Deep venous thrombosis (DVT) is clotting of blood in a deep vein of an extremity (usually calf or thigh) or the pelvis. DVT is the primary cause of pulmonary embolism. DVT results from conditions that impair venous return, lead to endothelial injury or dysfunction, or cause hypercoagulability. DVT may be asymptomatic or cause pain and swelling in an extremity; pulmonary embolism is an immediate complication. Diagnosis is by history and physical examination and is confirmed by objective testing, typically with duplex ultrasonography. d-Dimer testing is used when DVT is suspected; a negative result helps to exclude DVT, whereas a positive result is nonspecific and requires additional testing to confirm DVT. Treatment is with anticoagulants. Prognosis is generally good with prompt, adequate treatment. Common long-term complications include venous insufficiency with or without the postphlebitic syndrome.
DVT occurs most commonly in the lower extremities or pelvis (see Fig. 1: Peripheral Venous Disorders: Deep veins of the legs.). It can also develop in deep veins of the upper extremities (4 to 13% of DVT cases).
Lower extremity DVT is much more likely to cause pulmonary embolism (PE), possibly because of the higher clot burden. The superficial femoral and popliteal veins in the thighs and the posterior tibial and peroneal veins in the calves are most commonly affected. Calf vein DVT is less likely to be a source of large emboli but can propagate to the proximal thigh veins and from there cause PE. About 50% of patients with DVT have occult PE, and at least 30% of patients with PE have demonstrable DVT.
Many factors can contribute to DVT (see Table 1: Peripheral Venous Disorders: Risk Factors for Venous Thrombosis). Cancer is a risk factor for DVT, particularly in elderly patients and in patients with recurrent thrombosis. The association is strongest for mucin-secreting endothelial cell tumors such as bowel or pancreatic cancers. Occult cancers may be present in patients with apparently idiopathic DVT, but extensive workup of patients for tumors is not recommended unless patients have major risk factors for cancer or symptoms suggestive of an occult cancer.
Lower extremity DVT most often results from impaired venous return (eg, in immobilized patients), endothelial injury or dysfunction (eg, after leg fractures), or hypercoagulability.
Upper extremity DVT most often results from endothelial injury due to central venous catheters, pacemakers, or injection drug use. Upper extremity DVT occasionally occurs as part of superior vena cava (SVC) syndrome or results from a hypercoagulable state or subclavian vein compression at the thoracic outlet. The compression may be due to a normal or an accessory first rib or fibrous band (thoracic outlet syndrome) or occur during strenuous arm activity (effort thrombosis, or Paget-Schroetter syndrome, which accounts for 1 to 4% of upper extremity DVT cases).
DVT usually begins in venous valve cusps. Thrombi consist of thrombin, fibrin, and RBCs with relatively few platelets (red thrombi); without treatment, thrombi may propagate proximally or travel to the lungs.
Common complications include
Much less commonly, acute DVT leads to phlegmasia alba dolens or phlegmasia cerulea dolens, both of which, unless promptly diagnosed and treated, can result in venous gangrene.
In phlegmasia alba dolens, a rare complication of DVT during pregnancy, the leg turns milky white. Pathophysiology is unclear, but edema may increase soft-tissue pressure beyond capillary perfusion pressures, resulting in tissue ischemia and wet gangrene.
In phlegmasia cerulea dolens, massive iliofemoral venous thrombosis causes near-total venous occlusion; the leg becomes ischemic, extremely painful, and cyanotic. Pathophysiology may involve complete stasis of venous and arterial blood flow in the lower extremity because venous return is occluded or massive edema cuts off arterial blood flow. Venous gangrene may result.
Rarely, venous clots can become infected. Jugular vein suppurative thrombophlebitis (Lemierre syndrome), a bacterial (usually anaerobic) infection of the internal jugular vein and surrounding soft tissues, may follow tonsillopharyngitis and is often complicated by bacteremia and sepsis. In septic pelvic thrombophlebitis, pelvic thromboses develop postpartum and become infected, causing intermittent fever. Suppurative (septic) thrombophlebitis, a bacterial infection of a superficial peripheral vein, comprises infection and clotting that usually is caused by venous catheterization.
Symptoms and Signs
DVT may occur in ambulatory patients or as a complication of surgery or major medical illness. Among high-risk hospitalized patients, most deep vein thrombi occur in the small calf veins, are asymptomatic, and may not be detected.
