Review Article
Diagnosis and management of upper extremity deep-vein thrombosis
in adults
Jonathan D. Grant1; Scott M. Stevens2,3; Scott C. Woller2,3; Edward W. Lee4; Stephen T. Kee4; David M. Liu5; Derek G. Lohan6;
C. Gregory Elliott2,3
1Divisionof Radiation Oncology, MD Anderson Cancer Center, Houston, Texas, USA; 2Department of Medicine, Intermountain Medical Center, Murray, Utah, USA; 3Department of
Medicine, University of Utah School of Medicine, Salt Lake City, Utah, USA; 4Division of Interventional Radiology, Department of Radiology, David Geffen School of Medicine
at UCLA, Los Angeles, California, USA; 5Department of Radiology, Faculty of Medicine, University of British Columbia, Vancouver, British Columbia, Canada; 6Department of
Radiology, Galway University Hospitals, Galway, Ireland
Summary
Upper extremity deep-vein thrombosis (UEDVT) is common and can
cause important complications, including pulmonary embolism and
post-thrombotic syndrome. An increase in the use of central venous ca-
theters, particularly peripherally inserted central catheters has been as-
sociated with an increasing rate of disease. Accurate diagnosis is essen-
tial to guide management, but there are limitations to the available evi-
dence for available diagnostic tests. Anticoagulation is the mainstay of
therapy, but interventional treatments may be considered in select situ-
ations. The risk of UEDVT may be reduced by more careful selection of
Correspondence to:
Scott M. Stevens, MD
Department of Medicine, Intermountain Medical Center
5169 South Cottonwood Street, Murray, UT 84107, USA
Tel.: +1 801 507 3747, Fax: +1 801 507 3350
E-mail: scott.stevens@imail.org
Background
Upper extremity deep-vein thrombosis (UEDVT) refers to the
formation of fibrin clots within the subclavian, axillary, and bra-
chial veins of the arm (1). Two proximal superficial veins, the ba-
silic and cephalic, have anastomoses with the deep veins and can be
the location of superficial thrombophlebitis.
“Primary UEDVT” refers to thrombosis without apparent pre-
cipitant, or in the setting of anatomic variants. An image of pri-
mary UEDVT is shown in 씰Figure 1. Thrombosis occurring in the
setting of a central venous catheter (CVC), malignancy, pregnancy,
recent surgery or trauma is termed “secondary UEDVT” (2). An
image of secondary UEDVT, associated with a PICC device, is
shown in 씰Figure 2.
UEDVT can cause symptoms, including pain, swelling, and skin
discoloration, or can be detected by imaging studies in asympto-
matic patients.
Epidemiology and risk factors
Approximately 10% of all DVT occurs in the upper extremities,
which translates into an annual incidence of 0.4 to 1 case per 10,000
© Schattauer 2012 1097
patients who receive central venous catheters and by use of smaller ca-
theters. Herein we review the diagnosis, management and prevention
of UEDVT. Due to paucity of research, some principles are drawn from
studies of lower extremity DVT. We present a practical approach to
Downloaded by: State University of New York at Stony Brook. Copyrighted material.
diagnosing the patient with suspected deep-vein thrombosis of the
upper extremity.
Keywords
Venous thrombosis, diagnosis management, deep-vein thrombosis,
magnetic resonance, risk factors
Received: June 1, 2012
Accepted after major revision: August 30, 2012
Prepublished online: October 23, 2012
doi:10.1160/TH12-05-0352
Thromb Haemost 2012; 108: 1097–1108
persons. The incidence is increasing with time, due to increasing
numbers of secondary UEDVT due to medical devices (2).
The majority of primary UEDVT cases are caused by anatomic
abnormalities of the costoclavicular junction. Thoracic outlet syn-
drome, a disorder in which the nerves, artery and veins are com-
pressed passing through the costoclavicular space, is associated
with DVT (3). Effort thrombosis results from repetitive injury to
the vein in a tight thoracic outlet induced by arm movements, es-
pecially frequent use of arms above shoulder level. This entity is
also referred to by the eponym Paget-Schroetter syndrome (PSS)
(4). Idiopathic cases, in which no anatomic abnormalities are
identified, also occur. Primary UEDVT comprises about 20–30%
of total UEDVT (5).
CVCs are the predominate cause of secondary UEDVT, and the
overall incidence of UEDVT is increasing coincident to the in-
creasing use of CVCs, particularly peripherally-inserted central ca-
theters (PICCs). A venous catheter is present in about half of all
cases of UEDVT (6–8). CVCs precipitate thrombus formation via
stasis, platelet adherence, and endothelial trauma (9). The inci-
dence of symptomatic and asymptomatic UEDVT following CVC
placement is 2–6% and 11–19%, respectively (씰Table 1) (10–22).
Catheter diameter and type, tip location, and concurrent infection
have all been shown to affect the risk of DVT. In a recent analysis of
2,014 CVC placements, larger-diameter triple-lumen PICCs were
Thrombosis and Haemostasis 108.6/2012
1098 Grant et al. Upper extremity DVT

shown to carry a 20-fold higher risk for UEDVT compared with
single-lumen PICCs (23). Peripheral misplacement of the CVC tip
has been associated with a 46% rate of thrombosis, and concomi-
tant infection confers a relative risk of up to 17.6. PICCs appear to
carry higher risk of DVT than surgically tunneled catheters (10, 24,
25). Cardiac pacemaker systems with intravenous leads have also
been associated with DVT (26). Odds ratios (OR) for risk factors
are presented in 씰Table 2.
Malignancy is an independent risk factor for UEDVT, present in
about one-third of cases (6, 27). Cancer, especially of the lung and

Figure 1:
Primary UEDVT.
Multiple foci of thrombo-

Downloaded by: State University of New York at Stony Brook. Copyrighted material.
sis are seen in the right
subclavian vein with in-
numerable collaterals
draining the right upper
extremity to the central
venous outflow in this
33-year-old woman with
A B PSS and recurrent epi-
sodes of thrombosis.

A B

C D

Figure 2: Secondary UEDVT. 45-year-old woman requiring long-term IV
antibiotics with a 5-French double lumen PICC placed in the left basilic vein.
Four days later, patient developed left upper-extremity pain, US showing
non-occlusive thrombus in left axillary vein. A) Longitudinal gray scale US
shows presence of PICC line (arrow) in distal subclavian vein. B) Longitudinal
gray scale US shows echogenic mass (arrow) within the lumen of the axillary
vein, implying an acute thrombus. C) Transverse gray scale US shows a non-
occlusive hypoechoic mass (arrow) within the axillary vein. In the right
image, the vessel lumen shows no compressibility, indicating the presence of
an intra-luminal thrombus. D) Longitudinal colour Doppler US shows periph-
eral blood flow around the non-occlusive thrombus (arrow) with increased
central velocity due to the narrowed channel.

Thrombosis and Haemostasis 108.6/2012 © Schattauer 2012

 Grant et al. Upper extremity DVT 1099

Table 1: Incidence of Author Patients Surveillance-detected Symptomatic

catheter-related (n, indication) incidence incidence
thrombosis*.
Bern, 1990 82 (40). chemotherapy 23.2% (37.5%) 21.2% (32.5%)
Bonfils, 1996 78, chemotherapy 14.1% 3.4%
De Cicco, 1997 95, chemotherapy, TPN, or both 19%; 66% including fibrin sleeve 6.3%
Martin, 1999 60, critical care 11.6%; 47% including fibrin sleeve 1.7%
Luciani, 2001 145, chemotherapy 11.7% 2.8%
Mismetti, 2003 45, cancer care 16.9% 6.7%
Couban, 2005 255 (125), cancer care -- 4.3% (4.0%)
Verso, 2005 310 (155), chemotherapy 16.1% 18% 2.1% (3.1%)
Karthaus, 2006 439, chemotherapy (or 145) 3.7% (3.4%) + --
Niers, 2007 87 (46), chemotherapy 12.6% (9%) 1.1% (or 2.2%)
Cortelezzi, 2005 458, cancer care -- 1.5%

Downloaded by: State University of New York at Stony Brook. Copyrighted material.
Lee, 2006 444, cancer care -- 4.3%
Young, 2009 812 (404), chemotherapy -- 6% (6%)
*Numbers in parentheses represent rates in the control groups in studies which tested a prophylactic intervention. +Outcomes for
symptomatic and surveillance-detected thrombosis were not reported separately.

