You are on page 1of 6

The American Journal of Medicine (2005) Vol 118 (8A), 31S–36S

Endovascular management of acute deep vein thrombosis

Andrew Blum, MD,a Erin Roche, MSb
Division of Vascular Interventional Radiology, Midwest Heart Specialists, Elmhurst, Illinois, USA; and
Riverside, California, USA.

KEYWORDS: The goals of successful management of deep vein thrombosis (DVT) include relief of acute symptoms
Deep vein thrombosis; with restoration of venous patency, prevention of clot propagation and subsequent pulmonary embo-
Low-molecular-weight lism, and maintenance of venous valvular function. Valvular incompetence is the leading cause of
heparin; postthrombotic syndrome (PTS), which is characterized by chronic leg heaviness and aching, lower
Postthrombotic extremity edema, and impaired viability of subcutaneous tissues, which may lead to chronic trophic skin
syndrome; changes and venous ulceration.
Unfractionated heparin Anticoagulation with unfractionated or low-molecular-weight heparin followed by warfarin is rec-
ognized as the standard therapy for acute DVT. Although this approach may effectively prevent
recurrent thrombosis, it often fails to meet the other treatment goals. Recent studies have demonstrated
that early clot lysis through the use of catheter-directed thrombolytic therapy and other adjunctive endo-
vascular techniques rapidly restores venous patency, more effectively preserves valvular function, and
improves quality of life. When used in conjunction with anticoagulation, these minimally invasive endo-
vascular techniques have the potential to lead to improved long-term outcomes in patients with DVT.
© 2005 Elsevier Inc. All rights reserved.

Deep vein thrombosis (DVT) is a common condition, propagation and embolization. Anticoagulation typically
with ⬎250,000 new cases reported annually in the United consists of a short course of treatment with intravenous (IV)
States.1 When inadequately treated, DVT may be compli- unfractionated heparin (UFH) followed by a period of 3 to
cated acutely by the development of pulmonary embolism 6 months of oral warfarin.4 Results from several recent
(PE) and in the long-term by chronic venous insufficiency, randomized clinical trials demonstrate that outpatient anti-
also known as postthrombotic syndrome (PTS). As many as coagulant treatment with subcutaneous low-molecular-
7 million individuals experience complications of severe weight heparin (LMWH) is feasible, and may confer several
chronic venous disease,2 many cases of which are the direct advantages over inpatient treatment with IV UFH.5 The
result of previous DVT. The economic impact of compli- potential advantages of LMWH include once- or twice-daily
cations related to PTS have been estimated to account for as dosing due to its longer half-life and fixed dosages due to a
much as 75% of the total care cost of DVT.3 more predictable antithrombotic effect. The use of LMWH
therefore eliminates the need for laboratory monitoring,
reducing the costs associated with hospital stays in these
Current treatment of deep vein thrombosis patients and promoting earlier discharge.6
Although anticoagulation is useful in the reduction of
Historically, the standard treatment for DVT has been the
recurrent thrombotic events and the prevention of PE,7 used
administration of anticoagulant drugs to prevent thrombus
alone it is rarely able to facilitate clot lysis adequately.8
Preservation of valve function and prevention of PTS has
Requests for reprints should be addressed to Andrew Blum, MD,
Division of Vascular Interventional Radiology, Midwest Heart Specialists,
not been established with UFH use.8,9 Anticoagulation reg-
429 North York Road, Elmhurst, Illinois 60126. imens, along with their strengths and weaknesses, are fur-
E-mail address: ther explored by Merli elsewhere in this supplement.10 The

0002-9343/$ -see front matter © 2005 Elsevier Inc. All rights reserved.
32S The American Journal of Medicine, Vol 118 (8A), August 2005

objective of this article is to review the utility of thrombo- Table 1 Outcome of anticoagulation versus systemic
lytic therapy and other adjunctive endovascular techniques thrombolytic infusion for acute deep vein thrombosis: results
in the setting of more extensive iliofemoral DVT. of 13 studies


