Deep Venous Thrombosis (DVT) Treatment & Management

Updated: Jun 05, 2019
  • Author: Kaushal (Kevin) Patel, MD; Chief Editor: Barry E Brenner, MD, PhD, FACEP  more...
  • Print

Approach Considerations

The primary objectives for the treatment of deep venous thrombosis (DVT) are to prevent pulmonary embolism (PE), reduce morbidity, and prevent or minimize the risk of developing the postthrombotic syndrome (PTS).

The mainstay of medical therapy has been anticoagulation since the introduction of heparin in the 1930s. [112] Other anticoagulation drugs have subsequently been added to the treatment armamentarium over the years, such as vitamin K antagonists and low-molecular-weight heparin (LMWH). More recently, mechanical thrombolysis has become increasingly used as endovascular therapies have increased. Absolute contraindications to anticoagulation treatment include intracranial bleeding, severe active bleeding, recent brain, eye, or spinal cord surgery, pregnancy, and malignant hypertension. Relative contraindications include recent major surgery, recent cerebrovascular accident, and severe thrombocytopenia.

The immediate symptoms of DVT often resolve with anticoagulation alone, and the rationale for intervention is often reduction of the 75% long-term risk of PTS. Systemic IV thrombolysis once improved the rate of thrombosed vein recanalization; however, it is no longer recommended because of an elevated incidence of bleeding complications, slightly increased risk of death, and insignificant improvement in PTS. The lack of a significantly reduced incidence of PTS after systemic thrombolysis (40-60%) likely reflects the inadequacy of the relatively low threshold volume of thrombus removal that was considered successful.

Thrombolytic therapy is recommended (systemic preferred over catheter directed) in hypotensive individuals with an acute PE. [113]  Those with high-risk PE presenting in shock should undergo systemic thrombolysis; when thrombolysis is contraindicated owing to a high risk of bleeding, consider surgical thrombectomy or catheter direct thrombolysis. [114]

The bleeding risk of systemic thrombolysis is similar to that of catheter-directed thrombolysis, and the risk of PTS may further decrease risk. However, whether catheter-directed thrombolysis is preferred to anticoagulation has not been examined. The addition of percutaneous mechanical thrombectomy to the interventional options may facilitate decision-making, because recanalization may be achieved faster than before and with a decreased dose of lytic; therefore, the bleeding risk may be decreased.

Inpatient Versus Outpatient Treatment

Acute DVT may be treated in an outpatient setting with LMWH. Patients with low-risk PE may be safely discharged early from hospital or receive only outpatient treatment with LMWH, followed by vitamin K antagonists, although nonvitamin K-dependent oral anticoagulants may be as effective but safer than the LMWH/vitamin K antagonist regimen. [115]

Anticoagulant therapy is recommended for 3-12 months depending on site of thrombosis and on the ongoing presence of risk factors. If DVT recurs, if a chronic hypercoagulability is identified, or if PE is life threatening, lifetime anticoagulation therapy may be recommended. This treatment protocol has a cumulative risk of bleeding complications of less than 12%.

Most patients with confirmed proximal vein DVT may be safely treated on an outpatient basis. Exclusion criteria for outpatient management are as follows:

  • Suspected or proven concomitant PE

  • Significant cardiovascular or pulmonary comorbidity

  • Iliofemoral DVT

  • Contraindications to anticoagulation

  • Familial or inherited disorder of coagulation: antithrombin III (ATIII) deficiency, prothrombin 20210A, protein C or protein S deficiency, or factor V Leiden

  • Familial bleeding disorder

  • Pregnancy

  • Morbid obesity (>150 kg)

  • Renal failure (creatinine >2 mg/dL)

  • Unavailable or unable to arrange close follow-up care

  • Unable to follow instructions

  • Homeless

  • No contact telephone

  • Geographic (too far from hospital)

  • Patient/family resistant to outpatient therapy

Admitted patients may be treated with a LMWH, fondaparinux, or unfractionated heparin (UFH). Warfarin 5 mg PO daily is initiated and overlapped for about 5 days until the international normalized ratio (INR) is therapeutic >2 for at least 24 hours.

For admitted patients treated with UFH, the activated partial thromboplastin time (aPTT) or heparin activity level must be monitored every 6 hours while the patient is taking intravenous (IV) heparin until the dose is stabilized in the therapeutic range. Patients treated with LMWH or fondaparinux do not require monitoring of the aPTT.

Platelets should be monitored. Heparin or LMWH should be discontinued if the platelet count falls below 75,000. Fondaparinux is not associated with hepatin-induced thrombocytopenia (HIT).


Consultations with the following specialists are indicated:

  • Hematologist

  • Vascular surgeon

  • Radiologist

  • Interventional radiologist


General Principles of Anticoagulation

Anticoagulant therapy remains the mainstay of medical therapy for deep venous thrombosis (DVT) because it is noninvasive, it treats most patients (approximately 90%) with no immediate demonstrable physical sequelae of DVT, it has a low risk of complications, and its outcome data demonstrate an improvement in morbidity and mortality. Long-term anticoagulation is necessary to prevent the high frequency of recurrent venous thrombosis or thromboembolic events. Anticoagulation does have problems. Although it inhibits propagation, it does not remove the thrombus, and a variable risk of clinically significant bleeding is observed.

First-line therapy for non-high risk venous thromboembolism (VTE) or pulmonary embolism (PE) consists of direct oral anticoagulants (dabigatran, rivaroxaban, apixaban, or edoxaban) over vitamin K antagonists (VKAs). [113, 114]  VKAs are also recommended over low-molecular-weight heparin (LMWH), unless VTE is associated with malignancy, in which case LMWH is preferred over VKAs or any direct oral anticoagulants. [113]  

When the risk of VTE recurrence is high in patients with subsegmental PE without DVT, the American College of Chest Physicians (ACCP) recommends anticoagulation over surveillance; when the VTE recurrence risk is low in these patients, surveillance over anticoagulation is suggested. [113]

Inferior vena cava filters are not recommended in patients with acute VTE on anticoagulant therapy. [113]

Barring contraindications to aspirin therapy, aspirin is recommended to prevent recurrent VTE in patients with an unprovoked proximal DVT or PE following anticoagulation cessation. [113]

Park and Byun indicate that possibilities for advances in anticoagulant delivery systems include expansion of new oral agents and their antidotes, reducing the size of heparins, developing oral or topical heparins, and modifying physical or chemical formulations. [116]  For example, Ita suggests that transdermal delivery may potentially bypass known issues with heparin use, such as short half-life and unpredictable bioavailability, and offer improved patient compliance, convenience, ease of dosing termination, as well as avoid the first-pass effect. [112]

For more information, see General Principles of Anticoagulation in Deep Venous Thrombosis.