When present, symptoms and signs (eg, vague aching pain, tenderness along the distribution of the veins, edema, erythema) are nonspecific, vary in frequency and severity, and are similar in arms and legs. Dilated collateral superficial veins may become visible or palpable. Calf discomfort elicited by ankle dorsiflexion with the knee extended (Homans sign) occasionally occurs with distal leg DVT but is neither sensitive nor specific. Tenderness, swelling of the whole leg, > 3 cm difference in circumference between calves, pitting edema, and collateral superficial veins may be most specific; DVT is likely with a combination of ≥ 3 in the absence of another likely diagnosis (see Table 2: Peripheral Venous Disorders: Probability of Deep Venous Thrombosis Based on Clinical Factors).
Low-grade fever may be present; DVT may be the cause of fever without an obvious source, especially in postoperative patients. If PE occurs, symptoms may include shortness of breath and pleuritic chest pain (see Pulmonary Embolism: Symptoms and Signs).
|Probability of Deep Venous Thrombosis Based on Clinical Factors
Tenderness along distribution of the veins in calf or thigh
Swelling of entire leg
Calf swelling (> 3 cm difference in circumference between calves, measured 10 cm below tibial tuberosity)
Pitting edema greater in affected leg
Dilated collateral superficial veins
Cancer (including cases in which treatment was stopped within 6 mo)
Immobilization of lower extremity (eg, due to paralysis, paresis, casting, or recent long-distance travel)
Surgery leading to immobility for > 3 days within the past 4 wk
Probability equals the number of factors, subtracting 2 if another diagnosis is as likely as or more likely than DVT.
High probability: ≥ 3 points
Moderate probability: 1–2 points
Low probability: ≤ 0 points
Based on data from Anand SS, Wells PS, Hunt D, et al: Does this patient have deep vein thrombosis? Journal of the American Medical Association 279 (14):1094–1099, 1998.
Common causes of asymmetric leg swelling that mimic DVT are soft-tissue trauma, cellulitis, pelvic venous or lymphatic obstruction, and popliteal bursitis (Baker cyst) that obstructs venous return. Abdominal or pelvic tumors that obstruct venous or lymphatic return are less common causes. Use of drugs that cause dependent edema (eg, dihydropyridine Ca channel blockers, estrogen, high-dose opioids), venous hypertension (usually due to right heart failure), and hypoalbuminemia cause symmetric bilateral leg swelling; swelling may be asymmetric if venous insufficiency coexists and is worse in one leg.
Common causes of calf pain that mimic acute DVT include venous insufficiency and postphlebitic syndrome; cellulitis that causes painful erythema of the calf; ruptured popliteal (Baker) cyst (pseudo-DVT), which causes calf swelling, pain, and sometimes bruising in the region of the medial malleolus; and partial or complete tears of the calf muscles or tendons.
History and physical examination help determine probability of DVT before testing (see Table 2: Peripheral Venous Disorders: Probability of Deep Venous Thrombosis Based on Clinical Factors). Diagnosis is typically by ultrasonography with Doppler flow studies (duplex ultrasonography). The need for additional tests (eg, d-dimer testing) and their choice and sequence depend on pretest probability and sometimes ultrasonography results. No single testing protocol is best; one approach is described in Fig. 2: Peripheral Venous Disorders: One approach to testing for suspected deep venous thrombosis..
Ultrasonography identifies thrombi by directly visualizing the venous lining and by demonstrating abnormal vein compressibility or, with Doppler flow studies, impaired venous flow. The test is > 90% sensitive and > 95% specific for femoral and popliteal vein thrombosis but is less accurate for iliac or calf vein thrombosis.
d-Dimer is a byproduct of fibrinolysis; elevated levels suggest recent presence and lysis of thrombi. d-Dimer assays vary in sensitivity and specificity; however, most are sensitive and not specific. Only the most accurate tests should be used. For example, a highly sensitive test is enzyme-linked immunosorbent assay (ELISA), which has a sensitivity of about 95%.
If pretest probability of DVT is low, DVT can be safely excluded in patients with a normal d-dimer level on a sensitive test. Thus, a negative d-dimer test can identify patients who have a low probability of DVT and do not require ultrasonography. However, a positive test result is nonspecific; because levels can be elevated by other conditions (eg, liver disease, trauma, pregnancy, positive rheumatoid factor, inflammation, recent surgery, cancer), further testing is necessary.