GI tract may further increase the risk of DVT in patients with a
CVC (8, 28). UEDVT may also herald an undiagnosed cancer--one
series found a 23.7% rate of occult malignancy in apparently un-
provoked UEDVT; higher than that found in unprovoked lower
extremity DVT (29).
The association between hereditary and acquired thrombophi-
lia with UEDVT is unclear, with varying rates (11–60%) of throm-
bophilia in DVT cases reported. The utility of screening for throm-
bophilia in cases of UEDVT is controversial (3, 30), and identifying
a hereditary thrombophilia does not clearly impact management
decisions (31). Multiple case reports of DVT of the upper extrem-
ity in pregnant women have been described in the setting of as-
sisted reproductive techniques and ovarian hyperstimulation syn-
drome (32).
Prognosis and complications
Overall two- and 12-month mortality rates following diagnosis of
UEDVT are 30 and 40%, respectively; however, these studies in-
cluded patients with significant comorbidities (33, 34). Patients
with PSS generally have good functional status with longer life ex-
pectancies.
The reported rate of symptomatic PE in the setting of UEDVT
ranges from 3–12.4% (씰Table 3) (9, 35–38). The asymptomatic
incidence is several-fold higher. A study by Prandoni et al. used
prospective imaging to show evidence of PE in 36% of patients
with confirmed UEDVT (35). Of patients enrolled in the RIETE
registry, 9% with UEDVT had symptomatic PE at presentation,
versus 29% of patients with DVT of the lower extremities. The rate
of new PEs during follow-up, however, was similar between these
groups, and those with UEDVT had higher three-month mortality
(11% vs. 7% in the lower extremities, 95% confidence interval [CI]
1.18–2.21) (39). As noted above, UEDVT, even in unprovoked
cases, is substantially less likely to recur than lower extremity DVT
(7, 35, 40, 41). The annualised recurrence rate after a first UEDVT
ranges from 2.3 to 4.7% (7, 27). Patients with thrombophilia may
have a higher yearly risk of symptomatic recurrence than those
without thrombophilia (4.4% vs. 1.6%) (33); but data are limited.
PTS is a late complication characterised by chronic pain, oede-
ma, and functional limitation of the affected arm as a result of per-
sistent obstruction and valvular reflux. The reported incidence of
PTS following UEDVT is 7–46%, a wide range reflecting the lack of
a standardised definition in relevant studies (42). Residual throm-
bosis has been associated with a four-fold increased risk of devel-
oping PTS following UEDVT (4), but it is not clear whether endov-
ascular interventions alter the rate of this outcome. While com-
pression garments decrease the risk of PTS of the legs following
DVT, no such studies have been performed for UEDVT. An evi-
dence-based practice guideline suggests against using compression
sleeves to prevent PTS of the upper extremity; though it does sug-
gest a trial of compression therapy to treat symptoms of estab-
lished PTS of the arm (31).
Prevention
Risk reduction
A large body of literature focuses on the prevention of DVT in vari-
ous patient groups, though few trials have focused specifically on
UEDVT. Evidence-based clinical practice guidelines from the
American College of Chest Physicians (ACCP) provide specific
recommendations for prevention of venous thromboembolism in

© Schattauer 2012 Thrombosis and Haemostasis 108.6/2012

1100 Grant et al. Upper extremity DVT

Table 2: Significant risk factors for development of UEDVT. Randomised studies evaluating the benefit of heparin-bonded
catheters in preventing catheter-related thrombosis have centered
Author Risk factor OR 95% CI P-value
on the paediatric population. Two prospective trials comparing
Joffe, 2004 Indwelling CVC 9.7 7.8–12.2 -
heparin-bonded with standard non-heparin bonded catheters
Joffe CVC within 30 days 7.3 5.79–9.21 - with surveillance ultrasound found a decrease in both radio-
Saber, 2011 Implanted port (vs. PICC) 0.43 0.23–0.8 0.008 graphic and symptomatic thrombosis as well as infection in criti-
Saber Subclavian vein insertion site 2.16 1.07–4.34 0.029 cally-ill children (46, 47). A more recent study limited to infants ≤1
(vs. upper arm vein) year old with congenital heart disease found no such benefit (48).
Saber Catheter tip not in right atrium 1.92 1.22–3.02 0.034
or SVC/atrial junction
Saber Prior DVT 2.03 1.05–3.92 0.004
Evans, 2010 9.92 5.08–21.25 <0.001
Pharmacologic prophylaxis
Evans Double lumen PICC 7.54 1.51->100 <0.05
(vs. single-lumen) The use of prophylactic anticoagulation for the prevention of
Evans Triple lumen PICC 19.50 3.45->100 <0.01 CVC-related UEDVT is controversial. While early studies reported

(vs. single-lumen)
CI, confidence interval; CVC, central venous catheter; OR, odds ratio; PICC, pe-
ripherally-
inserted central catheter; SVC, subclavian vein catheter.
a number of clinical situations, but do not differentiate strategies
specifically targeted to prevent UEDVT. Recommendations
broadly focus on assessment of thrombosis and bleeding risk, and
selection of mechanical or pharmacologic prophylaxis based on
this assessment (43). The National Comprehensive Cancer Net-
work (NCCN) has also published guidelines on DVT prevention,
but likewise does not specify specific guidance for prevention of
UEDVT (44).
Because the majority of UEDVT cases are associated with
CVCs, carefully selecting patients for CVC placement is essential.
CVCs should only be used when definitively indicated, and in
whom the benefit of a CVC outweighs the risks. As noted above,
the diameter of the catheter and location of the catheter tip both
influence the rate of subsequent DVT (23, 25). Therefore, clini-
cians should choose the smallest diameter catheter compatible
with the indication, and assure appropriate position of the catheter
tip in the superior vena cava (SVC) or at the cavo-atrial junction.
Duration of catheter use may relate to risk of DVT and catheters
should be promptly removed when no longer needed.
A statistically significant reduction in UEDVT from 3.0% to
1.9% was reported following a multi-component intervention
which included use of fewer multi-lumen PICC and a reduction in
the diameter of triple-lumen PICC from 6 to 5 French (45).
Table 3: Rates of symptomatic PE in patients with UEDVT.
Author Patients, n PE
Munoz, 2008 512 9%
Horattas, 1998 539 12.4%
De Cicco, 1997 63 3%
Hingorani, 1997 170 7%
Kooij, 1997 78 12%
Kerr, 1990 693 8%
Downloaded by: State University of New York at Stony Brook. Copyrighted material.
significant benefit, recent prospective randomised trials have not.
A meta-analysis of nine randomised clinical trials found a trend to-
wards decreased incidence of symptomatic DVT in cancer patients
with a CVC treated with prophylactic heparin or low-dose warfa-
rin, but these results were not statistically significant (relative risk
[RR] 0.43, 95% CI 0.18–1.06 for heparin and RR 0.62, 95% CI
0.30–1.7 for low-dose warfarin) (49). The ACCP and NCCN
guidelines recommend against routine use of anticoagulant pro-
phylaxis solely on the basis of an indwelling CVC (31, 50).
Diagnosis
Clinical presentation
The most common clinical features of UEDVT are unilateral arm
swelling, discomfort, and erythema. Additional signs include di-
lated superficial veins, dyspnea, low-grade fever, and the develop-
ment of superior vena cava syndrome (8). The most common signs
and symptoms of PSS are venous distention (100%), swelling of
the arm (93%), blue discoloration (77%) and aching pain with ex-
ercise (66%) (51). As a tight thoracic outlet may also compromise
arterial circulation, physical examination maneuvers such as
Adson’s test (neck is rotated ipsilaterally and extended) and
Wright’s test (arm is hyperabducted) work to compress the tho-
racic outlet and may result in a diminished radial pulse (52).
Clinical evaluation has low specificity (30–64%) (41, 53, 54). A
clinical pre-test probability score has been reported (53) but has
yet to be used in a prospective management study.
Laboratory testing
In contrast to diagnosis of lower extremity DVT, sensitive D-dimer
testing has not been prospectively tested in high-quality manage-
ment studies, and a low quantitative (or negative qualitative)
D-dimer cannot be used to exclude UEDVT (55).
Imaging studies are required when UEDVT is suspected.