Heparin Use in Deep Venous Thrombosis

Heparin products used in the treatment of deep venous thrombosis (DVT) include unfractionated heparin and low molecular weight heparin (LMWH) The efficacy and safety of low-molecular-weight heparin (LMWH) for the initial treatment of DVT have been well established in several trials. Traditionally, heparin has been used only for admitted patients with DVT. Regular unfractionated heparin was the standard of care until the introduction of LMWH products. Heparin prevents extension of the thrombus and has been shown to significantly reduce (but not eliminate) the incidence of fatal and nonfatal pulmonary embolism and recurrent thrombosis.

Heparin is a heterogeneous mixture of polysaccharide fragments with varying molecular weights but with similar biological activity. The larger fragments exert their anticoagulant effect by interacting with antithrombin III (ATIII) to inhibit thrombin. ATIII, the body’s primary anticoagulant, inactivates thrombin and inhibits the activity of activated factor X in the coagulation process. The low-molecular-weight fragments exert their anticoagulant effect by inhibiting the activity of activated factor X. The hemorrhagic complications attributed to heparin are thought to arise from the larger higher-molecular-weight fragments. LMWH is prepared by selectively treating unfractionated heparin to isolate the low-molecular-weight (< 9000 Da) fragments.

Patients with recurrent venous thromboembolism (VTE) while on treatment with a non-LMWH anticoagulant should be switched to LMWH therapy. [113] Those who suffer recurrent VTE while on LMWH therapy should receive an increased dose of LMWH. [113]

For more information, see Heparin Use in Deep Venous Thrombosis.


Factor Xa and Direct Thrombin Inhibitors


Fondaparinux, a direct selective inhibitor of factor Xa, overcomes many of the aforementioned disadvantages of low-molecular-weight heparins (LMWHs). Pharmacokinetic studies of fondaparinux reveal that only a single-daily subcutaneous dose is required. Furthermore, a single dose of 7.5 mg is effective over a wide range of patient weights between 50 and 100 kg. Daily doses of 5 mg or 10 mg are appropriate for patients who weigh less or more than that weight range. Heparin-induced thrombocytopenia (HIT) has not been reported. Therapeutic monitoring of laboratory parameters such as the prothrombin time or activated partial thromboplastin time (aPTT) is also not required. In some regions, the cost of therapy with fondaparinux is less than enoxaparin when it is being used to bridge therapy to a vitamin K antagonist (VKA).

The combination of two factor Xa inhibitors may be an effective treatment strategy for acute venous thromboembolism (VTE). [117]  In an observational study, 80 of 87 consecutive Japanese patients with VTE who received SC fondaparinux for 7-10 days and then were switched to oral rivaroxaban for 7-14 days had treatment success. Both D-dimer levels and quantitative ultrasound thrombosis (QUT) scores were improved with the use of fondaparinux, and further reductions were achieved using rivaroxaban. [117]

Buller and his coauthors on behalf of the Matisse Investigators conducted a randomized, double-blind, international study of fondaparinux versus enoxaparin on 2,205 patients with objectively confirmed acute deep venous thrombosis (DVT) and found the two agents to be comparable in safety and efficacy. [6] Patients were randomly assigned to receive fondaparinux or enoxaparin therapy. Fondaparinux was administered as a single 7.5-mg subcutaneous daily dose, with adjustments made for those patients weighing less than 50 kg (5 mg) or greater than 100 kg (10 mg). Enoxaparin was given 1 mg/kg subcutaneously twice daily. Both agents were bridged with a VKA until a therapeutic international normalized ratio (INR) was achieved. Anticoagulation with a VKA was continued for 3 months. Efficacy was measured by the rate of recurrent VTE in the 3-month follow-up period after enrollment. Safety was assessed by the incidence of major bleeding and mortality over the same interval. [6]

The recurrence rate showed a nonsignificant trend in favor of fondaparinux (3.9%) compared with enoxaparin (4.1%) (absolute difference = 0.15%; 95% CI, 1.8% to -1.5%). [6] The conservative noninferiority margin was attained, and fondaparinux was determined to be equally as effective as enoxaparin for the treatment of DVT. Major bleeding rates were essentially identical, and mortality rates were also comparable. In a subgroup analysis, the authors also evaluated the relationship between the recurrence rate, the bleeding risks, and the patients’ body weight. In general, the safety and efficacy of fondaparinux were independent of body weight. However, patients with mild renal insufficiency and a low creatinine clearance had the same risk of bleeding in both the LMWH and fondaparinux groups. Overall, the authors concluded that once-daily fondaparinux was as effective and as safe as twice-daily, weight-adjusted enoxaparin. [6]

The Matisse DVT trial confirmed that fondaparinux and enoxaparin have similar safety and efficacy for the initial treatment of DVT. Only one fixed-dosage regimen for fondaparinux is required for patients who weigh between 50 kg and 100 kg, and only one subcutaneous dose per day is required. This greatly simplifies the treatment of DVT and facilitates outpatient therapy. In the original study, about one third of the patients were treated partially or entirely as outpatients without any increased risk when compared with those treated as inpatients.

In renal insufficiency with a creatinine clearance less than 30 mL/min, major bleeding occurred in 2 of 25 patients (8%) on fondaparinux versus 1 of 18 patients (5.6%) treated with enoxaparin (= NS). Because of the small sample size and the higher risk of bleeding, fondaparinux is contraindicated in patients with renal insufficiency and a creatinine clearance less than 30 mL/min.

