Epidemiology
Venous thromboembolism (VTE) is a relatively common medical problem with a yearly incidence of approximately 1 in every 1000 adults.[1][2][3] Approximately two-thirds present as DVT alone, and one third present as pulmonary embolism (with or without concomitant symptoms of DVT).[2]
The incidence of DVT during pregnancy or the postnatal period is approximately 1 per 1000 live births.[4] Other clinical characteristics confer widely variable incidences of DVT. For instance, orthopaedic surgery patients have a DVT incidence ranging from approximately 1% to 4% depending on the utilisation of pharmacological prophylaxis, while the incidence in acutely ill medical patients is approximately 0.5% to 6%, depending heavily on the method of diagnosis, inclusion of asymptomatic versus only symptomatic VTE, utilisation of pharmacological prophylaxis, and duration of follow-up.[5] In critically ill patients, an incidence as high as 37.2% has been reported.[6] The population incidence is increasing slowly as the proportion of the population that is older increases, and as testing for DVT using ultrasound and testing for pulmonary embolism using multi-detector chest computed tomographic angiography increases.
Risk factors
Major surgery is a particularly significant risk if the patient required general or regional anaesthesia, as this is a component of the Wells score.[12]
Approximately 18% of all incident venous thromboembolism occurs within 3 months of major surgery. Reasons include postoperative immobilisation, inflammation, underlying comorbidity, and injury to the venous system in selected cases (e.g., total knee replacement).[3]
Approximately 20% of all incident venous thromboembolism (VTE) occurs either during a medical hospitalisation or within 2 months of a hospitalisation of 4 or more days.[1][2]
Reasons are the combination of immobilisation with acute and chronic medical comorbidities that are associated with VTE development, such as acute infection, heart failure, stroke, respiratory failure, and inflammatory conditions.[19][20][21][22] The use of intravenous catheters predisposes to hospital-associated DVT, both in the upper and lower extremities.[23]
Active cancer is particularly significant if treatment is ongoing, within 6 months, or palliative; this is a component of the Wells score.[12][24][25][26][27]
Many malignancies increase the risk for thrombosis through a variety of mechanisms, including activation of the coagulation system and restriction of flow due to vein compression. DVT rates are likely to be 4- to 7.5-fold higher in patients with cancer than in the general population,[24] and patients with metastatic cancer at the time of diagnosis are at especially increased risk.[25] Cancer-related therapies including surgery, some chemotherapeutic and biological agents, and use of vascular access devices also increase the risk of DVT.
Previous venous thromboembolism (VTE) is a component of the Wells score.[12]
Previous VTE predicts the risk for future events, with the magnitude of the risk being dependent on the presence or absence of provoking factors at the time of the initial event, sex of the patient, and other factors. In one systematic review, the rate of recurrence was 3.3% per patient-year for patients with a previous DVT due to a transient risk factor, and 7.4% per patient-year for patients with an unprovoked DVT.[28]
Paralysis, paresis, or recent plaster immobilisation of the lower extremities is a component of the Wells score.[12]
Patients with severe trauma are at increased risk of DVT even when the lower extremities are not involved.[29][30]
Patients with lower-extremity injuries that require surgery, such as leg, femur, or hip fracture, are at particularly increased risk, owing to vein injury coupled with effects of immobilisation and surgery.[31]
Non-surgical injuries (e.g., a fracture that requires casting) also increase the risk.
The factor V Leiden (FVL) mutation creates a variant factor V that is resistant to activated protein C. The relative risk of developing venous thromboembolism (VTE; particularly DVT) is approximately 3 to 4 times greater in patients who carry 1 copy of the FVL mutation (heterozygotes) compared with patients without this mutation. However, the absolute lifetime risk of developing VTE is low.
There is a strong interaction between use of oral contraceptives or hormone replacement therapy containing oestrogen and presence of FVL, likely because oestrogen also confers resistance to activated protein C. The relative risk increase of VTE is approximately 12-fold that of a non-carrier who does not use oestrogen.[36]
Homozygous carriers have a substantially higher risk of developing VTE compared with heterozygotes.
FVL carriers appear to have increased risk for DVT only, not for pulmonary embolism, an observation called the 'factor V Leiden paradox'.[37]
The prothrombin gene mutation is caused by a single nucleotide polymorphism (G20210A). Affected patients produce an excess amount of prothrombin. The relative risk of developing venous thromboembolism (VTE) is approximately 4 times greater than the unaffected population, but the absolute risk of thrombosis remains low.
There is a strong interaction between use of oral contraceptives or hormone replacement therapy containing oestrogen and presence of the prothrombin variant, with an approximately 7-fold increase in the risk of VTE.[36]
Homozygous carriers of the prothrombin gene mutation have substantially greater risk of developing VTE compared with heterozygotes.