Thrombosis and Haemostasis 108.6/2012 © Schattauer 2012

 Grant et al. Upper extremity DVT 1101

Diagnostic imaging

Research evaluation of diagnostic tests may broadly be classified as

accuracy or management studies. The former compares a diag-
nostic test against a reference standard, while the latter focuses on
clinical outcome over a pre-defined period of follow-up. In
contrast to suspected lower extremity DVT, few management
studies for suspected UEDVT have been performed and the litera-
ture is largely limited to accuracy studies. Some of the following
studies relate to thrombosis of the lower extremity and must be
extrapolated for application to the upper extremities.

A
Ultrasound

Downloaded by: State University of New York at Stony Brook. Copyrighted material.
Ultrasound (US) is the most commonly used initial test for sus-
pected UEDVT. It is widely available, non-invasive and does not re-
sult in exposure to radiation. Assessment for UEDVT can be per-
formed with B-mode US with compression (henceforth compres-
sion US) and with assessment of venous flow using Doppler or col-
our Doppler for central veins not amenable to direct compression.
씰Figure 3 shows an example of images obtained during US.
On B-mode imaging, acute thrombi may appear as areas of
variable echogenicity within the vessel lumen. Non-compressibil-
ity of a venous segment defines the presence of DVT. Compression
US is performed by applying pressure to the overlying tissues to
compress the visualised vein in the transverse plane. Veins should B
easily collapse under the applied pressure, with lack of expected ve-
nous collapse indicating a thrombus. Because this technique
requires direct manual compression, it cannot be used to evaluate
the centrally-situated brachiocephalic vein and superior vena cava
(SVC), nor the medial segment of the subclavian vein underlying
the bony clavicle.
Doppler and colour-flow Doppler US assess venous blood flow.
Absence of flow, particularly in a vein which cannot be subjected to
compression, suggests DVT (41, 56). Alterations of the normal
biphasic pattern on pulsed-wave Doppler flow analysis can also
suggest the presence of DVT. Patel et al. reported that absence of
biphasic flow had 75% sensitivity and 100% specificity (compared
to contrast venography) for DVT not found on preliminary gray-
scale or color Doppler examination (57). C
A systematic review by di Nisio et al. evaluated nine studies using Figure 3: Example of venous ultrasound. A 77-year-old man status post
US to diagnose UEDVT; 75% of studies used CV as the reference orthotopic heart transplant with a left internal jugular central venous ca-
standard (58). Overall, the methodological quality of the included theter developed left upper-extremity swelling. US showing occlusive throm-
studies was poor. Summary estimates of sensitivity/specificity for bus in cephalic and basilic veins. A and B) Longitudinal colour Doppler US
each modality are presented in 씰Table 4. Due to limited negative shows echogenic material in the lumen of the cephalic and basilic veins with
predictive value, a normal US does not exclude DVT when there is complete lack of flow. Spectral waveform is markedly dampened. C) Trans-
high clinical suspicion, and additional tests are required. A recent verse gray scale US shows echogenic material in basilic vein (left image). On
evidence-based clinical practice guideline suggests use of compres- compression, basilic vein does not obliterate (arrow- right image), suggest-
ing the presence of a thrombus. Note compressibility of neighboring struc-
sion and Doppler US (termed “Combined Modality US” in the
tures.
guideline) as the initial diagnostic test for suspected UEDVT (59).

© Schattauer 2012 Thrombosis and Haemostasis 108.6/2012

1102 Grant et al. Upper extremity DVT
Contrast venography
Contrast venography (CV) has been the accepted reference stan-
dard for suspected DVT; although it has never been tested in a
management study of suspected UEDVT. A management study of
patients with suspected lower extremity DVT and a technically ad-
equate, negative venogram revealed a subsequent rate of DVT or
PE of 1.2% (95% CI, 0.2%-4.4%) during three months of follow-
up.(60) CV is invasive, requires patient exposure to radiation and
intravenous contrast, and is technically demanding. CV is there-
fore broadly reserved for instances when the results of initial non-
invasive imaging are insufficient to reach a diagnostic conclusion--
such as a technically inadequate US, or when an ultrasound is
negative for DVT but a high clinical suspicion for DVT persists. CV
visualises the veins at the thoracic outlet and can be used to con-
firm the diagnosis of tight thoracic outlet and planning of possible
interventional therapy (2).
To perform CV, contrast is injected into a distal vein of the af-
fected arm. Thrombus is defined by a constant intraluminal filling
defect present on more than one view. Non-filling of a venous seg-
ment on repeat injections suggests thrombosis, but is not diag-
nostic (61). Inter-observer agreement for interpreting venograms
in suspected UEDVT is 71–83% (kappa 0.48–0.71) (62).
In studies of suspected lower extremity DVT, CV are technically
inadequate in 10–15% of cases, largely due to insufficient contrast
filling and poor image quality (63). Corresponding failure rates for
venography for suspected UEDVT are not available.
In a small accuracy study with 22 subjects, CO2 venography,
which does not require iodinated contrast, had a sensitivity of 97%
and specificity of 85% versus standard CV (64). Further research of
this method is needed before it can be considered for clinical use.
Computed tomography venography
Computed tomography venography (CTV) is performed by im-
aging veins during the equilibrium phase of the contrast injection.
A recent systematic review of CTV in the diagnosis of lower-ex-
tremity DVT reported a pooled sensitivity of 95.9% (CI 93% to
97.8%) and specificity of 95.2% (CI 93.6–96.5%) (65). The refer-
ence standard was ultrasound in 12 and CV in one of the analysed
studies. Limited data exist regarding its use for the upper extrem-
ities. A small study by Kim et al. reported 100% concordance be-
tween CTV and standard CV in 27 patients with central venous ob-
Author Modality Sensitivity Specificity Reference modality
Di Nisio, 2010 Ultrasound 97% 96%
Compression 84% 94%
Colour-flow Doppler Duplex 91% 93%
Thomas, 2008 Computed tomography venography 95.9%) 95.2%
(lower extremity)
Sampson, 2007 Magnetic resonance venography (lower 91.5% 94.8%
extremity)
Thrombosis and Haemostasis 108.6/2012
struction (66). Drawbacks of CTV include additional ionising
radiation exposure to the patient, and risk associated with intra-
venous iodinated contrast. The relative performance of CTV vs.
CV in suspected tight thoracic outlet is not known. A sample image
from CTV is shown in 씰Figure 4.
Magnetic resonance venography
Magnetic resonance venography (MRV) may be performed using a
variety of techniques and methods of image processing, which may
be broadly categorised as contrast and non-contrast enhanced,
based on whether Gadolinium is utilised. MRV does not expose
patients to radiation or iodinated contrast; though methods which
Downloaded by: State University of New York at Stony Brook. Copyrighted material.
employ Gadolinium carry the risk of nephrogenic systemic fibro-
sis. MRV is able to produce images of the central veins of the tho-
rax, a region where ultrasound has limitations. In a recent meta-
analysis for MRV of the lower extremities, the pooled sensitivity
and specificity were 91.5% and 94.8%, respectively, with 13 of 14
studies using CV as a reference standard (67). While the possibility
of simultaneous evaluation of PE is attractive, a recent clinical trial
suggested that MRV is insufficiently sensitive to exclude suspected
PE (68). To date, however, no management studies of MRV for sus-
pected upper-extremity DVT have been performed.
MRV techniques which require contrast administration include
3-dimensional (3D) MRV, which utilises a systemic gadolinium
bolus injection and 3D spoiled gradient echo sequence with sub-
traction or timed imaging to eliminate arteries A small retrospec-
tive study reported good-quality images using partition analysis,
with accurate evaluation of central vein thrombosis in all seven
cases (69). 3D MRV requires a large amount of contrast agent. Di-
rect MRV uses a dilute bolus (typically 1:20 gadolinium dilution in
saline), which prevents T2 shortening and reduces the risk of toxic-
ity, injected distal to the vein of interest (70). The segment from the
axillary vein to the superior vena cava can typically be captured
with two 30-second breath holds (71). Two studies which evalu-
ated a total of 46 patients with suspected UEDVT reported a 100%
concordance between direct MRV and traditional imaging tech-
niques (mostly CV) (70, 71). In contrast, Baarslag et al., using non-
dilute contrast, reported only 50% sensitivity and 80% specificity
in a study of 17 patients comparing direct MRV with CV (72). An
example of a direct MRV image is shown in 씰Figure 5.
Techniques which do not require contrast include time-of-flight
(TOF) MRV which relies on the intrinsic properties of flowing blood
Table 4: Diagnostic
accuracy in UEDVT
Mostly contrast venography
(meta-analyses).
Mostly ultrasound (12 of 13 studies)
Mostly contrast venography (13 of 14
studies)
© Schattauer 2012
 Grant et al. Upper extremity DVT 1103