In the event of a major bleed, protamine sulfate partially reverses the anticoagulant effect of enoxaparin. However, no specific antidote to fondaparinux is available. A recent study revealed that a bolus dose of 90 mcg/kg of recombinant factor VIIa reversed the anticoagulant effect of fondaparinux, at least in healthy volunteers given a larger 10-mg dose. [118]


Rivaroxaban (Xarelto) is an oral factor Xa inhibitor approved by the FDA in November 2012 for treatment of DVT or pulmonary embolism (PE) and for reduction of the risk of recurrent DVT and PE after initial treatment. [7, 8, 9]  Approval for this indication was based on studies totaling 9478 patients with DVT or PE. Participants were randomly assigned to receive rivaroxaban, a combination of enoxaparin and a VKA (eg, warfarin), or a placebo. Study endpoints were designed to measure the number of patients who experienced recurrent symptoms of DVT, PE, or death after receiving treatment.

Data from a pooled analysis of the EINSTEIN-DVT [7] and EINSTEIN-PE [8] trials suggested that rivaroxaban is as effective in preventing VTE recurrence as enoxaparin followed by a VKA and may be associated with less bleeding [9] ; in addition, the data suggested that there are no grounds for avoiding rivaroxaban use in high-risk groups (eg, fragile patients, cancer patients, and patients with a large clot).

Approximately 2.1% of patients treated with rivaroxaban experienced recurrent DVT or PE, compared with 1.8-3% treated with the enoxaparin and VKA combination. [7, 8]

Additionally, results from extended treatment demonstrated a reduced risk of recurrent DVT and PE. Approximately 1.3% in the rivaroxaban group experienced recurrent DVT or PE, compared with 7.1% in the placebo group. [119, 120]


In March 2014, the FDA approved apixaban (Eliquis) for the additional indication of prophylaxis of DVT and PE in adults who have undergone hip- or knee-replacement surgery. Support for this new indication was a result of the ADVANCE 1, 2, and 3 clinical trials that enrolled nearly 12,000 patients. [121, 122, 123] Apixaban was originally approved by the FDA in December 2012 for the prevention of stroke and systemic embolism in patients with nonvalvular atrial fibrillation.

In August 2014, apixaban was approved for treatment of DVT and PE. [124] The approval for treatment of PE and prevention of recurrence was based on the outcome of the AMPLIFY (Apixaban for the Initial Management of Pulmonary Embolism and Deep-Vein Thrombosis as First-Line Therapy) and AMPLIFY-EXT (extended treatment) studies, in which apixaban therapy was compared with enoxaparin and warfarin treatment. The AMPLIFY study showed that, in comparison with the standard anticoagulant regimen, apixaban therapy resulted in a 16% reduction in the risk of a composite endpoint that included recurrent symptomatic venous thromboembolism (VTE) or VTE-associated death. [125, 126]

Data from the AMPLIFY-EXT trial showed that extended anticoagulation (12 months) with apixaban shortened hospital stays, reduced symptomatic recurrent venous thromboembolism or all-cause death without an associated increase in major episodes of hemorrhage when compared with placebo. [127]


Dabigatran (Pradaxa) inhibits free and clot-bound thrombin and thrombin-induced platelet aggregation. This agent was FDA approved in 2010 to reduce the risk of stroke and systemic embolism in patients with nonvalvular atrial fibrillation. In April 2014, it was approved for the treatment of DVT and PE in patients who have been treated with a parenteral anticoagulant for 5-10 days. Additionally, it was approved to reduce the risk of DVT and PE recurrence in patients who have been previously treated. Approval was based on results from 4 global phase III trials that showed dabigatran was noninferior to warfarin and had a lower risk of major or clinically relevant bleeding compared with warfarin. [128, 129, 130] There have been reports of severe and fatal bleeding in users of the drug.

The RE-COVER and RE-COVER II trials included patients with DVT and PE who were treated with parenteral anticoagulant therapy for 5-10 days. Results showed dabigatran was noninferior to warfarin in reducing DVT and PE after a median of 174 days of treatment with a lower risk of bleeding compared with warfarin. [128, 129]

The RE-SONATE trial and RE-MEDY trials included patients (n=2856) with acute DVT and PE who had completed at least 3 months of anticoagulant therapy. Results from this trial showed dabigatran was noninferior to warfarin in the extended treatment of VTE and carried a lower risk of major or clinically relevant bleeding than warfarin. [130]


Edoxaban (Savaysa) was approved by the FDA in January 2015 for the treatment of DVT and PE in patients who have been initially treated with a parenteral anticoagulant for 5-10 days. [131] Approval was based on the Hokusai-VTE study that included 4,921 patients with DVT and 3,319 patients with PE. [131, 132]

Among patients with PE, 938 had right ventricular dysfunction, as assessed by measurement of N-terminal pro-brain natriuretic peptide (NT-proBNP) levels. [132] There was a 3.3% rate of recurrent VTE in this subgroup in those who received edoxaban compared to 6.2% in the group that received warfarin. The investigators concluded that edoxaban was not only noninferior to high-quality standard warfarin therapy but also caused significantly less bleeding in a broad spectrum of patients with VTE, including those with severe PE. [132]


Betrixaban (Bevyxxa), a FXa inhibitor, was approved by the FDA in June 2017. [133] It is indicated for the prophylaxis of VTE in adults hospitalized for acute medical illness who are at risk for thromboembolic complications owing to moderate or severe restricted mobility and other risk factors that may cause VTE. [133]

Approval of betrixaban was based on data from the phase 3 APEX studies. [134, 135] These randomized, double-blind, multinational clinical trials compared extended-duration betrixaban (35-42 days) to short-duration enoxaparin (6-14 days) for VTE in 7,513 acutely medically ill hospitalized patients with VTE risk factors. [133, 134, 135] Patients in the betrixaban group received an initial dose of 160 mg orally on day 1, followed by 80 mg once daily for 35-42 days, and received a placebo injection once daily for 6-14 days. Patients in the enoxaparin group received 40 mg subcutaneously once daily for 6-14 days and took an oral placebo once daily for 35-42 days. [133, 134, 135]

Efficacy was measured in 7,441 patients using a composite outcome score composed of the occurrence of asymptomatic or symptomatic proximal DVT, nonfatal PE, stroke, or VTE-related death. [133, 134, 135]  Those who received betrixaban showed significant decreases in VTE events (4.4%) compared with patients in the enoxaparin group (6%).


