The major complications of iliofemoral deep venous thrombosis (IFDVT) are pulmonary embolism and post-thrombotic syndrome. Although anticoagulation may prevent thrombus propagation, the dissolution of the clot is dependent on the body's own intrinsic fibrinolytic system. Surgical thrombectomy has had limited success rates, with early recurrent thrombosis. The systemic infusion of a lytic drug is fraught with a threefold increase in major bleeding complication rate. Acute IFDVT (<10 days) responds well to catheter-directed thrombolytic therapy. The lytic drug is delivered directly into the clot burden, thereby providing a higher thrombolytic efficiency, reduction in the overall dose of the lytic drug needed, decrease in infusion time, and rapid clearance.
Dr. Eclavea is a clinical research fellow in the Section of
Vascular & Interventional Radiology at the Indiana
University School of Medicine, Indianapolis, IN.
Dr. Patel is an associate professor in the Section of
Vascular & Interventional Radiology at the Indiana
University School of Medicine, Indianapolis, IN.
Hospital- and population-based studies, using objective
confirmation, report annual incidence ranging from 48 to 182 new
cases of acute deep vein thrombosis per 100,000 Western
Clinical features of acute iliofemoral deep venous thrombosis
(IFDVT) include massive unilateral leg edema, pain, and cyanosis.
As initially described by Virchow,
three factors are of primary importance in the development of deep
venous thrombosis (DVT): (a) abnormalities of blood flow, (b)
abnormalities of blood, and (c) vessel wall injury. Several of the
risk factors associated with DVT can be related to one or more of
these elements (Table 1). However, it is now recognized that most
thromboses arise from the interaction of multiple inherited and
acquired risk factors, giving rise to the concept of "thrombotic
potential" as proposed by Rosendaal et al.
When the cumulative thrombotic potential of all risk factors
exceeds a certain threshold, clinical thrombosis is likely.
Sequelae and natural history of acute deep venous
The short-term sequela of DVT is propagation of the thrombus to
a more clinically significant portion of the venous system and
embolization (figure 1) of thrombus to the pulmonary tree.
With anticoagulant therapy, clot propagation is deterred and the
intrinsic fibrinolytic system begins to clear the clot. This
recanalization process results in a series of pathological events.
Thrombi that are not dissolved by intrinsic fibrinolysis become
organized--connective tissue cells invade the thrombus, incorporate
it into the vein wall, and form scar tissue that may damage or
destroy vein valves. Alternatively, a new channel may develop
through the thrombus, recanalizing the vein. But the new channel is
irregular and does not permit free blood flow. This impedance often
results in recurrent deep venous thrombosis.
The major long-term sequela of DVT is post-thrombotic syndrome
This is manifested as chronic edema, pain, and muscle fatigue
(venous claudication), varicosities, hyperpigmentation of the skin,
subcutaneous fibrosis, and venous stasis skin ulcers (figure 2).
PTS becomes apparent only months to years after an episode of DVT.
Patients with IFDVT are at a high risk; 66% develop PTS, and 10% to
15% develop venous stasis skin ulcers.
PTS is the result of chronic elevation of venous pressures (venous
hypertension) in the lower limb due to venous outflow obstruction
and/or valvular incompetence. The single most important implicated
cause of PTS is valvular insufficiency in the popliteal and
superficial femoral veins.
Anticoagulant therapy and its shortcomings
The standard treatment of DVT is anticoagulation therapy with
supportive care (bed rest, leg elevation, and venous compression
stockings). Traditionally, the patient with DVT is admitted to the
hospital for intravenous unfractionated heparin while converting to
long-term oral anticoagulation using warfarin. With the
introduction of low-molecular weight heparin (LMWH) compounds
(i.e., enoxaparin, 1 mg/kg, bid sq; Lovenox, Rhone-Poulenc,
Collegeville, PA), outpatient anticoagulation therapy can be
administered subcutaneously. This method offers substantial
cost-savings since no hospitalization or laboratory monitoring of
the partial thromboplastin time (PTT) is required.
A common misconception among physicians is the notion that
systemic anticoagulation dissolves thrombus. By forming a complex
with antithrombin III, heparin inhibits the formation of thrombin
and thus suppresses thrombus formation and propagation. Heparin
does not act on the formed thrombus. With anticoagulation therapy,
recanalization of the occluded vessel depends solely on the
effectiveness of the patient's own fibrinolytic system.