for signal acquisition. Studies evaluating TOF MRV for thrombosis

detection and prediction of CVC access sites in the upper extremities
report 71–97% sensitivity and 89–94% specificity (73–75). The ref-
erence standard for these studies, however, is inconsistent. TOF MRV
is clinically limited due to its long examination time and cumber-
some interpretation. Magnetic resonance direct thrombus imaging
(MRDTI) detects the formation of methemoglobin from hemoglo-
bin within a blood clot, which manifests as T1 shortening. MRDTI
can distinguish recent from chronic thrombus in the evaluation of
recurrent DVT, can visualise smaller vessels and complete occlu-
sions, and has a fast scanning time (76). No studies have evaluated
A
MRDTI for UEDVT. In a study of 101 patients with suspected DVT
of the lower extremities, the overall sensitivities and specificities were
96% and 94%, respectively (76).
A general approach to selecting a diagnostic modality for sus-

Downloaded by: State University of New York at Stony Brook. Copyrighted material.
pected UEDVT is provided in 씰Figure 6. An evidence-based clini-
cal practice guideline has been published, which includes a section
on diagnosis of UEDVT (59). Data for the usefulness of both CTV
and MRV to assess for UEDVT at this time are limited and the pos-
sible benefits of gadolinium in MRV as a means of avoiding X-ray
contrast nephrotoxicity should be weighed against its implication
in causing nephrogenic systemic fibrosis. High quality manage-
ment studies are needed. Therapeutic outcomes of various treat-
ments are presented in 씰Table 5.
B

Figure 4: Example of CTV study. Axial contrast-enhanced CTV of the chest

Treatment
Goals of therapy can be divided into the “active treatment” phase,
defined as the first three months following diagnosis, and the sec-
ondary prevention phase, which refers to therapy beyond three
months (77). Acute treatment phase goals are to prevent DVT pro-
gression and embolisation. Interventions in the acute phase may
also reduce the risk of post-thrombotic syndrome (PTS). The goal
of therapy in the secondary prevention phase is to prevent new epi-
sodes of DVT. Due to limited data regarding treatment of UEDVT,
many treatment recommendations are extrapolated from trials of
lower extremity DVT (1). In addition to anticoagulation, a number
of interventions may be useful in the treatment of select cases, the
indications for which continue to evolve.
Anticoagulation
Anticoagulation − typically consisting of therapeutically-dosed
parenteral anticoagulant bridged to an oral anticoagulant − is the
mainstay treatment for acute DVT (31). Parenteral anticoagu-
lation is achieved with therapeutically-dosed low-molecular-
weight heparin (LMWH), fondaparinux, subcutaneous unfrac-
tionated heparin or intravenous unfractionated heparin (UFH)
and is initiated at the time of diagnosis. Subcutaneous regimens
are suggested over intravenous UFH, except in cases of severe renal
insufficiency (31). Warfarin may be initiated on the same day, and
shows occlusive thrombus in the left subclavian vein (arrows) in a 46-year-
old woman with anticardiolipin antibodies.
the parenteral anticoagulant is continued for at least five days over-
lapping with warfarin, and until the International Normalised
Ratio (INR) is 2.0 or higher on two measurements, separated by at
least 24 hours (h) (31). The acute phase of treatment consists of
three months of therapeutic anticoagulation.
Though no randomised-controlled trials exist for treatment of
UEDVT, a retrospective analysis comparing conservative measures
to anticoagulation reported a 48% and 70% rate of symptomatic
resolution, respectively (78). Therapeutic anticoagulation de-
creases the risk of pulmonary embolism (PE) and recurrence by
preventing thrombus extension. Complications from anticoagu-
lation include a 2–4% rate of major bleeding (2).
Indications for extending anticoagulation therapy beyond three
months into the secondary prevention phase, and the optimal
treatment duration for UEDVT are uncertain. Unprovoked DVT
has a lower rate of recurrence in the upper extremity than in the
lower extremity (79–82) and therefore is not a compelling indi-
cation for long-term anticoagulation. When the provocation for
DVT is ongoing, such as active cancer, or when a CVC is not re-
moved, extended duration anticoagulation may be merited (31). A
guidance statement from the International Society of Thrombosis
and Haemostasis (ISTH) also suggests considering long-term anti-
coagulation for primary UEDVT with tight thoracic outlet which
has not been surgically corrected (83).