Duration of Anticoagulation

For the first episode of deep venous thrombosis (DVT), patients should be treated for 3-6 months. Recurrent episodes should be treated for at least 1 year.

The American College of Chest Physicians (ACCP) recommends cessation of anticoagulant therapy after 3 months of treatment in those with (1) surgery-associated acute proximal DVT, (2) an acute proximal DVT or PE provoked by a nonsurgical transient risk factor, and (3) a first unprovoked VTE and a high risk of bleeding. [113] (In those with a low or moderate bleeding risk, extend anticoagulation without a scheduled stop date.) [113]

Prandoni et al found that the use of ultrasonography to determine the duration of anticoagulation can reduce recurrences of venous thromboembolism after a first episode of acute proximal DVT. In the study, 538 consecutive outpatients who had completed an uneventful 3-month period of anticoagulation were randomized to receive either fixed-duration anticoagulation (< 9 months for secondary DVT and up to 21 months for unprovoked thrombosis) or flexible-duration anticoagulation, with treatment discontinued once ultrasound showed recanalization of the affected veins. Recurrent venous thromboembolism developed in 17.2% of the patients allocated to fixed-duration anticoagulation and 11.9% of the patients allocated to flexible-duration anticoagulation; no significant difference was noted in the rate of major bleeding. [136]

Patients with cancer have a particularly higher rate of DVT recurrence than noncancer patients. Long-term therapy for DVT is strongly recommended. Studies have shown a lower rate of venous thromboembolism (VTE) recurrence without increasing the risk of bleeding with low-molecular-weight heparin (LMWH) therapy. Reports also describe that the LMWH compounds may decrease the all-cause mortality rate. The author recommends LMWH therapy alone without crossover to warfarin if the patient’s insurance covers it.

Indefinite therapy is recommended for patients with recurrent episodes of venous thrombosis regardless of the cause. The risk of recurrent thromboembolism during a 4-year follow-up period was reduced from 21% to 3% with continued anticoagulation. However, the incidence of major bleeding increased from 3% to 9%. [137]

Long-term therapy with LMWH has been shown to be as effective as warfarin in the treatment of venous thrombosis, except in those patients with a concurrent malignancy. In this subgroup, LMWH was shown to be more effective than oral therapy. [138, 139] Initial studies have also shown LMWH to be effective in pregnant patients, but long-term, large randomized trials have yet to be completed. [140]


Complications of Anticoagulant Therapy

Hemorrhagic complications are the most common adverse effects of anticoagulant therapy. Anticoagulation therapy for 3-6 months results in major bleeding complications in 3-10% of patients. [141] High-risk populations (>65 y with a history of stroke, gastrointestinal [GI] bleed, renal insufficiency, or diabetes) have a 5-23% risk of having major hemorrhage at 90 days. Patients who require yearlong or indefinite anticoagulation (because of chronic risk factors) have double the risk of hemorrhage. Significant bleeding (ie, hematemesis, hematuria, GI hemorrhage) should be thoroughly investigated because anticoagulant therapy may unmask a preexisting disease (eg, cancer, peptic ulcer disease, arteriovenous malformation).

The treatment of hemorrhage while taking heparin depends on the severity of the bleeding and the extent to which the activated partial thromboplastin time (aPTT) is elevated above the therapeutic range. Patients who hemorrhage while receiving heparin are best treated by discontinuing the drug. The half-life is relatively short, and the aPTT usually returns to the reference range within a few hours. Treatment with fresh frozen plasma or platelet infusions is ineffective. For severe hemorrhage, such as intracranial or massive gastrointestinal bleeding, heparin may be neutralized by protamine at a dose of 1 mg for every 100 units. Protamine should be administered at the same time that the infusion is stopped.

The treatment of major hemorrhage associated with low-molecular-weight heparin (LMWH) is similar to heparin. However, the half-life of these agents is longer (4-6 h). As with heparin, fresh frozen plasma or platelet transfusions are ineffective. Protamine may be used, but it only reverses 60% of the drug’s effects.

The risk of bleeding on warfarin is not linearly related to the elevation of the international normalized ratio (INR). The risk is conditioned by other factors, including poor follow-up, drug interactions, age, and preexisting disorders that predispose to bleeding.

Patients who hemorrhage while receiving oral warfarin are treated by withholding the drug and administering vitamin K. Severe life-threatening hemorrhage is managed with fresh frozen plasma in addition to vitamin K. Recombinant factor VIIa is another option especially for central nervous system hemorrhage.

Additional complications include the following:

  • Systemic embolism

  • Chronic venous insufficiency

  • Postthrombotic syndrome (ie, pain and edema in the affected limb without new clot formation)

  • Soft tissue ischemia associated with massive clot and very high venous pressures - phlegmasia cerulea dolens


Emerging Anticoagulant Agents

The qualities desired in the ideal anticoagulant are ease of administration, efficacy and safety (with minimal complications or adverse effects), rapid onset, a therapeutic half-life, and minimal or no monitoring. Predictable and reversible action, with few drug or dietary interactions, and cost also are important. Achieving all these criteria in a single agent has not yet been achieved. Each of the anticoagulant agents available today has generally been able to incorporate some, but not all, of these characteristics.

In patients with deep venous thrombosis (DVT), anticoagulation remains the cornerstone of treatment. The relatively recent development of novel oral anticoagulants has provided clinicians with an expanding set of options for DVT treatment. [142]

Current research in anticoagulants involves investigations into drugs that act on various phases of the coagulation cascade. Drugs under investigation that act in the initiation phase include tissue factor pathway inhibitors (TFPIs) and nematode anticoagulant peptide (NAPc2). Drugs that act on the third stage of the coagulation cascade, the thrombin activity phase, include the direct thrombin inhibitors. A partial listing of these emerging new anticoagulants includes razaxaban, idraparinux, bivalirudin, lepirudin, and ximelagatran.