Goals of deep vein thrombosis treatment None/Worse Partial Significant/

Treatment (N) (%) (%) Complete (%)
Without adequate treatment, many patients with DVT ex- Heparin (254) 82 14 4
perience persistent venous outflow obstruction and valvular Thrombolysis (337) 37 18 45
incompetence, both of which are known to contribute to the
Adapted with permission from Semin Vasc Surg.13
development of PTS. It is estimated that 6 million to 7
million patients in the United States have chronic venous
insufficiency as a result of PTS, including about 500,000
lyzing the conversion of plasminogen into its active form,
who may eventually progress to ulceration.2 Patients with
plasmin, which in turn breaks down fibrin into soluble
more extensive DVT involving the proximal segments, in-
byproducts. In a randomized trial that compared thrombol-
cluding the iliac veins and inferior vena cava (IVC), are
ysis and anticoagulation with anticoagulation alone in pa-
more likely to develop PTS.
tients with iliofemoral DVT, thrombolysis was associated
Successful treatment of DVT and prevention of post-
with improved patency (72% vs. 12%; P ⬍0.001) and
thrombotic complications require realization of the follow-
venous valvular competence (89% vs. 59%; P ⫽ 0.04) at 6
ing 5 goals: (1) prevention of clot propagation; (2) preven-
months.14 Although most studies of venous thrombolysis
tion of PE and recurrent thrombosis; (3) restoration of
have used urokinase, more recent data suggest that treat-
venous patency and flow; (4) preservation of valvular func-
ment with tissue t-PA also may be effective.13
tion; and (5) elimination of clinical symptoms associated
Although systemic thrombolysis (via a peripheral IV
with PTS.
infusion) is more effective than treatment with anticoagu-
Although studies of new anticoagulation regimens focus
lation alone, this method results in significantly less throm-
on short-term goals, such as prevention of PE, few consider
bus dissolution in obstructed segments than in partially
the long-term aim of preservation of valvular function as a
thrombosed segments (10% vs. 52%).15 This is likely due to
significant therapeutic end point. It has been demonstrated
reduced diffusion of these agents into larger venous thrombi
that early recanalization of thrombi correlates with preven-
under conditions of little or no flow.13 There is also a greater
tion of valvular reflux and preservation of valve function.11
risk of bleeding associated with systemic thrombolysis than
These factors are considered to be highly important in the
with standard anticoagulation alone. The limitations of sys-
prevention of PTS12 and its associated complications, and
temic thrombolytic therapy have prompted studies of cath-
thrombolytic agents are better able than standard anticoag-
eter-directed, “local” infusions of thrombolytic agents in
ulation alone to achieve this end point (Table 1).2,8,13
order to enhance clot dissolution and minimize bleed-
ing.16 –18 These interventional techniques offer the advan-
tages of more concentrated delivery of the thrombolytic
Catheter-directed thrombolysis and agent into the clot, with less risk of systemic bleeding.
adjunctive endovascular treatment options These methods also allow for imaging of the affected veins
with the option of adjunctive endovascular techniques such
Practitioners who manage patients with DVT should be as stent placement to ensure optimal patency and flow.
familiar with the various interventional and surgical treat- In 1994, Semba and colleagues17 reported their initial
ment strategies, which include pharmacologic thrombolysis, experience with catheter-directed thrombolysis in 21 pa-
percutaneous mechanical thrombectomy, adjunctive stent- tients with iliofemoral DVT. These authors reported a tech-
ing, and surgical thrombectomy. The endovascular strate- nical and clinical success rate of 85%. Based on these
gies are often used in combination, allowing for better results, the National Venous Thrombosis Registry was ini-
resolution of venous clot burden than when a single modal- tiated, through which 287 patients with large obstructed
ity is used alone. Results with surgical thrombectomy have venous segments (71% with iliofemoral DVT) were re-
improved with the advent of the Fogarty balloon and other cruited from 63 participating centers and observed over a
refined techniques, yet the procedure has generally fallen period of 1 year after treatment with catheter-directed uroki-
out of favor because of the associated operative morbidity. nase. Complete resolution of thrombus was achieved in
Surgery remains a valid option, however, in the event of 31% of cases, and partial (50%–99%) thrombus resolution
impending venous gangrene or failure of endovascular tech- was reported in 52%.18 At 1 year, primary patency was
niques. 60%. The study enrolled patients with both acute and
Streptokinase, urokinase, and tissue plasminogen activa- chronic forms of DVT, and success rates were considerably
tor (t-PA) are the primary thrombolytic agents used in the higher in the population with acute DVT. Only 6 patients in
treatment of DVT. These agents dissolve thrombi by cata- the study developed PE, and, of those cases, 2 resulted in
Blum and Roche Endovascular Management of Acute Deep Vein Thrombosis 33S