© Schattauer 2012 Thrombosis and Haemostasis 108.6/2012

1104 Grant et al. Upper extremity DVT
Novel oral anticoagulants, such as direct thrombin inhibitors
and Xa inhibitors have not yet been studied for the treatment of
UEDVT.
Endovascular therapy
Because anticoagulation has no direct thrombolytic effect, recanal-
ization depends on endogenous fibrinolysis. In a study of com-
bined upper and lower extremity DVT, 82% of patients treated
with anticoagulation alone had no change or worsening of the
thrombus load (84). Residual clot has been associated with com-
promised valvular function and a higher rate of PTS and DVT re-
currence (85). Various interventional therapies have been evalu-
ated as a means of attempting to accelerate the process of recanal-
ization, which has been hypothesised to reduce the risk of PTS and
perhaps the risk of DVT recurrence.
Thrombolysis
Thrombolysis involves the infusion of thrombolytic agents which
activate fibrin-bound plasminogen. Systemic thrombolysis has
largely been abandoned in favor of catheter-directed thrombolysis
(CDT) due to fewer bleeding complications and superior rates of
valvular competence. Studies have found no major differences in
safety or efficacy between urokinase, alteplase, or reteplase (86) .
CDT utilises a multi-port catheter embedded directly into the
thrombus under image guidance. Venous access is typically ob-
tained peripheral to the obstruction, and the thrombus is traversed
Figure 5: Example of MRV study. Contrast-enhanced MRV of the chest
and neck using 1.5 Tesla demonstrates filling of the SVC through intercostal
veins when a right-sided injection of 37 ml of gadolinium is performed
(arrow) in a 77-year-old woman with breast cancer.
Thrombosis and Haemostasis 108.6/2012
with a guidewire. For select cases of lower extremity DVT, the So-
ciety of Interventional Radiology recommends a continuous high-
volume drip regimen (25–100 ml/h) with dilute thrombolytic and
concomitant unfractionated heparin (86). However, no published
guideline specifies thrombolytic dosing for cases of UEDVT. When
undergoing CDT, the patient is typically monitored in an ICU or
stepdown unit and follow-up venograms are obtained every 8–24
h with repositioning of the infusion catheter into residual throm-
bus (87).
Published reports demonstrate initial thrombus clearance rates
between 72 and 91% for CDT of DVT of the upper extremities and
torso (88, 89). Organised clots older than two weeks are generally
less susceptible to thrombolysis (85). No randomised comparisons
between thrombolysis and anticoagulation exist, but a retrospec-
tive long-term analysis of 95 UEDVT patients showed a 60% ad-
Downloaded by: State University of New York at Stony Brook. Copyrighted material.
justed reduced risk for residual thrombosis compared with anti-
coagulation alone (95% CI 0.2–0.9) (90). However, there was no
significant difference detected between these groups in the fre-
quency of PTS.
The most common complication of thrombolysis is bleeding,
typically at the access site. In large cohorts examining CDT of
lower extremity DVT, published rates of major bleeding for CDT
of the lower extremities are 8–11% (85, 91). This rate has decreased
in recent years with improved techniques and better patient selec-
tion (92). Risk factors for bleeding include recent surgery, liver dys-
function, thrombocytopenia, advanced age, and prior stroke (85).
In a recent study, the presence of malignancy did not affect the fre-
quency of bleeding complications (89).
Because thrombolysis increases the risk for major bleeding, and
because high-quality evidence demonstrating long-term reduc-
tion of recurrent DVT and PTS symptoms are lacking, appropriate
case selection is challenging. Evidence based practice guidelines
from the ACCP suggest thrombolysis be considered in cases meet-
ing the following criteria: “severe symptoms, thrombus involving
most of the subclavian vein and the axillary vein, symptoms for14
days, good functional status, life expectancy of 1 year, and low risk
for bleeding”(31). NCCN guidelines recommend CDT in appro-
priate candidates based on institutional expertise (44).
Percutaneous mechanical thrombectomy
Percutaneous mechanical thrombectomy (PMT) refers to a group
of catheter-based devices that remove clot through aspiration,
fragmentation, or maceration. PMT can be used alone or in com-
bination with pharmacologic thrombolysis (pharmacomechanical
catheter-directed thrombolysis or PCDT) (87). There are two
single-session systems in broad use. The AngioJet (Possis Medical,
Minneapolis, MN, USA) which macerates the thrombus with a
high-velocity jet and the Trellis-8 (Covidien, Boulder, CO, USA)
which utilises a mechanical dispersion wire (85, 87).
Published data for these devices in the treatment of UEDVT are
very limited. In a report of eight patients with subclavian and axil-
lary vein thrombosis treated with the AngioJet Powerpulse system,
© Schattauer 2012