Unfortunately, the body's natural fibrinolytic system cannot
completely remove the large clot burden of extensive IFDVT.
Subsequently, the thrombus organizes leading to permanent
obstruction and occlusion of the vein with development of valvular
The clinical consequence is the onset of PTS.
Krupski et al
found that only 6% of patients with acute extensive lower-extremity
DVT showed complete lysis of the thrombus within 10 days when
treated with heparin alone. Propagation of the thrombus occurs in
38% of these patients despite having therapeutic levels of heparin.
Johnson et al
found 59% of patients to be asymptomatic at a mean follow-up of 3
years after an episode of deep vein thrombosis. Valve function and
venous patency was shown by Doppler ultrasonography to have
normalized in only 18% of the symptomatic limbs and 3% of the
asymptomatic limbs. The frequency of reflux and obstruction in the
asymptomatic limbs was similar to that in symptomatic limbs.
However, when residual obstruction accompanied reflux, PTS was more
likely to develop. In a 5-year follow-up study, 95% of symptomatic
patients with IFDVT treated with anticoagulation alone had muscle
pump function and severe compromise of valvular competency.
Furthermore, 50% developed venous claudication and job disability,
and 86% developed venous stasis skin ulcers. The recurrence rate of
deep vein thrombosis is 7% to 30% in patients successfully treated
with anticoagulant therapy.
Although heparin is of benefit in reducing the incidence of
pulmonary embolism (PE), it appears to be of little value in
reducing the incidence of PTS.
Surgical thrombectomy and systemic thrombolytic
In contrast to anticoagulant therapy, surgical thrombectomy
(figure 3) and thrombolytic therapy directly remove the clot
burden. Unfortunately, surgical thrombectomy is hindered by early
rethrombosis rates (up to 72%) and plagued by recurrent DVT.
This, in part, is related to incomplete removal of the thrombus
burden and venous endothelial damage incurred by the Fogarty
The primary goal for thrombolytic therapy includes eliminating
the thrombus in order to restore venous blood return while
preserving valvular function.
reviewed the results of a multi-institutional study comparing the
acute benefits of systemic heparin therapy with systemic
thrombolytic therapy using streptokinase (SK). He found that 81% of
the patients in the heparin group had persistent venographic
occlusion at 1 week compared with 33% in the SK group (33%);
partial lysis was 13% in the heparin group compared with 20% in the
SK group; and complete lysis was 6% in the heparin group compared
with 47% in the lysis group. Overall, 67% of the patients treated
with systemic thrombolysis had partial or complete restoration of
patency, compared with only 19% in the heparin group. Two
prospective studies comparing heparin and SK have assessed the
long-term benefit following thrombolysis. Arneson et al
found the incidence of PTS (at 6.5 years) was 67% in the
heparin-treated patient group compared with 23% in the SK-treated
patient group. Furthermore, the symptoms were severe (with
ulceration) in 17% of the heparin-treated patient group and 0% of
the SK-treated patient group. Elliot et al
showed that 90% of the patients were symptomatic in the
heparin-treated group with 24% having severe symptoms, whereas 45%
of the patients in the SK-treated group were symptomatic but only
9% had severe symptoms. Comerota and Aldridge
performed a meta-analysis of 13 major studies comparing systemic
anticoagulant therapy with systemic thrombolytic therapy; 254
patients treated with heparin therapy showed complete clot lysis in
4% and partial clot lysis in 14%, whereas, 337 patients treated
with systemic thrombolytic therapy showed complete clot lysis in
45% and partial clot lysis in 18%.
Despite these encouraging results, the threefold increased risk
in major bleeding complication rates,
recently reported at 5% (12 of 250) compared with none using
has limited the widespread acceptance of systemic thrombolytic
therapy for DVT. Also, systemic thrombolytic therapy met with only
limited success in patients with extensive IFDVT because only the
exposed surfaces of the clot burden interacted with the lytic drug.
Acute thrombus is more responsive to lysis with systemic infusion
of a lytic drug than is chronic thrombus. Theiss et al
found acute occlusions <3 days old responded to lysis 94% of the
time, while occlusions 1 to 2 weeks old responded 80% of the time.