If pretest probability of DVT is moderate or high, d-dimer testing can be done at the same time as duplex ultrasonography. A positive ultrasound result confirms the diagnosis regardless of the d-dimer level. If ultrasonography does not reveal evidence of DVT, a normal d-dimer level helps exclude DVT. Patients with an elevated d-dimer level should have repeat ultrasonography in a few days or additional imaging, such as venography, depending on clinical suspicion..
Contrast venography was the definitive test for the diagnosis of DVT but has been largely replaced by ultrasonography, which is noninvasive, more readily available, and almost equally accurate for detecting DVT. Venography may be indicated when ultrasonography results are normal but pretest suspicion for DVT is high. The complication rate is 2%, mostly because of contrast dye allergy.
Noninvasive alternatives to contrast venography are being studied. They include MRI venography and direct MRI of thrombi using T1-weighted gradient-echo sequencing and a water-excitation radiofrequency pulse; theoretically, the latter test can provide simultaneous views of thrombi in deep veins and subsegmental pulmonary arteries (for diagnosis of PE).
If symptoms and signs suggest PE, additional imaging (eg, ventilation/perfusion [V/Q] scanning or CT pulmonary angiography) is required.
Determination of cause
Patients with confirmed DVT and an obvious cause (eg, immobilization, surgical procedure, leg trauma) need no further testing. Testing to detect hypercoagulability is controversial but is sometimes done in patients who have idiopathic (or unprovoked) DVT or recurrent DVT, in patients who have a personal or family history of other thromboses, and in young patients with no obvious predisposing factors. Some evidence suggests that presence of hypercoagulability does not predict DVT recurrence as well as clinical risk factors.
Screening patients with DVT for cancer has a low yield. Selective testing guided by complete history and physical examination aimed at detecting cancer is probably adequate.
Without adequate treatment, lower extremity DVT has a 3% risk of fatal PE; death due to upper extremity DVT is very rare. Risk of recurrent DVT is lowest for patients with transient risk factors (eg, surgery, trauma, temporary immobility) and greatest for patients with persistent risk factors (eg, cancer), idiopathic DVT, or incomplete resolution of past DVT (residual thrombus). A normal d-dimer level obtained after warfarin is stopped may help predict a relatively low risk of DVT or PE recurrence. Risk of venous insufficiency is difficult to predict. Risk factors for postphlebitic syndrome include proximal thrombosis, recurrent ipsilateral DVT, and body mass index (BMI) ≥ 22 kg/m2.
Treatment is aimed primarily at PE prevention (see also Pulmonary Embolism: Prevention) and secondarily at symptom relief and prevention of DVT recurrence, chronic venous insufficiency, and postphlebitic syndrome. Treatment of lower and upper extremity DVT is generally the same.
All patients with DVT are given anticoagulants, initially an injectable heparin (unfractionated or low molecular weight), followed by warfarin started within 24 to 48 h. Inadequate anticoagulation in the first 24 to 48 h may increase risk of recurrence or PE. Acute DVT can be treated on an outpatient basis unless severe symptoms require parenteral analgesics, other disorders preclude safe outpatient discharge, or other factors (eg, functional, socioeconomic) might prevent the patient from adhering to prescribed treatments.
General supportive measures include pain control with analgesics, which may include short (3- to 5-day) courses of an NSAID. Extended treatment with NSAIDs and aspirin should be avoided because their antiplatelet effects may increase the risk of bleeding complications. In addition, elevation of legs (supported by a pillow or other soft surface to avoid venous compression) is recommended during periods of inactivity. Patients may be as physically active as they can tolerate; there is no evidence that early activity increases risk of clot dislodgement and PE and may help to reduce the risk of the postphlebitic syndrome.
The anticoagulants most often used are the following:
LMWHs (eg, enoxaparin, dalteparin, tinzaparin—see Table 3: Pulmonary Embolism: Some Low Molecular Weight Heparin* Options in Thromboembolic Disease) are the initial treatment of choice because they can be given on an outpatient basis. LMWHs are as effective as UFH for reducing DVT recurrence, thrombus extension, and risk of death due to PE. Like UFH, LMWHs catalyze the action of antithrombin (which inhibits coagulation factor proteases), leading to inactivation of coagulation factor Xa and, to a lesser degree, factor IIa. LMWHs also have some antithrombin–mediated anti-inflammatory properties, which facilitate clot organization and resolution of symptoms and inflammation.