Table 2 Contraindications for thrombolytic therapy Contraindications for thrombolytic therapy

and risk of pulmonary embolism
Absolute Contraindications Relative Contraindications
● Active or recent internal ● Recent major surgery/ Primary contraindications for thrombolytic therapy include
bleeding organ biopsy concomitant medical conditions that increase the risk of
● Recent cerebrovascular ● CPR bleeding complications, such as recent surgery, stroke, or
accident ● Trauma gastrointestinal bleeding (Table 2). In addition to hemor-
● Known intracranial ● Pregnancy rhage, potential complications of catheter-directed throm-
neoplasm ● Diabetic hemorrhagic bolysis include migration of the thrombus and subsequent
● Recent craniotomy or retinopathy PE. In published thrombolytic series, these complications
spinal surgery ● Uncontrolled
are rare, and routine prophylactic IVC filter placement has
● Bacterial endocarditis
not been necessary during these infusions. The recent intro-
● Hepatic failure duction of temporary retrievable filters, however, may lower
● Renal failure the threshold for placement, as they may be removed fol-
lowing clot dissolution.
CPR ⫽ cardiopulmonary resuscitation.
In general, patients who might benefit from IVC filters
include those for whom anticoagulation therapy is contra-
indicated, has resulted in complications, or has failed to
prevent embolic events. This includes patients at high risk
death. Major bleeding complications were primarily related
for bleeding, e.g., those with extensive trauma or malig-
to catheter-insertion site bleeding, and were noted in 11% of
nancy and those undergoing certain surgical procedures.4
Under certain circumstances, IVC filter placement and an-
In a subsequent study involving a subset of this registry’s ticoagulation may be used in conjunction. This approach is
patients, Comerota and coworkers19 demonstrated better often used for patients with severe cardiopulmonary disease
physical functioning and health-related quality of life in whom a recurrent embolic event may otherwise prove to
(HRQOL) in patients with iliofemoral DVT treated by cath- be fatal.
eter-directed thrombolysis than in those patients treated
with anticoagulation alone. Patients successfully treated
with urokinase reported better overall physical functioning
(P ⫽ 0.046), less health distress (P ⫽ 0.022), and fewer May-Thurner syndrome
incidents of PTS (P ⫽ 0.006), compared with patients
Some patients develop DVT for anatomic rather than phys-
treated with heparin alone. Patients in the urokinase group
iologic reasons. May-Thurner syndrome, most commonly
who failed to achieve adequate lysis had HRQOL scores
found in young women with DVT, is a condition in which
similar to those of patients treated with heparin.
the left common iliac vein is narrowed by extrinsic com-
Data from the National Venous Thrombosis Registry pression by the overlying right common iliac artery. Intralu-
have established the optimal catheter-directed treatment ap- minal venous webs may develop as a result. Before the more
proach and patient population.18 An antegrade catheter- widespread use of imaging studies, this anomaly often went
directed approach using urokinase in patients with acute undetected. Venography following thrombolytic infusion
iliofemoral DVT duration of ⬍10 days and no prior history may identify such a culprit lesion, allowing for subsequent
of DVT leads to the best result, with complete lysis in 65% endovascular stenting or surgical repair (Figure 1).20
of those patients.
Analysis of long-term clinical outcome is pending, but
early results suggest improved valve function and fewer
symptoms at 1 year in patients with complete thrombolysis. Phlegmasia cerulea dolens
These promising data should serve as the basis for future
Phlegmasia cerulea dolens (PCD) refers to an ischemic
randomized trials of catheter-directed thrombolysis for the condition caused by massive venous thrombosis that often
treatment of acute DVT. involves the IVC and extends to the collateral veins and
Recently, the development of percutaneous mechanical small postcapillary venules. The condition is characterized
thrombectomy devices has added to the interventional ar- by massive edema with cyanosis due to arterial stasis and
mamentarium in the management of DVT. These devices compromise, and it may lead rapidly to acute compartment
are a natural extension of traditional open surgical tech- syndrome and venous gangrene (Figure 2). There are mul-
niques. A variety of low-profile mechanical devices are tiple triggering factors, including May-Thurner syndrome,
currently in investigational use in this setting. Early results malignancy, and surgery.
with these devices are promising, with dramatically reduced The initial treatment for mild PCD should be conserva-
infusion times and bleeding complications, and may obviate tive measures, including steep limb elevation, anticoagula-
the need for thrombolysis in some circumstances. tion with IV heparin, and fluid resuscitation.21 Strong con-
34S The American Journal of Medicine, Vol 118 (8A), August 2005