Patients with a well-defined deficiency in protein C or protein S have a 5- to 6-fold greater risk of developing venous thromboembolic events, although the magnitude of this risk varies according to the degree of functional loss present.[36] The absolute risk of first pregnancy-associated venous thromboembolism (VTE) is 7.8% in protein C-deficient women and 4.8% in protein S-deficient women.[38] Thrombosis risk increases in a multiplicative fashion in the presence of other thrombophilic disorders. Both disorders are rare.
The prevalence of antithrombin deficiency disorders is low in cohorts of patients with venous thromboembolism (VTE; <1%). The magnitude of thrombosis risk varies depending on the degree of functional loss present. The absolute risk of first pregnancy-associated VTE in antithrombin-deficient women is 16.6%.[38]
Defined as an association of persistently detectable antiphospholipid antibodies with specified clinical features consisting of thrombosis and/or pregnancy-related morbidity, antiphospholipid antibody syndrome is often over-diagnosed, likely due to the relatively complex criteria for diagnosis and the possibility of false-positive laboratory tests. Criteria have been updated and clarified.[39]
In contrast to the hereditary thrombophilias, antiphospholipid syndrome likely predicts a higher risk for both initial and recurrent venous thromboembolism (VTE), and may be useful in informing decisions regarding use of anticoagulation for secondary prevention after the initial treatment of a VTE event.[40] Antiphospholipid antibodies have been reported in up to 14% of patients presenting with a VTE.[41]
Increased venous thromboembolism (VTE) risk occurs especially with inflammation, infection, and immobility.
Case reports and small series show greater incidence in patients with sickle cell anaemia, inflammatory bowel disease, Behcet's disease, HIV, primary pulmonary hypertension, hyperlipidaemia, diabetes mellitus, myeloproliferative diseases, and others, including systemic lupus erythematosus.[42]
Several medical disorders, especially heart failure, respiratory disease, acute ischaemic stroke, and acute infections, are associated with hospital-acquired DVT, though the incidence of DVT continues to accumulate for about 2 months after hospital admission.[19][20][21][22]
Liver disease, even when causing prolongation of coagulation times, increases the risk for thrombosis.[43]
Mechanical ventilation has been associated with increased DVT in many studies of critically ill patients, though may be a marker of disease severity and underlying respiratory disease.[6][44][45]
Coronavirus disease 2019 (COVID-19), an infection caused by the novel coronavirus severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has been associated with risk for VTE, though the risk for pulmonary embolism may predominate over the risk for DVT.[9] See our topic Coronavirus disease 2019 (COVID-19).
The absolute risk of developing a DVT in women who take oestrogen-containing oral contraceptives is low. The risk of developing an oral contraceptive-related DVT is associated with the presence of a classic thrombophilia and is also associated with obesity and smoking. Risk is greatest in the first year of use.
For contraceptives, all preparations that contain oestrogen are associated with risk for venous thromboembolism (VTE). Mechanism of delivery (oral, transdermal, transvaginal) and the 'generation' of oral combined contraceptives are associated with broadly similar risks.[46][47] However, oral contraceptive pills containing third-generation progestins (such as desogestrel) or fourth-generation progestins (such as drospirenone) may be associated with greater risk of VTE compared with levonorgestrel.[48][49] When used for the indication of hormone replacement, the transdermal route appears to confer less risk than the oral route.[46][47]
Tamoxifen and raloxifen are associated with a 2- to 3-fold relative risk of developing DVT, particularly in patients with a thrombophilic condition, such as factor V Leiden.
Thalidomide most commonly causes DVT when used as a cancer chemotherapeutic agent. Many other chemotherapeutic agents can also increase the risk. Concomitant prophylaxis is suggested for certain diseases and regimens.[50]
Erythropoietin is associated with an increased risk of DVT in patients with cancer.
Patients who develop antibodies to adalimumab (a tumour necrosis factor alpha inhibitor) frequently develop venous thrombosis.[51]
Androgen-deprivation therapies used in prostate cancer increase VTE risk between about 1.5- and 2.5-fold depending upon the agent.[52]
Testosterone replacement therapy, both in men with and without demonstrable hypogonadism, has been associated with a 2-fold increased VTE risk.[53]
Non-steroidal anti-inflammatory drugs (NSAIDs), as a class, are associated with an increased rate of VTE. Risk attributable to individual NSAIDs is unknown.[54][55]
Randomised clinical trials and retrospective cohort studies have shown that high body mass index (especially >30 kg/m²) is associated significantly with DVT development.[32][56]
Mechanisms may include relative immobilisation, reduced venous flow rates, underlying inflammatory state, and greater frequency of co-existing comorbidities.
The Emerging Risk Factors Collaboration (731,728 participants) found a correlation between current smoking and the risk of venous thromboembolism, with a hazard ratio of 1.38.[32]
The absolute risk with air travel appears to be small. The risk appears to be increased in patients with an elevated baseline risk of venous thromboembolism (VTE) such as those with a previous VTE, recent surgery or trauma, obesity, limited mobility, or advanced age. The duration of flight associated with increased risk is uncertain, but flights longer than about 4 hours are likely associated with elevated risk.[57]
Use of this content is subject to our disclaimer