LMWHs are typically given sc in a standard weight-based dose (eg, enoxaparin 1.5 mg/kg sc once/day or 1 mg/kg sc q 12 h or dalteparin 200 units/kg sc once/day). Patients with renal insufficiency may be treated with UFH or with reduced doses of LMWH. Monitoring is not reliable because LMWHs do not significantly prolong the results of global tests of coagulation. Furthermore, they have a predictable dose response, and there is no clear relationship between the anticoagulant effect of LMWH and bleeding. Treatment is continued until full anticoagulation is achieved with warfarin. However, evidence suggests that LMWH is effective for long-term DVT treatment in high-risk patients, such as those with cancer. Thus, LMWH may become an acceptable alternative to warfarin for some patients, although warfarin is likely to be the treatment of choice for most patients because of its low cost and oral route of administration.
UFH may be used instead of LMWH for hospitalized patients and for patients who have renal insufficiency or failure (creatinine clearance 10 to 30 mL/min) because UFH is not cleared by the kidneys. UFH is given as a bolus and infusion (see Fig. 2: Pulmonary Embolism: Weight-based heparin dosing.) to achieve full anticoagulation, (eg, activated PTT [aPTT] 1.5 to 2.5 times that of the reference range). For outpatients, UFH 333 units/kg initial bolus, then 250 units/kg sc q 12 h can be substituted for IV UFH to facilitate mobility; the dose does not appear to need adjustment based on aPTT. Treatment is continued until full anticoagulation has been achieved with warfarin.
Complications of heparin include bleeding, thrombocytopenia (less common with LMWHs), urticaria, and, rarely, thrombosis and anaphylaxis. Long-term use of UFH causes hypokalemia, liver enzyme elevations, and osteopenia. Rarely, UFH given sc causes skin necrosis. Inpatients and possibly outpatients should be screened for bleeding with serial CBCs and, where appropriate, testing for occult blood in stool.
Bleeding due to overheparinization can be stopped with protamine sulfate. The dose is 1 mg protamine for each milligram of LMWH given as 1 mg in 20 mL of normal saline infused slowly over 10 to 20 min. If a 2nd dose is required, it should be one half the first dose. However, the precise dose is undefined because protamine only partially neutralizes LMWH inactivation of factor Xa. During all infusions, patients should be observed for hypotension and a reaction similar to an anaphylactic reaction. Because UFH given IV has a half-life of 30 to 60 min, protamine is not given to patients receiving UFH (eg, if UFH was given > 60 min beforehand) or is given at a dose based on the amount of heparin estimated to be remaining in plasma, based on the half-life of UFH.
Fondaparinux, a selective factor Xa inhibitor, may be used as an alternative to UFH or LMWH for the initial treatment of DVT or PE. It is given in a fixed dose of 7.5 mg sc once/day (10 mg for patients > 100 kg, 5 mg for patients < 50 kg). It has the advantage of fixed dosing and is less likely to cause thrombocytopenia.
Vitamin K antagonists, including warfarin, are the drugs of choice for long-term anticoagulation for all patients except pregnant women (who should continue to take heparin) and patients who have had new or worsening venous thromboembolism during warfarin treatment (who may be candidates for an inferior vena cava filter). Warfarin 5 to 10 mg can be started immediately with heparin. The elderly and patients with a liver disorder typically require lower warfarin doses. Therapeutic goal is an INR of 2.0 to 3.0. INR is monitored weekly for the first 1 to 2 mo of warfarin treatment and monthly thereafter; the dose is increased or decreased by 0.5 to 3 mg to maintain the INR within this range. Patients taking warfarin should be informed of possible drug interactions, including interactions with foods and nonprescription medicinal herbs (see Table 4: Pulmonary Embolism: Drug, Herbal Preparation, and Food Interactions With Warfarin).
Duration of treatment varies. Patients with transient risk factors for DVT (eg, immobilization, surgery) can usually stop taking warfarin after 3 to 6 mo. Patients with nonmodifiable risk factors (eg, hypercoagulability), idiopathic (or unprovoked) DVT with no known risk factors, or recurrent DVT should take warfarin for at least 6 mo and, in selected patients, probably for life unless complications occur.