Figure 6: Suggested diagnostic algorithm for suspected UEDVT.
Adapted from Bates et al. (32). US = ultrasound; CV= contrast venography;
CTV= computed tomography venography; MRV=magnetic resonance ve-
nography. *Suspicion is based on clinician gestalt as formal clinical predic-
all patients were treated in a single setting and had patent central
veins at six-month follow-up. Two of the eight procedures were
complicated by puncture site haematomas (93).
Published rates of major bleeding from PCDT are 3–4% for
DVT of both upper and lower extremities (85). Other potential
complications include endothelial damage, traumatic haemolysis,
and pulmonary microemboli. As with CDT, the balance of risks
and benefits remain uncertain and guidelines suggest similar prin-
ciples be applied regarding case selection.
Angioplasty and endovascular stenting
Balloon angioplasty of veins, along with endovascular stenting, has
undergone limited evaluation for UEDVT. A published series of 22
Table 5: Summary of Treatment
therapeutic outcomes
Anticoagulation
for UEDVT.
Catheter-directed thrombolysis
Pharmacomechanical catheter-
directed thrombolysis
SVC filter placement
© Schattauer 2012
Grant et al. Upper extremity DVT 1105
Downloaded by: State University of New York at Stony Brook. Copyrighted material.
tion tools to quantify the pretest probability of UEDVT are not well validated.
£
CTV or MRV may be chosen in lieu of CV based on institutional experience,
or patient-specific factors (such as contrast allergy which would favor MRV).
patients with PSS reported stent occlusion in all patients within six
weeks (94). The location of occlusion in PSS may subject stents to
repetitive forces in the arm which promote stent fracture, com-
pression, and migration. A series of 65 stent placements for central
and peripheral limb obstructions of various causes reported a
clinical success rate at the peripheral sites of 75% and 42% at 12
and 24 months, respectively. Most patients underwent repeat inter-
vention (95). Angioplasty use has been reported in conjunction
with surgical decompression for PSS (see below) (96). Indications
for these interventions are uncertain, and long-term outcome data
is lacking.
Unless contraindicated, a conventional course of anticoagu-
lation should follow all interventional treatments for UEDVT (31).
To date, studies of endovascular treatment of UEDVT have not
demonstrated a lower rate of PTS or recurrent DVT than anti-
coagulation alone.
Benefit Major complications, %
48–70% symptom resolution 2–4%
72–91% initial thrombus clearance 8–11% (data from lower extremities)
8/8 treated patients had patent central 3–4% (data from both upper and lower
veins at six months extremities)
Uncertain benefit 3.8%
Thrombosis and Haemostasis 108.6/2012
1106 Grant et al. Upper extremity DVT
SVC filter
The placement of a filter device in the SVC has been reported in pa-
tients with UEDVT and a contraindication to anticoagulation. No
Food and Drug Administration (FDA)-approved SVC filter exists.
Greenfield filters have been employed because of their smaller caval
footprint. The Gunther Tulip filter has been reported to carry a lower
relative risk of strut perforation (97). When utilised, an SVC filter is
typically deployed at the convergence of the left and right brachio-
cephalic veins, with the hooks positioned above the origin of the azy-
gous vein to avoid occlusion in the event of filter thrombosis. An SVC
diameter greater than 28 mm is considered a contraindication by
most authors due to the risk of filter migration (6).
A recent systematic review by Owens et al. analysed 209 pub-
lished SVC filter placements (6). The major complication rate was
3.8%, largely caused by filter strut perforation through the SVC; in
some cases leading to cardiac tamponade, aortic perforation, and
pneumothorax. Given the known risks, and uncertain benefit pro-
vided by an SVC filter against PE, they have little role in the man-
agement of UEDVT. Evidence-based practice guidelines suggest
SVC filters should be reserved for “exceptional circumstances in
specialised centres”(31).
Treatment modification based on type and location
of DVT
PSS/primary UEDVT
Surgical correction with resection of the first rib aims to correct the
extrinsic vein compression in PSS, which occurs at the thoracic
outlet, with the goal of reducing the risk of recurrent thrombo-
sis. The rationale for surgical correction arise from the observation
that anticoagulation was associated with persistent symptoms and
recurrent thrombosis in more than 50% of cases (98). Urschel and
Razzuk reported outcomes in a registry of 294 patients with a
mean age of 32 years in three groups (initial anticoagulation alone,
initial CDT, initial CDT plus prompt first rib resection). The latter
groups required fewer subsequent thrombectomies, but assess-
ments and interventions were not dictated by a study protocol and
were performed at clinician discretion (51). As noted above, angio-
plasty use has been described in combination with surgical decom-
pression (96). ACCP guidelines suggest that surgical correction be
reserved for highly select cases in specialised centres (31). High-
quality studies assessing long-term outcomes are lacking.
Secondary UEDVT due to CVC
In CVC-related UEDVT, an initial treatment decision is whether to
remove the CVC. A prospective study of 72 cancer patients with
symptomatic CVC-related thrombosis evaluated the consequences
of anticoagulation while leaving the catheter in place and continu-
ing to use it. Patients were treated with dalteparin for 5–7 days as a
bridge to oral warfarin, and up to two administrations of tPA were
Thrombosis and Haemostasis 108.6/2012
allowed per blocked lumen. No recurrent DVTs or line blockages
were reported, and all 42 patients who had a persistent need main-
tained a functional CVC at three months (99). No events of PE
were reported. In the absence of infection or catheter damage,
guidelines support leaving a functional CVC in place if it has an
ongoing indication for use (31, 83). If the catheter is no longer
needed, however, the National Comprehensive Cancer Network
(NCCN) does recommend removal (44). The ACCP and ISTH
both recommend anticoagulation for three months when a CVC is
removed shortly after diagnosis of UEDVT in patients with or
without cancer (83). While not the subject of a specific guidance
statement, the ACCP text indicates that a CVC may be removed
either immediately after DVT diagnosis or after a short period of
initial anticoagulation, but favours the former (31). In patients
with cancer and an indwelling catheter, the NCCN recommends
Downloaded by: State University of New York at Stony Brook. Copyrighted material.
anticoagulation during catheter use and for a course of at least
three months after the catheter is removed (44).
Conclusions
The incidence of UEDVT is increasing, likely attributable to the in-
creasing use of CVCs. Pharmacologic prophylaxis has not proven
effective in reducing the risk of CVC associated UEDVT.
Clinical prediction scores and D-dimer assays have not under-
gone sufficient study, so imaging tests are required for diagnosis.
US is the initial imaging test recommended, and CV should be con-
sidered if a negative US is discordant with clinical suspicion. Data
regarding CTV and MRV are presently limited. High quality man-
agement studies of diagnostic strategies for suspected UEDVT are
needed.
Anticoagulation is the mainstay treatment for UEDVT, and du-
ration of anticoagulation is limited to the acute phase (three
months) in most cases. Interventional techniques are gaining sup-
portive evidence, but appropriate case selection is challenging.
Thrombolysis with or without mechanical thrombectomy can
achieve high rates of early vein patency, but whether this changes
long-term rates of PTS or DVT recurrence is presently not known.
UEDVT results in symptomatic PE in about one in 10 cases.
DVT recurs in about 2% annually and PTS in about a fifth of cases.
No preventive measures have been proven to reduce the risk of
PTS, but compression may improve symptoms in established
cases.
Evidence-based practice guidelines from the NCCN, last full
revision in 2008 and last updated in 2011, the ACCP, last updated
in 2012, and a guidance statement from the ISTH, published in
2012, are available. Further research should focus on high-quality
diagnostic management studies, case selection for long-term anti-
coagulation, use of novel anticoagulants, and long-term outcomes
from endovascular and surgical interventions.
Conflicts of interest
None declared.
© Schattauer 2012