Figure 1 May-Thurner syndrome (iliac vein compression syn-


sideration should be given to thrombolytic therapy in more

advanced cases, with surgical thrombectomy an option if
there is no immediate response. Thrombectomy alone can-
not open the small venules affected in venous gangrene, and
some investigators have used thrombolysis via an intra-
arterial infusion as an alternative that delivers the thrombo- Figure 2 Phlegmasia cerulea dolens.
lytic agent to the arterial capillaries and subsequently to the
venules.13,21 Data from the National Venous Thrombosis
Registry has shown that this procedure restored venous the occluded IVC filter (Figure 5). A thrombolytic infusion
outflow and arrested tissue ischemia in a number of pa- of urokinase was administered via these catheters for a
tients.22 period of 18 hours.

Case study

History and presentation

A 75-year-old woman with a confirmed history of DVT and

several previous recurrences presented with severe bilateral
lower extremity edema and impending PCD or venous gan-
grene 6 months after IVC filter placement. Venography
revealed extensive thrombus within bilateral femoral veins
(Figure 3). There was a filling defect and no flow within the
IVC filter, which is consistent with caval thrombosis and
occlusion at that level (Figure 4).


Infusion catheters were advanced from both popliteal veins

across the thrombosed femoral segments and extended into Figure 3 Thrombosed femoral vein (right side shown).
Blum and Roche Endovascular Management of Acute Deep Vein Thrombosis 35S

Figure 4 Thrombotic occlusion of inferior vena cava at level of

inferior vena cava filter.

Clinical response

Follow-up venography after thrombolysis demonstrated

clearance of the thrombus and restoration of patency and
flow within the filter, IVC, and femoral veins (Figure 6).
There was rapid resolution of clinical symptoms.

Figure 5 Placement of infusion catheters for thrombolysis.

Anticoagulation remains the standard treatment regimen in more active patient, the risk of PTS and its complications
patients with acute DVT. With the advent of LMWH, out- remains high with anticoagulation alone. In these situations,
patient treatment is initially more cost-effective. However, anticoagulation and thrombolysis are complementary. Cath-
in more severe cases with extensive clot, or in the younger, eter-directed thrombolysis, aided by adjunctive endovascu-

Figure 6 (A) Patency of inferior vena cava postthrombolysis. (B and C) Patency of femoral vein postthrombolysis.
36S The American Journal of Medicine, Vol 118 (8A), August 2005