For more information, see Emerging Anticoagulant Agents in Deep Venous Thrombosis.


Reversal of Anticoagulation

Anticoagulation-related major bleeding is associated with an increased risk of death and thrombotic events, independent of the class of anticoagulant used. Although older agents of anticoagulation and their reversal are well studied, the newer agents lack similar antidotes. With the increasing use of non–vitamin K antagonist oral anticoagulants (NOAC), the number of patients who require reversal of their anticoagulant effects can be expected to rise. The following section describes the reversal agents for both older and new anticoagulants.


Heparin has a relatively short half-life of about 60–90 minutes and, therefore, the anticoagulant effect of therapeutic doses of heparin will mostly be eliminated at 3-4 hours after termination of continuous intravenous administration.

For a more immediate neutralization of heparin, protamine sulfate can be administered at a dose of 1 mg for every 100 units of heparin. Protamine was originally isolated from fish sperm and binds to heparin to form a stable, biologically inactive complex. [143, 144]

Lower molecular weight heparins

Currently, there are no specific antidotes to low molecular weight heparins. Recombinant FVIIa (rVIIA) has been shown to stop bleeding in patients anticoagulated with fondaparinux; however, no randomized controlled trials on such patients have been conducted.


Vitamin K

In patients with clinically significant bleeding, vitamin K can be used to reverse the anticoagulant effect of vitamin K antagonists (VKA). Vitamin K can be given orally or intravenously. The parenteral route has a more rapid onset; however, it is associated with a slightly increased risk of allergic reaction.

Fresh frozen plasma (FFP)

In case of a life-threatening emergency, FFP can be used for the reversal of VKA. FFP contains all the coagulation factors in normal concentrations. However, FFP should be used with caution, as it has the potential to cause volume overload, allergic reaction, and transfusion-related reactions (eg, transfusion-related acute lung injury). [145]

Prothrombin complex concentrates (PCCs)

In the case of serious and life-threatening bleeding, immediate correction of the international normalized ratio (INR) can be achieved by the administration of PCCs. These contain 3 or 4 of the vitamin K–dependent coagulation factors, as well as proteins C and S. In a prospective study, administration of PCCs has been shown to result in sustained hemostasis in patients using VKA.

Non–vitamin K antagonist oral anticoagulants (NOACs)

The new oral anticoagulant factor Xa or IIa inhibitors have numerous advantages over traditional VKAs, including rapid therapeutic effectiveness, ease of dosing, and lack of monitoring. Until recently, there were no approved drug-specific reversal agents for the NOACs.  A number of drugs are currently under development. [146, 147]

Due to the short half-life of FXa inhibitors, discontinuation of the drugs suffice in clinical situations in which there is time to await spontaneous clearance.

Currently, PCCs can be used to address severe bleeding in patients taking NOACs when administered in high enough dosages. Some guidelines suggest an initial dose of 25 to 50 U/kg of PCCs in life-threatening emergencies, to be repeated if necessary.

Idarucizumab (Praxbind)

Idarucizumab is a humanized antibody fragment directed against dabigatran. This agent has been shown to completely reverse the anticoagulant effect of dabigatran within minutes; on October 16, 2015, it was approved by the FDA as an antidote for dabigatran. [148, 149, 150, 151]

Andexanet alfa

Andexanet alfa is a recombinant, modified FXa molecule that acts as a decoy protein that is catalytically inactive but has a high affinity for FXa inhibitors. It is being developed as an antidote for apixaban, edoxaban, and rivaroxaban. Andexanet alfa has been shown to reverse the anticoagulant effects of apixaban and rivaroxaban in human volunteers, and more studies are ongoing. [152]

Aripazine (PER977, ciraparantag)

Aripazine is a synthetic small molecule that has broad activity against both old (heparin, low molecular weight heparin) and new oral anticoagulants (dabigatran, rivaroxaban, apixaban, edoxaban). A 2014 study of human volunteers demonstrated that administration of aripazine reversed the prolonged clotting time caused by edoxaban. Further human trials are ongoing. [153]


Pharmacologic Thrombolysis

Use of thrombolytic medications to lyse deep venous thrombosis can cause intracranial bleeding, though this is infrequent, and death or impairment can result. Accordingly, careful assessment of the indications for lysis against the possibility of bleeding must be carried out before pharmacologic thrombolysis is attempted.

The need should be compelling when thrombolysis is considered in a setting of known contraindications. Factors such as recent surgery, stroke, gastrointestinal or other bleeding, and underlying coagulopathy increase the bleeding risk when the thrombolytic medication is administered. The process of obtaining informed consent should include a discussion of these risks.


General Principles of Endovascular Intervention

Percutaneous transcatheter treatment of patients with deep venous thrombosis (DVT) consists of thrombus removal with catheter-directed thrombolysis, mechanical thrombectomy, angioplasty, and/or stenting of venous obstructions. Consensus has been reached regarding indications for the procedure, although it is based on midlevel evidence from nonrandomized controlled trials. The goals of endovascular therapy include reducing the severity and duration of lower-extremity symptoms, preventing pulmonary embolism, diminishing the risk of recurrent venous thrombosis, and preventing postthrombotic syndrome. A randomized controlled trial comparing catheter-directed thrombolysis to conventional anticoagulation demonstrated a lower incidence of postthrombotic syndrome and improved iliofemoral patency in patients with a high proximal DVT and low risk of bleeding. [154]

Indications for intervention include the relatively rare phlegmasia or symptomatic inferior vena cava thrombosis that responds poorly to anticoagulation alone, or symptomatic iliofemoral or femoropopliteal DVT in patients with a low risk of bleeding. Contraindications are the same as those for thrombolysis in general. Absolute contraindications include active internal bleeding or disseminated intravascular coagulation, a cerebrovascular event, trauma, or neurosurgery within 3 months. Unfortunately, most patients with DVT have absolute contraindications to thrombolytic therapy. The American College of Chest Physicians (ACCP) consensus guidelines recommend thrombolytic therapy only for patients with massive ileofemoral vein thrombosis associated with limb ischemia or vascular compromise. [113, 155]

For more information, see Inferior Vena Caval Thrombosis.