Occlusions 5 to 8 weeks old had only a 14% response rate to
Catheter-directed thrombolytic therapy
In 1994, Semba and Dake
provided the initial insight on the potential role of
catheter-directed thrombolysis (CDT) using urokinase (UK) in the
treatment of IFDVT. They successfully treated 25 of 27 limbs in 21
consecutive patients with CDT. Complete lysis was achieved in 18
patients (72%), partial lysis was achieved in 5 patients (20%), and
lysis was unsuccessful in 2 (8%) patients. There were no major
complications or clinically detectable PE. The technical and
clinical success rates were 85%. At 3 months, 12 limbs were studied
with Doppler ultrasonography, and veins were patent in 92%. One
patient with a chronic occlusion experienced a recurrent
Comerota et al
also advocated that the initial treatment of a patient with acute
IFDVT should be CDT. This minimally invasive endovascular technique
entails embedding a multi-sidehole infusion catheter directly into
the thrombus burden for local, intrathrombotic delivery of the
lytic drug. The benefits of CDT over systemic thrombolytic therapy
are: (a) rapid clearance of the thrombus burden; (b) higher doses
of the lytic drug introduced directly into the clot burden, thereby
increasing the lytic efficiency; (c) reduction in overall dose and
duration of lytic infusion; and (d) decrease in the risk of
producing a systemic fibrinolytic state, thereby reducing the major
bleeding complication rate.
These encouraging reports resulted in the establishment of the
North American Venous Registry to determine the impact of CDT using
UK in the treatment of symptomatic lower-extremity DVTs.
Between January 1995 and December 1996, 473 patients with
symptomatic lower-limb DVT were entered into the registry
prospectively. CDT was used for 306 infusions, most commonly via an
ipsilateral popliteal venous approach. From conventional
venography, clot lysis was graded as follows: Grade I (<50%
lysis), Grade II (>50% lysis) or Grade III (complete lysis).
Based on follow-up duplex ultrasound examinations, 1-year
cumulative primary patencies were calculated for different
subgroups and compared using life-table methods. With CDT,
significant lysis (>50%) was found in 83%; Grade III lysis in
31%; Grade II in 52%; and Grade I in 17%. In acute DVT limbs, Grade
II lysis occurred in 34% and Grade III lysis in 60%, which was
significantly greater compared with that achieved in the chronic
DVT limbs. The lysis grade obtained after lysis and stenting was a
major predictor of continued patency. For Grade III, Grade II, and
Grade I lytic outcomes, the 1-year patencies were 79%, 58%, and
32%, respectively (
<0.001). At 6 months, approximately 70% of limbs with
significant lysis (>50%) remained patent; moreover, 90% of limbs
with complete lysis remained patent. Reflux at follow-up at 6
months was found in <30% of those with complete lysis, 45% of
those with >50% lysis, and >60% with <50% lysis.
Based on health-related quality of life (HRQOL) evaluation,
Comerota et al
recently reported on the benefits of CDT in patients entered in the
venous registry with IFDVT. After thrombolytic therapy, patients
reported better overall physical functioning, less stigma, less
health distress, and fewer post-thrombotic symptoms compared with
similar patients treated with anticoagulation alone. Successful
lysis was directly correlated with improved HRQOL, with patients
who were classified as thrombolytic failures having similar
outcomes to patients treated with heparin alone.
Thrombolytic therapy utilizes a group of medications (classified
as plasminogen activators) that enzymatically cleaves circulating
serum protein plasminogen into the active protease plasmin. Plasmin
functions to disrupt the proteinacious fibrin cross-linked strands
that provide the framework for platelet and red blood cell
aggregation in a thrombus.
Urokinase (UK) (Abbokinase, Abbott Laboratories, Abbott Park,
IL) acts directly on plasminogen and is derived from neonatal renal
cell tissue culture. UK has a relatively lower fibrin
affinity/specificity and a longer half-life (14 min) with the
reputation of being consistent, predictable, efficacious, and safe.
UK has been the lytic drug of choice over streptokinase and tissue
plasminogen activator for CDT for IFDVT when fibrin specificity,
immunogenicity, major bleeding complication rate, and cost are all
A common criticism of CDT using UK for IFDVT has been cost, since
it often costs hospitals several thousand dollars for the lytic
drug alone and generates large charges to the third party payers.