lar techniques, followed by outpatient management with venous thrombosis: an interim report of a prospective trial. Br J Surg.
anticoagulation is highly recommended in such instances. 1979;66:838 – 843.
9. Arnesen H, Høiseth A, Ly B. Streptokinase or heparin in the treatment
Placement of temporary or permanent IVC filters before
of deep vein thrombosis: follow-up results of a prospective study. Acta
catheter-directed thrombolysis may benefit patients at high Med Scand. 1982;211:65– 68.
risk for PE and is a stand-alone option for those with 10. Merli G. Anticoagulants in the treatment of deep vein thrombosis.
complications from or contraindications for anticoagulation Am J Med. 2005;118(Suppl 8A)13S–20S.
therapy and thrombolysis. Surgical thrombectomy is re- 11. Meissner MH, Manzo RA, Bergelin RO, et al. Deep venous insuffi-
served for patients who have failed therapy with these ciency: the relationship between lysis and subsequent reflux. J Vasc
less-invasive techniques. With continuing development and Surg. 1993;18:596 – 608.
12. Killewich LA, Martin R, Cramer M, et al. An objective assessment of
a growing body of experience with these new and/or refined
the physiological changes in the postthrombotic syndrome. Arch Surg.
methods to treat DVT, the goals of successful management 1985;120:424 – 426.
are closer to being met. 13. Comerota A, Aldridge S. Thrombolytic therapy for acute deep venous
thrombosis. Semin Vasc Surg. 1992;5:76 – 81.
14. Elsharawy M, Elzayat E. Early results of thrombolysis vs. anticoagu-
lation in iliofemoral venous thrombosis: a randomized clinical trial.
References Eur J Vasc Endovasc Surg. 2002;24:209 –214.
15. Meyerovitz MF, Polak JF, Goldhaber SZ. Short-term response to
1. White RH. The epidemiology of venous thromboembolism. Circula- thrombolytic therapy in deep venous thrombosis: predictive value of
tion [review]. 2003;107:I4 –I8. venographic appearance. Radiology. 1992;184:345–348.
2. Meissner MH. Thrombolytic therapy for acute deep vein thrombosis 16. Schweizer J, Kirch W, Koch R, et al. Short- and long-term results after
and the venous registry. Rev Cardiovasc Med. 2002;3(suppl 2):S53– thrombolytic treatment of deep venous thrombosis. J Am Coll Cardiol.
S60. 2000;36:1336 –1343.
3. Bergqvist D, Jendteg S, Johansen L, et al. Cost of long-term compli- 17. Semba CP, Dake MD. Iliofemoral deep venous thrombosis: aggressive
cations of deep venous thrombosis of the lower extremities: an analysis therapy with catheter-directed thrombolysis. Radiology. 1994;191:487–
of a defined patient population in Sweden. Ann Intern Med. 1997;126: 494.
454 – 457. 18. Mewissen MW, Seabrook GR, Meissner MH, et al. Catheter-directed
4. Hyers TM, Agnelli G, Hull RD, et al. Antithrombotic therapy for
thrombolysis of lower extremity deep venous thrombosis: report of a
venous thromboembolic disease. Chest. 2001;119:176S–193S.
national multicenter registry. Radiology. 1999;211:39 – 49.
5. Dolovich LR, Ginsberg JS, Douketis JD, et al. A meta-analysis com-
19. Comerota AJ, Throm RC, Mathias SD, et al. Catheter-directed throm-
paring low-molecular-weight heparins with unfractionated heparin in
bolysis for iliofemoral deep venous thrombosis improves health-re-
the treatment of venous thromboembolism. Arch Intern Med. 2000;
160:181–188. lated quality of life. J Vasc Surg. 2000;32:130 –137.
6. Levine M, Gent, M, Hirsh J, et al. A comparison of low-molecular- 20. O’Sullivan GJ, Semba CP, Bittner CA, et al. Endovascular manage-
weight heparin administered primarily at home with unfractionated ment of iliac vein compression (May-Thurner) syndrome. J Vasc
heparin administered in the hospital for proximal deep-vein thrombo- Intern Radiol. 2000;11:823– 836.
sis. N Engl J Med. 1996;334:677– 681. 21. Liem T, Rahhal D. Phlegmasia alba and cerulean dolens. Emedicine
7. Ziegler S, Schillinger M, Maca TH, Minar E. Postthrombotic syn- [serial online]. October 24, 2003. Available at: http://www.
drome after primary event of deep venous thrombosis 10 to 20 years Accessed May 18, 2005.
ago. Thromb Res. 2001;101:23–33. 22. Meissner MH. Thrombolytic therapy for acute deep vein thrombosis
8. Elliot MS, Immelman EJ, Jeffery P, et al. A comparative randomized and the venous registry. Rev Cardiovasc Med. 2002;3(suppl 2):S53–
trial of heparin versus streptokinase in the treatment of acute proximal S60.