Percutaneous mechanical thrombectomy devices are a popular adjunct to catheter-directed thrombolysis. Although these devices may not completely remove thrombus, they are effective for debulking and for minimizing the dose and time required for infusing a thrombolytic. Percutaneous mechanical thrombectomy has developed as an attempt to shorten treatment time and avoid costly ICU stays during thrombolytic infusion. The most basic mechanical method for thrombectomy is thromboaspiration, or the aspiration of thrombus through a sheath. Mechanical disruption of venous thrombosis has the potential disadvantage of damaging venous endothelium and valves, in addition to thrombus fragmentation and possible pulmonary embolism.

For more information, see Percutaneous Transcatheter Treatment of Deep Venous Thrombosis.


Surgical Thrombectomy

Surgical thrombus removal has traditionally been used in patients with massive swelling and phlegmasia cerulea dolens. In many patients, fibrinolysis alone is highly effective, and it has become the primary treatment of choice for many forms of venous and arterial thrombosis. Unfortunately, when thrombosis is extensive, fibrinolysis alone may be inadequate to dissolve the volume of thrombus present. Even when the bulk of the thrombus is not excessive, many patients with thrombosis are poor candidates for fibrinolysis because of recent surgery or trauma involving the central nervous system or other noncompressible areas.

Precisely defining the location and extent of thrombosis before considering any surgical approach to the problem is important. Duplex ultrasonography may sometimes be sufficient for this purpose, but venography (including routine contralateral iliocavography) is a more reliable guide to the anatomy and the particular pathology that must be addressed.

The patient must be heparinized before the procedure. Traditional venous thrombectomy is performed by surgically exposing the common femoral vein and saphenofemoral junction through a longitudinal skin incision. A Fogarty catheter is passed through the clot, and the balloon is inflated and withdrawn, along with the clot. However, care must be taken to avoid dislodging the clot or breaking it into small fragments because pulmonary embolus will result.

A proximal balloon or a temporary caval filter may be used to reduce the likelihood of embolization. Venography is mandatory to confirm the clearance of the thrombus. Back bleeding does not indicate clot clearance because a patent valve can block flow, or flow can be present with patent tributaries.

Venous valves may sometimes prevent the passage of a catheter in a retrograde direction down the leg. When this happens, the leg may be wrapped tightly with an Esmarch bandage in an attempt to force clot extrusion. After the thrombus has been removed, construction of a small arteriovenous fistula may assist in maintaining patency by increasing the flow velocity through a thrombogenic iliofemoral venous segment and promoting collateral development. The fistula is usually performed between the saphenous vein and the femoral vein. To reduce the likelihood of rethrombosis, heparin anticoagulation is usually initiated before surgery, continued during the procedure, and maintained for 6-12 months afterward. Leg compression devices are useful to maintain venous flow.

Outcomes from multiple studies have shown rethrombosis rates around 12% when a temporary arteriovenous fistula is used. Optimal results were found in thrombosis less than 7 days, clearance of thrombus from the external and internal iliac veins, intraoperative venography, early ambulation, and religious use of compression stockings. In a prospective randomized study from Sweden comparing surgery with anticoagulation, at 5 years, 37% of operated patients were asymptomatic, compared with just 18% in the anticoagulation group. Vein patency was 77% in the surgical group compared with just 30% in the anticoagulation group. [156]

Table. Surgical Thrombectomy with Temporary Arteriovenous Fistula in Early Iliac Vein Patency [157] (Open Table in a new window)

Study and Number of Patients

Patent Iliac Vein

Delin (13)


Plate (31)


Piquet (92)


Einarsson (51)


Juhan (42)


Vollmar (93)


Kniemeyer (185)


Neglen (48)




Total (555)



Placement of Inferior Vena Cava Filters

Inferior vena cava filters are not recommended in patients with acute venous thromboembolism (VTE) on anticoagulant therapy. [113]  These filters were developed in an attempt to trap emboli and minimize venous stasis. In most patients with deep venous thrombosis (DVT), prophylaxis against the potentially fatal passage of thrombus from the lower extremity or pelvic vein to the pulmonary circulation is adequately accomplished with anticoagulation. An inferior vena cava filter is a mechanical barrier to the flow of emboli larger than 4 mm.

In the past, inferior vena cava filters were placed in 4.4% of patients. Recent use was documented in 14% of patients with DVT; this rate was perhaps due to broadened indications with the introduction of removable filters. Temporary or removable filters, all of which may also be left as permanent, permit transient mechanical pulmonary embolism (PE) prophylaxis. This option may be useful in the setting of polytrauma, head injury, hemorrhagic stroke, known VTE, or major surgery when PE prophylaxis must be maintained during a short-term contraindication to anticoagulation.

In a randomized trial, the addition of an inferior vena cava filter to anticoagulation for DVT increased the risk of recurrent DVT (11.6% to 20.8%) and did not improve the 2-year survival rate. However, the filter group had significantly fewer PEs (1.1% vs 4.8%). Of note was the risk of major bleeding at 3 months (10.5%). This result agrees with other reports and highlights the usual trade-off of prophylaxis with a filter versus anticoagulation and the respective complication risks of new DVT (peripheral to the filter) versus major hemorrhage. In the elderly patient with an increased risk of bleeding, and particularly if the patient is at risk for trauma, the risk and benefits may favor use of a filter.

Catheter-directed thrombolysis does not add to the risk of PE to warrant routine filter placement. However, for patients with contraindications to pharmacologic lysis in whom a percutaneous mechanical thrombectomy device is to be used, a filter may be a useful adjunct. [158]

The ideal vena cava filter would trap venous emboli while maintaining normal venous flow. Many different filter configurations have been used, but the current benchmark remains the Greenfield filter with the longest long-term data. Patency rates greater than 95% and recurrent embolism rates of less than 5% have been demonstrated by numerous studies. The conical shape allows central filling of emboli while allowing blood on the periphery to flow freely. Numerous other filters with similar track records have since been developed, including filters that can be removed.