Bleeding is the most common complication. Risk factors for severe bleeding (defined as life-threatening hemorrhage or loss of ≥ 2 units of blood in ≤ 7 days) include age ≥ 65; history of prior GI bleeding or stroke; recent MI; and coexisting anemia (Hct < 30%), renal insufficiency (serum creatinine > 1.5 mg/dL), or diabetes. In patients who are actively bleeding or may be at increased risk of bleeding, anticoagulation can be reversed with vitamin K; the dose is 1 to 2.5 mg po if INR is 5 to 9, 2.5 to 5 mg po if INR is > 9, and 5 to 10 mg IV (given slowly to avoid anaphylaxis) if hemorrhage occurs. If hemorrhage is severe, a transfusion of coagulation factors, fresh frozen plasma, or prothrombin complex concentrate should also be given. Selected patients with overanticoagulation (INR 5 to 9) who are neither actively bleeding nor at increased risk of bleeding can be managed by omitting 1 or 2 warfarin doses and monitoring INR more frequently, then giving warfarin at a lower dose. Rarely, warfarin causes skin necrosis in patients with protein C or S deficiency or factor V Leiden mutations.
Other anticoagulants, such as direct thrombin inhibitors (DTIs, eg, bivalirudin, desirudin, argatroban, given parenterally; dabigatran given po) and oral factor Xa inhibitors (eg, rivaroxaban, apixaban), are being evaluated for the treatment of DVT (see see Table 6: Pulmonary Embolism: Some Anticoagulation Options Other Than Heparin in Thromboembolic Disease).
Inferior vena cava filter (IVCF)
An IVCF may help prevent PE in patients with lower extremity DVT who have contraindications to anticoagulant therapy or in patients with recurrent DVT (or emboli) despite adequate anticoagulation. An IVCF is placed in the inferior vena cava just below the renal veins via catheterization of an internal jugular or femoral vein. Some IVCFs are removable and can be used temporarily (eg, until contraindications to anticoagulation subside or resolve). IVCFs reduce risk of acute embolic complications but can have longer-term complications (eg, venous collaterals can develop, providing a pathway for emboli to circumvent the IVCF, and increased risk of recurrent DVT). Also, IVCFs can dislodge or become obstructed by clot. Thus, patients with recurrent DVT or nonmodifiable risk factors for DVT may still require anticoagulation despite the presence of an IVCF. A clotted filter may cause bilateral lower extremity venous congestion (including acute phlegmasia cerulea dolens), lower body ischemia, and acute renal failure. Treatment for a dislodged filter is removal, using angiographic, or, if necessary, surgical methods. Despite widespread use of IVCFs, efficacy in preventing PE is unstudied and unproved. IVCFs should be removed whenever possible.
Thrombolytic (fibrinolytic) drugs
Streptokinase, urokinase, and alteplase lyse clots and may be more effective to prevent postphlebitic syndrome than heparin alone, but the risk of bleeding is higher than with heparin. Their use is under ongoing study, especially in patients with PE and right ventricular dysfunction and in combination with mechanical thrombolysis for extensive proximal DVT. Thrombolytic therapy alone may be indicated for large proximal thrombi, especially those in the iliofemoral veins, and for phlegmasia alba or cerulea dolens. Local perfusion with an indwelling catheter may be preferable to IV administration.
Surgery is rarely needed. However, thrombectomy, fasciotomy, or both are mandatory for phlegmasia alba dolens or phlegmasia cerulea dolens unresponsive to thrombolytics to try to prevent limb-threatening gangrene.
Patients at low risk of DVT (eg, those who are undergoing minor surgery but have no clinical risk factors for DVT, those who must be temporarily inactive for long periods, as during an airplane flight) should be encouraged to walk or otherwise move their legs periodically; no medical treatment is needed. Dorsiflexion 10 times/h is probably sufficient.
Patients at higher risk of DVT (eg, those undergoing minor surgery if they have clinical risk factors for DVT; those undergoing major surgery, especially orthopedic surgery, even without risk factors; bedbound patients with major medical illnesses) require additional preventive treatment (see Table 3: Peripheral Venous Disorders: Risk of Deep Venous Thrombosis and Pulmonary Embolism in Surgical Patients). Most of these patients can be identified and should receive DVT prophylaxis. Hospitalization itself is not considered a risk factor, and hospitalized patients not in one of these categories do not require routine DVT prophylaxis.