References
1. Kearon C, et al. Antithrombotic therapy for venous thromboembolic disease:
American College of Chest Physicians Evidence-Based Clinical Practice Guide-
lines (8th Edition). Chest 2008; 133 (6 Suppl): 454S-545S.
2. Kucher N. Clinical practice. Deep-vein thrombosis of the upper extremities. N
Engl J Med 2011; 364: 861–869.
3. Linnemann B, et al. Hereditary and acquired thrombophilia in patients with
upper extremity deep-vein thrombosis. Results from the MAISTHRO registry.
Thromb Haemost 2008; 100: 440–446.
4. Joffe HV, Goldhaber SZ. Upper-extremity deep vein thrombosis. Circulation
2002; 106: 1874–1880.
5. Hingorani A, et al. Upper extremity deep venous thrombosis and its impact on
morbidity and mortality rates in a hospital-based population. J Vasc Surg 1997;
26: 853–860.
6. Owens CA, et al. Pulmonary embolism from upper extremity deep vein thrombo-
sis and the role of superior vena cava filters: a review of the literature. J Vasc Interv
Radiol 2010; 21: 779–787.
7. Spencer FA, et al. Upper extremity deep vein thrombosis: a community-based
perspective. Am J Med 2007; 120: 678–684.
8. Joffe HV, et al. Upper-extremity deep vein thrombosis: a prospective registry of
592 patients. Circulation 2004; 110: 1605–1611.
9. Kuter DJ. Thrombotic complications of central venous catheters in cancer pa-
tients. Oncologist 2004; 9: 207–216.
10. Luciani A, et al. Catheter-related upper extremity deep venous thrombosis in cancer
patients: a prospective study based on Doppler US. Radiology 2001; 220: 655–660.
11. Martin C, et al. Upper-extremity deep vein thrombosis after central venous cathe-
terization via the axillary vein. Crit Care Med 1999; 27: 2626–2629.
12. De Cicco M, et al. Central venous thrombosis: an early and frequent complication
in cancer patients bearing long-term silastic catheter. A prospective study.
Thromb Res 1997; 86: 101–113.
13. Bonfils P, et al. Prospective study of thromboses of the subclavian vein after setting
implantable venous access systems in cervico-facial cancerology. Ann Oto-laryn-
gol Chir Cervicofac 1996; 113: 425–429.
14. Bern MM, et al. Very low doses of warfarin can prevent thrombosis in central ve-
nous catheters. A randomized prospective trial. Ann Intern Med 1990; 112:
423–428.
15. Mismetti P, et al. Low-molecular-weight heparin (nadroparin) and very low doses
of warfarin in the prevention of upper extremity thrombosis in cancer patients
with indwelling long-term central venous catheters: a pilot randomized trial.
Haematologica 2003; 88: 67–73.
16. Couban S, et al. Randomized placebo-controlled study of low-dose warfarin for
the prevention of central venous catheter-associated thrombosis in patients with
cancer. J Clin Oncol 2005; 23: 4063–4069.
17. Verso M, et al. Enoxaparin for the prevention of venous thromboembolism as-
sociated with central vein catheter: a double-blind, placebo-controlled, random-
ized study in cancer patients. J Clin Oncol 2005; 23: 4057–4062.
18. Karthaus M, et al. Dalteparin for prevention of catheter-related complications in
cancer patients with central venous catheters: final results of a double-blind,
placebo-controlled phase III trial. Ann Oncol 2006; 17: 289–296.
19. Niers TM, et al. Prevention of catheter-related venous thrombosis with nadropa-
rin in patients receiving chemotherapy for hematologic malignancies: a random-
ized, placebo-controlled study. J Thromb Haemost 2007; 5: 1878–1882.
20. Cortelezzi A, et al. Incidence of thrombotic complications in patients with hae-
matological malignancies with central venous catheters: a prospective multi-
centre study. Br J Haematol 2005; 129: 811–817.
21. Lee AY, et al. Incidence, risk factors, and outcomes of catheter-related thrombosis
in adult patients with cancer. J Clin Oncol 2006; 24: 1404–1408.
22. Young AM, et al. Warfarin thromboprophylaxis in cancer patients with central ve-
nous catheters (WARP): an open-label randomised trial. Lancet 2009; 373:
567–574.
23. Evans RS, et al. Risk of symptomatic DVT associated with peripherally inserted
central catheters. Chest 2010; 138: 803–810.
24. van Rooden CJ, et al. Infectious complications of central venous catheters increase
the risk of catheter-related thrombosis in hematology patients: a prospective
study. J Clin Oncol 2005; 23: 2655–2660.
25. Saber W, et al. Risk factors for catheter-related thrombosis (CRT) in cancer pa-
tients: a patient-level data (IPD) meta-analysis of clinical trials and prospective
studies. J Thromb Haemost 2011; 9: 312–319.
© Schattauer 2012
Grant et al. Upper extremity DVT 1107
26. Korkeila P, et al. Progression of venous pathology after pacemaker and cardio-
verter-defibrillator implantation: A prospective serial venographic study. Ann
Med 2008; 31: 1–8.
27. Isma N, et al. Upper extremity deep venous thrombosis in the population-based
Malmo thrombophilia study (MATS). Epidemiology, risk factors, recurrence risk,
and mortality. Thromb Res 2010; 125: e335–338.
28. Anderson AJ, et al. Thrombosis: the major Hickman catheter complication in pa-
tients with solid tumor. Chest 1989; 95: 71–75.
29. Girolami A, et al. Venous thromboses of upper limbs are more frequently associ-
ated with occult cancer as compared with those of lower limbs. Blood Coagul Fi-
brinolysis 1999; 10: 455–457.
30. Ellis MH, et al. Risk factors and management of patients with upper limb deep
vein thrombosis. Chest 2000; 117: 43–46.
31. Kearon C, et al. Antithrombotic Therapy for VTE Disease: Antithrombotic Ther-
apy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians
Evidence-Based Clinical Practice Guidelines. Chest 2012; 141 (2 Suppl):
e419S-494S.
32. Chan WS, Ginsberg JS. A review of upper extremity deep vein thrombosis in preg-
nancy: unmasking the 'ART' behind the clot. J Thromb Haemost 2006; 4:
Downloaded by: State University of New York at Stony Brook. Copyrighted material.
1673–1677.
33. Martinelli I, et al. Risk factors and recurrence rate of primary deep vein thrombo-
sis of the upper extremities. Circulation 2004; 110: 566–570.
34. Hingorani A, et al. Risk factors for mortality in patients with upper extremity and
internal jugular deep venous thrombosis. J Vasc Surg 2005; 41: 476–478.
35. Prandoni P, et al. The long term clinical course of acute deep vein thrombosis of
the arm: prospective cohort study. Br Med J 2004; 329: 484–485.
36. Horattas MC, et al. Changing concepts of deep venous thrombosis of the upper
extremity--report of a series and review of the literature. Surgery 1988; 104:
561–567.
37. Kooij JD, et al. Pulmonary embolism in deep venous thrombosis of the upper ex-
tremity: more often in catheter-related thrombosis. Netherl J Med 1997; 50:
238–242.
38. Kerr TM, et al. Upper extremity venous thrombosis diagnosed by duplex scan-
ning. Am J Surg 1990; 160: 202–206.
39. Munoz FJ, et al. Clinical outcome of patients with upper-extremity deep vein
thrombosis: results from the RIETE Registry. Chest 2008; 133: 143–148.
40. Lechner D, et al. Comparison between idiopathic deep vein thrombosis of the
upper and lower extremity regarding risk factors and recurrence. J Thromb Hae-
most 2008; 6: 1269–1274.
41. Prandoni P, et al. Upper-extremity deep vein thrombosis. Risk factors, diagnosis,
and complications. Arch Intern Med 1997; 157: 57–62.
42. Elman EE, Kahn SR. The post-thrombotic syndrome after upper extremity deep ve-
nous thrombosis in adults: a systematic review. Thromb Res 2006; 117: 609–614.
43. Kahn SR, et al. Prevention of VTE in nonsurgical patients: Antithrombotic Ther-
apy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians
Evidence-Based Clinical Practice Guidelines. Chest 2012; 141 (2 Suppl):
e195S-226S.
44. Wagman LD, et al. Venous thromboembolic disease. NCCN. Clinical practice
guidelines in oncology. J Natl Compr Canc Netw 2008; 6: 716–753.
45. Evans RS, et al. Reduction of Peripherally Inserted Central Catheter Associated
Deep Venous Thrombosis. Chest 2012; epub ahead of print.
46. Krafte-Jacobs B, et al. Catheter-related thrombosis in critically ill children: com-
parison of catheters with and without heparin bonding. J Ped 1995; 126: 50–54.
47. Pierce CM, et al. Heparin-bonded central venous lines reduce thrombotic and in-
fective complications in critically ill children. Inten Care Med 2000; 26: 967–972.
48. Anton N, et al. Heparin-bonded central venous catheters do not reduce thrombo-
sis in infants with congenital heart disease: a blinded randomized, controlled trial.
Pediatrics 2009; 123: e453–458.
49. Akl EA, et al. Thromboprophylaxis for patients with cancer and central venous ca-