UK was the most widely used lytic drug until the U.S. Food and Drug
Administration (FDA) banned the distribution in late 1998 due to
the theoretical risk of transmitting viral disease.
Streptokinase (Streptase, AstraZeneca Laboratories, Westboro,
MA) is a purified derivative from Group C beta-hemolytic
streptococci and indirectly activates plasminogen conversion.
Anistreplase (Eminase, Roberts Pharmaceuticals, Eatontown, NJ) is
the anisoylated stabilized intermediate complex of streptokinase
and plasminogen. These complexes are inactivated by circulating
antibodies and can be typically used only once in the same patient
due to the risk of drug-related fever and antibody reaction (serum
sickness) when administered repeatedly. Both are considered
first-generation lytic drugs and are not regarded as the optimal
replacement for UK due to their antigenicity and inferior efficacy.
Alteplase (Activase, Genentech, Inc., South San Francisco, CA)
is a recombinant engineered human tissue plasminogen activator
(t-PA) analog, which has high fibrin affinity/specificity and a
short circulating half-life (5 min). Tissue plasminogen activator
is a naturally occurring serine protease, which is secreted by
vascular endothelial cells and plays a major role in maintaining
the normal homeostasis of the blood vessel. The gene sequence that
expresses t-PA is isolated and removed from the human melanoma cell
line and introduced into the ovarian cell of the Chinese hamster,
where sufficient quantities of the protein are produced for
commercial use. Only a few trials exist using rt-PA as a systemic
lytic drug for the treatment of DVT; however, experience with CDT
using rt-PA is emerging.
While clinical studies have shown efficacy of rt-PA, there remains
generalized concern about the risk of adverse bleeding using rt-PA
compared with UK, especially in lieu of early trials in which
extraordinarily high doses of rt-PA (>2 mg/hr with full
heparinization) were utilized for prolonged overnight infusions.
Because of these concerns, the Advisory Panel on Catheter-Directed
Thrombolytic Therapy of the Society of Cardiovascular and
Interventional Radiology recently published initial practice
guidelines regarding the dosing of rt-PA and recommends doses far
below the initial published experiences of the late 1980s.
Reteplase (Retavase, Centocor/ Johnson & Johnson, Malvern,
PA) is a new recombinant derivative of tissue plasminogen
activator. It is a single-chain protein consisting of 355 of the
527 amino acid sequence in rt-PA. However, by genetically removing
the finger, growth factor, and 1 kringle domain, reteplase (R-PA)
has less fibrin affinity and increased half-life (15 min) compared
with alteplase (rt-PA). The decreased fibrin affinity of R-PA has
been perceived as allowing the lytic drug to seep through the clot
more easily than rt-PA, but large-scale trials in acute myocardial
infarction have shown rt-PA and R-PA to be equivalent in terms of
safety, efficacy, and long-term outcomes.
Early clinical experience using R-PA as an alternative to UK is
also emerging, and it is likely that its performance will be safe
and efficacious for CDT.
Tenecteplase (TNKase, Genentech, Inc.), the newest recombinant
lytic drug, represents a triple-mutated version of the parent
molecule rt-PA. Compared to rt-PA, TNK-tPA possesses enhanced
fibrin specificity, a longer half-life, and increased resistance to
plasminogen activator inhibitor-1. This lytic drug may have
potential impact on thrombolytic therapy in the near future.
Protocol for CDT for acute (extensive) IFDVT
IFDVT should be documented by conventional venography and/or
duplex Doppler ultrasonography (figure 4), although magnetic
resonance imaging and computed tomography are also useful
modalities. An inferior vena cava filter should be reserved for
those patients with recurrent PE despite anticoagulation or with
"free-floating" thrombus in the iliac veins or inferior vena
Baseline coagulation parameters (platelet count, prothrombin
time, partial thromboplastin time, serum fibrinogen, hemoglobin,
and hematocrit) are obtained, a peripheral arm intravenous line is
started, and a Foley urinary bladder catheter is inserted. The
ipsilateral popliteal vein is accessed antegrade with a 4F
micropuncture set (Cook, Inc., Bloomington, IN) under ultrasound
guidance, so the risk of valvular damage is minimized. If there is
extensive calf vein and popliteal thrombus, then the posterior
tibial or lesser saphenous vein can be used for access.
