
To write a letter to the editor, please email: Group subscriptions at heavily discounted rates are also available.
jagodamd@ebmedicine.net Contact Cory Shrider, Account Manager, at 678-366-7933 x 316 or
cs@ebmedicine.net for more information.

Emergency Medicine Practice (ISSN Print: 1524-1971, ISSN Online: 1559-3908, ACID-FREE) is published monthly (12 times per year) by EB Medicine (5550 Triangle Parkway, Suite 150,
Norcross, GA 30092). Opinions expressed are not necessarily those of this publication. Mention of products or services does not constitute endorsement. This publication is intended as
a general guide and is intended to supplement, rather than substitute, professional judgment. It covers a highly technical and complex subject and should not be used for making specific
medical decisions. The materials contained herein are not intended to establish policy, procedure, or standard of care. Emergency Medicine Practice is a trademark of EB Medicine. Copyright
© 2015 EB Medicine. All rights reserved. No part of this publication may be reproduced in any format without written consent of EB Medicine. This publication is intended for the use of the
individual subscriber only and may not be copied in whole or part or redistributed in any way without the publisher’s prior written permission — including reproduction for educational purposes
or for internal distribution within a hospital, library, group practice, or other entity.

Copyright © 2015 EB Medicine. All rights reserved. 28 www.ebmedicine.net • March 2015