theters: a systematic review and a meta-analysis. Cancer 2008; 112: 2483–2492.
50. Streiff MB. The National Comprehensive Cancer Center Network (NCCN)
guidelines on the management of venous thromboembolism in cancer patients.
Thromb Res 2010; 125 (Suppl 2): S128–133.
51. Urschel HC, Jr., Razzuk MA. Paget-Schroetter syndrome: what is the best manage-
ment? Ann Thorac Surg 2000; 69: 1663–1669.
52. Sanders RJ, et al. Diagnosis of thoracic outlet syndrome. J Vasc Surg 2007; 46:
601–604.
53. Constans J, et al. A clinical prediction score for upper extremity deep venous
thrombosis. Thromb Haemost 2008; 99: 202–207.
Thrombosis and Haemostasis 108.6/2012
1108 Grant et al. Upper extremity DVT
54. Knudson GJ, et al. Color Doppler sonographic imaging in the assessment of
upper-extremity deep venous thrombosis. Am J Roentgenol 1990; 154: 399–403.
55. Merminod T, et al. Limited usefulness of D-dimer in suspected deep vein throm-
bosis of the upper extremities. Blood Coagul Fibrinolysis 2006; 17: 225–226.
56. Baarslag HJ, et al. Prospective study of color duplex ultrasonography compared
with contrast venography in patients suspected of having deep venous thrombo-
sis of the upper extremities. Ann Intern Med 2002; 136: 865–872.
57. Patel MC, et al. Subclavian and internal jugular veins at Doppler US: abnormal
cardiac pulsatility and respiratory phasicity as a predictor of complete central oc-
clusion. Radiology 1999; 211: 579–583.
58. Di Nisio M, et al. Accuracy of diagnostic tests for clinically suspected upper extrem-
ity deep vein thrombosis: a systematic review. J Thromb Haemost 2010; 8: 684–692.
59. Bates SM, et al. Diagnosis of DVT: Antithrombotic Therapy and Prevention of
Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clini-
cal Practice Guidelines. Chest 2012; 141 (2 Suppl): e351S-418S.
60. Hull R, et al. Clinical validity of a negative venogram in patients with clinically
suspected venous thrombosis. Circulation 1981; 64: 622–625.
61. Rabinov K, Paulin S. Roentgen diagnosis of venous thrombosis in the leg. Arch
Surg 1972; 104: 134–144.
62. Baarslag HJ, et al. Deep vein thrombosis of the upper extremity: intra- and inter-
observer study of digital subtraction venography. Eur Radiol 2003; 13: 251–255.
63. Leizorovicz A, et al. The assessment of deep vein thromboses for therapeutic trials.
Angiology 2003; 54: 19–24.
64. Heye S, et al. Upper-extremity venography: CO2 versus iodinated contrast ma-
terial. Radiology 2006; 241: 291–297.
65. Thomas SM, et al. Diagnostic value of CT for deep vein thrombosis: results of a
systematic review and meta-analysis. Clin Radiol 2008; 63: 299–304.
66. Kim H, et al. Role of CT venography in the diagnosis and treatment of benign tho-
racic central venous obstruction. Korean J Radiol 2003; 4: 146–152.
67. Sampson FC, et al. The accuracy of MRI in diagnosis of suspected deep vein
thrombosis: systematic review and meta-analysis. Eur Radiol 2007; 17: 175–181.
68. Stein PD, et al. Gadolinium-enhanced magnetic resonance angiography for pul-
monary embolism: a multicenter prospective study (PIOPED III). Ann Intern
Med 2010; 152: 434–443, W142–3.
69. Oxtoby JW, et al. 3D gadolinium-enhanced MRI venography: evaluation of cen-
tral chest veins and impact on patient management. Clin Radiol 2001; 56:
887–894.
70. Tanju S, et al. Direct contrast-enhanced 3D MR venography evaluation of upper
extremity deep venous system. Diagn Interv Radiol 2006; 12: 74–79.
71. Thornton MJ, et al. A three-dimensional gadolinium-enhanced MR venography
technique for imaging central veins. AJR Am J Roentgenol 1999; 173: 999–1003.
72. Baarslag HJ, et al. Magnetic resonance venography in consecutive patients with
suspected deep vein thrombosis of the upper extremity: initial experience. Acta
Radiol 2004; 45: 38–43.
73. Baarslag HJ, et al. Diagnosis and management of deep vein thrombosis of the
upper extremity: a review. Eur Radiol 2004; 14: 1263–1274.
74. Rose SC, et al. MR angiography for mapping potential central venous access sites
in patients with advanced venous occlusive disease. Am J Roentgenol 1996; 166:
1181–1187.
75. Hartnell GG, et al. Magnetic resonance angiography of the central chest veins. A
new gold standard? Chest 1995; 107: 1053–1057.
76. Fraser DG, et al. Diagnosis of lower-limb deep venous thrombosis: a prospective
blinded study of magnetic resonance direct thrombus imaging. Ann Intern Med
2002; 136: 89–98.
77. Kearon C. A conceptual framework for two phases of anticoagulant treatment of
venous thromboembolism. J Thromb Haemost 2012; 10: 507–511.
78. Sajid MS, et al. Upper limb deep vein thrombosis: a literature review to streamline
the protocol for management. Acta Haematol 2007; 118: 10–18.
Thrombosis and Haemostasis 108.6/2012
79. Laerum F, Holm HA. Postphlebographic thrombosis: a double-blind study with
methylglucamine metrizoate and metrizamide. Radiology 1981; 140: 651–654.
80. Heijboer H, et al. Clinical utility of real-time compression ultrasonography for
diagnostic management of patients with recurrent venous thrombosis. Acta Radi-
ologica 1992; 33: 297–300.
81. Cronan JJ, Leen V. Recurrent deep venous thrombosis: limitations of US. Radiol-
ogy 1989; 170: 739–742.
82. Linkins LA, et al. Interobserver agreement on ultrasound measurements of resid-
ual vein diameter, thrombus echogenicity and Doppler venous flow in patients
with previous venous thrombosis. Thromb Res 2006; 117: 241–247.
83. Baglin T, et al. Duration of anticoagulant therapy after a first episode of unpro-
voked pulmonary embolus or deep vein thrombosis: guidance from the scientific
and standardization committee of the international society on thrombosis and
haemostasis. J Thromb Haemost 2012; 10: 698-702.
84. Goldhaber SZ, et al. Pooled analyses of randomized trials of streptokinase and he-
parin in phlebographically documented acute deep venous thrombosis. Am J
Med 1984; 76: 393–397.
85. Vedantham S. Interventions for Deep Vein Thrombosis: Reemergence of a Prom-
ising Therapy. Am J Med 2008; 121: S28-S39.
Downloaded by: State University of New York at Stony Brook. Copyrighted material.
86. Vedantham S, et al. Society of Interventional Radiology position statement: treat-
ment of acute iliofemoral deep vein thrombosis with use of adjunctive catheter-
directed intrathrombus thrombolysis. J Vasc Interv Radiol 2006; 17: 613–616.
87. Murphy EH, et al. Endovascular intervention for lower extremity deep venous
thrombosis. In: Peripheral Endovascular Interventions. 3rd ed. Springer, New
York; 2010; pp. 203–213.
88. Sharafuddin MJ, et al. Endovascular management of venous thrombotic diseases
of the upper torso and extremities. J Vasc Interv Radiol 2002; 13: 975–990.
89. Maleux G, et al. Catheter-directed thrombolytic therapy for thoracic deep vein
thrombosis is safe and effective in selected patients with and without cancer. Eur
Radiol 2010; 20: 2293–2300.
90. Enden T, et al. Catheter-directed thrombolysis vs. anticoagulant therapy alone in
deep vein thrombosis: results of an open randomized, controlled trial reporting
on short-term patency. J Thromb Haemost 2009; 7: 1268–1275.
91. Wicky ST. Acute deep vein thrombosis and thrombolysis. Tech Vasc Interv Radiol
2009; 12: 148–153.
92. Comerota AJ. Randomized trial evidence supporting a strategy of thrombus re-
moval for acute DVT. Semin Vasc Surg 2010; 23: 192–198.
93. Sheynzon V, et al. Power pulse spray pharmacomechanical thrombectomy for the
treatment of upper extremity deep vein thrombosis. J Vasc Interv Radiol 2008;
Supplement: S132.
94. Urschel HC, Jr., Patel AN. Paget-Schroetter syndrome therapy: failure of intra-
venous stents. Ann Thorac Surg 2003; 75: 1693–1696.
95. Oderich GS, et al. Stent placement for treatment of central and peripheral venous
obstruction: a long-term multi-institutional experience. J Vasc Surg 2000; 32:
760–769.
96. Schneider DB, et al. Combination treatment of venous thoracic outlet syndrome:
open surgical decompression and intraoperative angioplasty. J Vasc Surg 2004; 40:
599–603.
97. Sadaf A, et al. Significant caval penetration by the celect inferior vena cava filter:
attributable to filter design? J Vasc Interv Radiol 2007; 18: 1447–1450.
98. Adams JT, DeWeese JA. “Effort” thrombosis of the axillary and subclavian veins.
J Trauma 1971; 11: 923–930.
99. Kovacs MJ, et al. A pilot study of central venous catheter survival in cancer pa-
tients using low-molecular-weight heparin (dalteparin) and warfarin without ca-
theter removal for the treatment of upper extremity deep vein thrombosis (The
Catheter Study). J Thromb Haemost 2007; 5: 1650–1653.
© Schattauer 2012