Regardless of the type of filter placed, the technique remains the same. Local anesthetic is used to anesthetize either the groin for a femoral vein approach or the neck for a jugular vein approach. A single wall needle is used under ultrasonic guidance to enter the target vein, and a 0.035-inch guidewire is passed into the inferior vena cava. A venogram is performed to identify the renal veins and measure the diameter of the vena cava to ensure the cava is not too big for the filter. Intravascular ultrasound (IVUS) can also be used for this purpose. It has the added benefit of not only allowing for bedside filter placement in sick intensive care unit (ICU) patients, but it also obviates the need for IV contrast. The correct filter location traditionally entails an infra-renal fixation with central filter extension to the level of the renal veins. Placement in the suprarenal inferior vena cava or superior vena cava may be indicated in some situations.

American Heart Association recommendations for inferior vena cava filters include the following [10] :

  • Confirmed acute proximal DVT or acute PE in patient with contraindication for anticoagulation (this remains the most common indication for inferior vena cava filter placement)

  • Recurrent thromboembolism while on anticoagulation

  • Active bleeding complications requiring termination of anticoagulation therapy

Relative contraindications include the following:

  • Large, free-floating iliofemoral thrombus in high-risk patients

  • Propagating iliofemoral thrombus while on anticoagulation

  • Chronic PE in patient with pulmonary hypertension and cor pulmonale

  • Patient with significant fall risk

For more information, see Inferior Vena Caval Thrombosis and Inferior Vena Cava Filters.


Replacement of Venous Valves

Percutaneously placed bioprosthetic venous valves are under development and may provide a minimally invasive therapy to the long-term complication of postthrombotic syndrome due to valve destruction. If successful, this approach may provide a percutaneous therapeutic alternative for patients with primarily palliative options to manage their venous reflux symptoms. An effective therapy should diminish one of the primary indications for aggressive thrombolytic therapy for acute deep venous thrombosis.


Use of Elastic Compression Stockings

Postthrombotic syndrome (PTS) affects approximately 50% of patients with deep venous thrombosis (DVT) after 2 years. Elderly patients and patients with recurrent ipsilateral DVT have the highest risk. Below-the-knee elastic compression stockings (ECS) assist the calf muscle pump and reduce venous hypertension and venous valvular reflux. This reduces leg edema, aids the microcirculation, and prevents venous ischemia.

In a randomized controlled study from an Italian university setting involving 180 patients who presented with a first episode of symptomatic proximal DVT, Prandoni and colleagues found below-the-knee ECS to have value for the prevention of PTS. After conventional anticoagulation with heparin, patients were discharged on therapeutic warfarin for 3-6 months and randomly assigned to the control group (no ECS) or the ECS group. Graduated compression stockings with ankle pressures of 30-40 mm Hg were given to the participants, who were required to wear them daily on the affected leg or legs over 2 years. Ninety percent of trial participants were compliant (wore the stockings for at least 80% of daytime hours), and 5-year cumulative data was evaluated to compare the incidence of PTS between the groups. [159]

A standardized validated scale was used to assess symptoms, severity, and/or progression of PTS. PTS occurred in 26% of patients who wore ECS compared with 49% of patients without ECS. All patients with PTS except one developed manifestations of the syndrome within the first 2 years after the initial diagnosis of DVT. The number of patients who need to be treated with ECS was estimated at 4.3 to prevent one case of PTS. The adjusted hazard ratio was 0.49 (CI 0.29-0.84, P = .011) in favor of ECS. Almost 50% of their patients with proximal DVT developed PTS within 2 years.

The regular use of graduated elastic compression stockings reduced the incidence of PTS by 50%. The authors also noted that the benefit conferred by ECS was not related to the rate of recurrent DVT, which was identical in both groups. The authors strongly recommended the early use and widespread implementation of graduated elastic stockings with adequate anticoagulant therapy for symptomatic proximal DVT to prevent the development of PTS.

The Eighth ACCP Conference on Antithrombotic and Thrombolytic Therapy observed that PTS occurs in 20-50% of patients with objectively confirmed DVT and assigned a grade 1A recommendation for the use of graduated elastic compression stockings for 2 years after the onset of proximal DVT. [144, 160] With the adoption of outpatient therapy for proximal DVT, the initial management of DVT increasingly becomes the responsibility of the emergency physician.

More recently, the ACCP recommends against the routine use of compression stockings in patients with acute DVT to prevent postthrombotic syndrome. [113]



Controversy exists regarding the role of ambulation in the therapy of deep venous thrombosis (DVT). A study by Partsch reviewed the myths surrounding immediate ambulation and compression in the patient with newly diagnosed DVT and concluded that early ambulation and compression is not associated with any significant risk of pulmonary embolism (PE). [161] It is well recognized from the older literature that almost 50% of patients with acute proximal DVT have evidence, based on V/Q pulmonary scanning, of asymptomatic PE at baseline. Analyzing the effect of ambulation and compression in this patient cohort focused on the development of a new PE, the relief of pain and swelling, and the reduction in the incidence and severity of postthrombotic syndrome (PTS).

The authors cited 2 small previous studies that demonstrated that the incidence of a new PE after initiation of anticoagulant therapy with a low-molecular-weight heparin (LMWH) did not increase significantly in patients treated with early ambulation and compression. They had previously reported their own prospective cohort study of 1289 patients with acute DVT treated as outpatients with LMWH, early ambulation, and compression. Partsch et al reported that only 77 of 1289 patients (5.9%) developed a new PE, only 6 of 1289 patients (0.4%) of these were symptomatic, and only 3 deaths (0.23%) were attributed to the PE. This was not significantly different than historical controls.

A systematic review by Kahn et al found that in patients with acute DVT, early walking exercise is safe and may help to reduce acute symptoms and that in patients with previous DVT, exercise training does not increase leg symptoms acutely and may help to prevent or improve the postthrombotic syndrome. [162]

In Europe, early ambulation and compression has been the mainstay of adjunctive treatment for DVT. In North America, the unsubstantiated fear of dislodging clots by ambulation led clinicians to recommend bed rest and leg elevation to their patients. The authors explained that bed rest promotes venous stasis, which is a major risk factor for DVT and, therefore, may actually enhance thrombus propagation and the risk of subsequent PE.