A complete ascending venogram is performed (figure 5). A vascular
sheath (5F or 6F, depending on the infusion system to be used) is
placed. A hydrophilic catheter and hydrophilic guidewire are
advanced under fluoroscopy into the inferior vena cava (IVC) and a
venocavagram is performed. A multi-sidehole (i.e., Cragg-McNamara
infusion catheter, Micro Therapeutics, Inc., Irvine, CA) or
multi-slit (i.e., Unistep infusion catheter, Angiodynamics, Inc.,
Queensbury, NY) infusion catheter with an infusion length that
spans the entire thombus burden is placed. Either 0.5 to 1.5 mg/hr
of alteplase (rt-PA) or 0.5 to 1.0 mg/hr of reteplase (R-PA) is
infused with the dose split between the vascular sheath and the
infusion catheter. Heparin should not be mixed with these lytic
drugs because of the possibility of precipitation. Subtherapeutic
heparin (2500 U bolus, 500 U/hr maintenance) may be administered
through the peripheral arm intravenous line. Serum fibrinogen level
and partial thromboplastin time are obtained every 6 to 8 hours and
are maintained >100 mg/dL and <60 sec, respectively. However,
it is important to note that serum fibrinogen values may not be
predictive of adverse bleeding.
Several scenarios may be seen at the initial follow-up
venography (Table 2), usually performed 6 to 12 hours after
initiating thrombolysis. After complete or near-complete lysis of
the deep vein thrombosis, a residual obstructive lesion is found in
the common iliac vein in the majority of the cases, and synechiae
may be found in the superficial femoral vein. The synechiae may be
disrupted with an angioplasty balloon or mechanical device (figure
6). However, the iliac vein lesion rarely responds to angioplasty
alone, and placement of a self-expanding endovascular stent is
generally required (figure 7).
Ideally, a stent diameter oversized by 20% deployed flush with
the iliocaval junction is an easily attainable goal. The stent is
then expanded using the appropriately sized angioplasty balloon.
Stenting below the inguinal ligament should typically be avoided,
although Semba and Dake
have not found any untoward sequelae as long as the stent does not
extend below the lesser trochanter of the femur.
The overall role of adjunctive mechanical thrombectomy devices
is not well defined. Most interventionalists use a maximum duration
of therapy with rt-PA of 48 hours but hope to stop earlier since
bleeding complications increase over time.
The patients should be anticoagulated (INR between 2 and 3) for 6
months following successful lysis. Many authors advocate giving
antiplatelet therapy (75 mg Plavix, Bristol-Myers Squibb/Sanofi
Pharmaceuticals, New York, NY) for 6 weeks in patients in whom a
stent was placed.
The patient should be followed clinically with serial duplex
Doppler sonographic imaging performed on the morning after
completion of thrombolytic therapy, and then at 6 and 12 months,
and annually thereafter.
Although the factors leading to venous thrombosis have been
known for over a century, Virchow's initial model of thrombosis has
been refined extensively. Activated coagulation is now recognized
to be of primary importance in venous thrombogenesis; the concept
of venous injury has been expanded to include molecular changes in
the endothelium; and stasis has been redefined as a largely
permissive factor. Furthermore, it is now clear that venous thrombi
undergo a dynamic evolution beginning soon after their formation.
The natural history of acute DVT is a balance between recurrent
thrombotic events and processes that restore the venous lumen, both
of which have important implications for the development of
Although pulmonary embolism is clearly the most life-threatening
complication of acute DVT, the long-term socioeconomic consequences
of PTS have perhaps been underemphasized in clinical trials. The
development of post-thrombotic manifestations is related to both
residual venous obstruction and valvular incompetence. Recognition
of the factors that contribute to a poor outcome, including
recurrent thrombotic events, the rate of recanalization, the global
extent of venous reflux, and the anatomic distribution of reflux
and obstruction, is important, as there may be therapeutic
alternatives to alter the natural history of acute DVT. CDT for
acute deep vein thrombosis could provide rapid thrombus clearance
with improved lytic efficiency while preserving valvular function
and decreasing bleeding complications. The treatment alternatives
will continue to expand with the introduction of new therapeutic
drugs and mechanical thrombectomy devices. *