After surgery, elevating the legs and avoiding prolonged immobility, which places the legs in a dependent position thereby impeding venous return, can help. Additional treatment may involve low-dose UFH, LMWH, warfarin, fondaparinux, new oral anticoagulants, compression devices or stockings, or a combination, depending on patient's risk level, type of surgery (if applicable), projected duration of preventive treatment, contraindications, adverse effects, relative cost, ease of use, and local practice. Low-dose UFH 5000 units sc is given 2 h before surgery and q 8 to 12 h thereafter for 7 to 10 days or until patients are fully ambulatory. Bedbound patients who are not undergoing surgery are given 5000 units sc q 12 h until risk factors are reversed.
LMWHs are more effective than low-dose UFH for preventing DVT and PE, but widespread use is limited by cost. For example, enoxaparin 30 mg sc q 12 h, dalteparin 5000 units sc once/day, and tinzaparin 4500 units sc once/day appear to be are equally effective. Fondaparinux 2.5 mg sc once/day is at least as effective as LMWH in patients who are undergoing nonorthopedic surgery and is possibly more effective than LMWHs after orthopedic surgery.
Warfarin, using a target INR of 2.0 to 3.0, is proven to be effective in orthopedic surgery but is being used less frequently because alternative anticoagulants such as LMWHs and new oral anticoagulants are easier to administer.
New oral anticoagulants (eg, dabigatran, rivaroxaban, apixaban) are at least as effective and safe as LMWH for preventing DVT and PE after hip or knee replacement surgery but are more expensive than warfarin, and their cost-effectiveness requires further study. Aspirin is better than placebo but likely worse than LMWH and warfarin for preventing DVT and PE and is not recommended as the first-line method of prevention in most patients (see Table 5: Pulmonary Embolism: Risk of Deep Venous Thrombosis and Pulmonary Embolism in Surgical Patients).
Intermittent pneumatic compression (IPC) uses a pump to cyclically inflate and deflate hollow plastic leggings, providing external compression to the lower legs and sometimes thighs. IPC may be used instead of or in combination with anticoagulants after surgery. IPC is recommended for patients undergoing surgery associated with a high risk of bleeding in whom anticoagulant use may be contraindicated. IPC is probably more effective for preventing calf than proximal DVT. IPC is contraindicated in some obese patients who may be unable to apply the devices properly.
The benefit of graded compression stockings is questionable except for low-risk surgical patients and selected hospitalized medical patients. However, combining stockings with other preventive measures may be more protective than any single approach.
For elective neurosurgery, spinal cord injury, or multiple trauma, low-dose UFH (eg, 5000 units sc q 8 h), LMWH, or adjusted-dose warfarin is recommended.For hip and other lower extremity orthopedic surgery, LMWH, fondaparinux, or adjusted-dose warfarin is recommended. For patients undergoing total knee replacement and some other high-risk patients, IPC is also beneficial. For orthopedic surgery, preventive treatment may be started before or after surgery and continued for at least 14 days. Fondaparinux 2.5 mg sc once/day appears to be more effective to prevent DVT than LMWH for orthopedic surgery but may be associated with an increased risk of bleeding. For neurosurgery patients, physical measures (IPC, elastic stockings) have been used because intracranial bleeding is a concern; however, LMWH appears to be an acceptable alternative. Limited data support the combination of IPC, elastic stockings, and LMWH in patients with spinal cord injury or multiple trauma.
For patients who are at very high risk of venous thromboembolism and bleeding (eg, after major trauma) IPC is recommended until the bleeding risk subsides and anticoagulants can be given. The use of IVCF should be avoided unless DVT has been confirmed, except in highly selected patients.
Preventive treatment is also indicated for patients who have a major medical illnesses requiring bed rest (eg, MI, ischemic stroke, heart failure). Low-dose UFH is effective in patients who are not already receiving IV heparin or thrombolytics; IPC, elastic stockings, or both may be used when anticoagulants are contraindicated. After a stroke, low-dose UFH or LMWH can be used; IPC, elastic stockings, or both may be beneficial.
In patients with symptomatic DVT, primary prevention of venous insufficiency and postphlebitic syndrome is recommended; knee-high compression stockings providing 30 to 40 mm Hg pressure are used, although stockings with lower tension (20 to 30 mm Hg) can be considered if patients are unable to tolerate the higher tension stockings.
Last full review/revision August 2012 by Alexander G.G. Turpie, MD