The authors also cited a number of other studies that revealed a significant decrease in leg swelling (using leg circumference measures) and pain (analog pain scales and quality of life scores) with early ambulation and compression. They also recognized the limited data that are available to assess the effect of early ambulation and compression on the subsequent development of PTS. In their own small trial, they reported a trend toward a lower incidence of PTS. They conceded that a larger, long-term study would be required. Nevertheless, they strongly recommended early ambulation for their patients in addition to elastic compression stockings.

The ACCP Consensus Conference on Antithrombotic and Thrombolytic Therapy for venous thromboembolism also recommended ambulation as tolerated for patients with DVT. [144, 160] Therefore, early ambulation on day 2 after initiation of outpatient anticoagulant therapy in addition to effective compression is strongly recommended. Early ambulation without ECS is not recommended. The fear of dislodging clots and precipitating a fatal PE is unfounded.


Treatment of Superficial Thrombophlebitis

Superficial thrombophlebitis is often associated with deep venous thrombosis (DVT) in two specific settings. The following high-risk groups require further evaluation for DVT:

  • Superficial thrombophlebitis in the absence of coexisting venous varices and no other obvious etiology

  • Involvement of the greater saphenous vein above the knee, especially if it extends to the saphenofemoral junction (These latter patients should be treated as having proximal vein DVT and treated with full anticoagulant therapy.)

Uncomplicated superficial thrombophlebitis may be treated symptomatically with heat, nonsteroidal anti-inflammatory agents (NSAIDs), and compression hose. Bed rest is not recommended.

Some centers recommend full anticoagulation for high-risk patients with isolated superficial thrombophlebitis. Some physicians may anticoagulate high-risk patients with negative initial study results until follow-up surveillance studies are completed. An alternative approach involves symptomatic care alone with close follow-up and repeated noninvasive testing in 1 week. Full anticoagulation is then reserved only for those patients with proven proximal vein DVT.


Treatment of Axillary and Subclavian Vein Thrombosis

This was first described by Paget in 1875 and von Schrötter in 1884 and is sometimes referred to as Paget–von Schrötter syndrome. The pathophysiology is similar to that of deep venous thrombosis (DVT), and the etiologies overlap. The incidence is lower than that of lower extremity DVT because of decreased hydrostatic pressure, fewer venous valves, higher rates of blood flow, and less frequent immobility of the upper arm.

Thoracic outlet compression from cervical ribs or congenital webs may precipitate axillary/subclavian venous thrombosis. Catheter-induced thrombosis is increasingly a common cause of this condition. The increased use of subclavian catheters for chemotherapy and parenteral nutrition has resulted in a dramatic increased incidence of proven thrombosis. Similarly, pulmonary artery catheters are associated with a high incidence of internal jugular and subclavian vein thrombosis. Pulmonary embolism (PE) occurs in approximately 10% of patients. Fatal or massive PE is extremely rare.

Ultrasonography and venography are the diagnostic tests of choice. Ultrasonographic findings may be falsely negative because of collateral blood flow. Duplex ultrasonography is accurate for the evaluation of the internal jugular vein and its junction with the subclavian vein where the innominate vein begins.

Thrombolytic therapy is the treatment of choice for axillary/subclavian venous thrombosis. Restoration of venous patency is more critical for the prevention of chronic venous insufficiency in the upper extremity. Thrombolysis is best accomplished with local administration of the thrombolytic agent directly at the thrombus. After completion of a venographic study, a catheter is floated up to the site of the clot, and the thrombolytic agent is administered as a direct infusion. Venographic assessment for clot lysis is repeated every 4-6 hours until venous patency is restored. Heparin is usually given concurrently to prevent rethrombosis.

In the presence of anatomic abnormalities, surgical therapy is recommended to minimize long-term morbidity and recurrence. Catheter-induced thrombosis may require removal of the device. Locally infused thrombolytic agents have been used successfully and are currently the treatment of choice.


Prophylaxis of Deep Venous Thrombosis

Prevention of deep venous thrombosis (DVT) has long been studied in various clinical situations with varying degrees of success. Primary prophylaxis is directed toward acting on one or more components of the Virchow triad, affecting blood flow, coagulation, or vessel wall endothelium. Methods of prophylaxis may be generally divided into mechanical and pharmacologic. Many pharmacologic agents are currently available to prevent thrombosis. Agents that retard or inhibit the process belong under the general heading of anticoagulants. Agents that prevent the growth or formation of thrombi are properly termed antithrombotics and include anticoagulants and antiplatelet drugs, whereas thrombolytic drugs lyse existing thrombi.

Surgical patients undergoing general anesthesia have been extensively studied. Studies of pneumatic compression in cardiac surgery and neurosurgical patients have shown a distinct improvement in the incidence of DVT without the added risk of bleeding. [163, 164] However, the effect is less impressive in higher-risk patients, and compliance can be difficult. Routine use of anticoagulant prophylaxis after cardiac surgery is discouraged. [165]  Kolluri et al showed no benefit of prophylactic postoperative fondaparinux following after coronary artery bypass graft (CABG) surgery. [165]  

Timing and duration of prophylactic agents has also been determined to have a significant effect on the development of DVT. Early prophylaxis in surgical patients with low molecular weight heparin has been associated with significant reductions in postoperative venous thrombosis. If surgery is delayed, then prophylaxis with low-dose unfractionated heparin or low molecular weight heparin should be initiated at the time of admission and discontinued prior to surgery.

Major surgical and high-risk orthopedic procedures place patients at risk for deep venous thrombosis and venous thromboembolism, including pulmonary embolism. Complications of DVT include postphlebitic syndrome or death from pulmonary embolism. Therefore, prophylaxis with anticoagulant medications, as well as the adjunct use of mechanical devices, is essential. The most effective treatment protocol for a patient must be determined on a case-by-case basis and account for the risk-benefit ratio in each situation. A risk stratification protocol, such as that developed by the American College of Chest Physicians (ACCP), is recommended to determine the appropriate level and method of treatment.

For more information, see Deep Venous Thrombosis Prophylaxis.