Overview of the causes of venous thrombosis

INTRODUCTION — The most common presentations of venous thrombosis are deep vein thrombosis (DVT) of the lower extremity and pulmonary embolism. The causes of venous thrombosis can be divided into two groups: hereditary and acquired, and are often multiple in a given patient.

The inherited and acquired causes of venous thrombosis will be reviewed here (table 1) [1,2]. The diagnostic approach to the patient with suspected venous thrombosis, the evaluation and treatment of patients with documented venous thrombosis, and the various causes of upper extremity venous thrombosis are discussed separately. (See "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity" and "Evaluating adult patients with established venous thromboembolism for acquired and inherited risk factors" and "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)" and "Primary (spontaneous) upper extremity deep vein thrombosis" and "Brachial plexus syndromes" and "Catheter-related upper extremity venous thrombosis in adults".)

VIRCHOW TRIAD — A major theory delineating the pathogenesis of venous thromboembolism (VTE), often called Virchow triad [3,4], proposes that VTE occurs as a result of:

●Alterations in blood flow (ie, stasis)

●Vascular endothelial injury

●Alterations in the constituents of the blood (ie, inherited or acquired hypercoagulable state)

A risk factor for thrombosis can be identified in over 80 percent of patients with venous thrombosis. Furthermore, there is often more than one factor at play in a given patient. As examples:

●Fifty percent of thrombotic events in patients with inherited thrombophilia are associated with the additional presence of an acquired risk factor (eg, surgery, prolonged bed rest, pregnancy, oral contraceptives). Some patients have more than one form of inherited thrombophilia or more than one form of acquired thrombophilia and appear to be at even greater risk for thrombosis (see 'Multiple inherited thrombotic defects' below) [5].

●In a population-based study of the prevalence of venous thromboembolism (VTE), 56 percent of the patients had three or more of the following six risk factors present at the time of VTE: >48 hours of immobility in the preceding month; hospital admission, surgery, malignancy, or infection in the past three months; or current hospitalization (see 'Multiple acquired risk factors' below) [6].

Accordingly, many patients with VTE fulfill most or all of Virchow triad of stasis, endothelial injury, and hypercoagulability [7-9].

SUPERFICIAL VEIN THROMBOSIS — Superficial vein thrombosis (SVT), a less severe disorder than deep vein thrombosis (DVT), occurs in both inherited and acquired thrombophilic states and may progress to DVT and/or pulmonary embolism (PE) [10-13]. In 63 patients presenting with ultrasonically-confirmed SVT of the lower extremities as a first thrombotic episode, and in whom DVT, varicose veins, malignancy, and autoimmune disorders were absent, the following observations were made [14]:

●Twenty patients (32 percent) developed DVT at a median elapsed interval of four years.

●Fifteen patients (24 percent) had recurrent episodes of SVT.

●The odds ratios for the development of SVT in patients with factor V Leiden, the prothrombin G20210A mutation, or a deficiency of antithrombin, protein S or C were 6.1, 4.3, and 12.9, respectively.

In another meta-analysis of 21 studies of patients with SVT, the prevalence of DVT was 18 percent and PE was 7 percent [15].

This subject is discussed in depth separately. (See "Superficial vein thrombosis and phlebitis of the lower extremity veins".)

INHERITED THROMBOPHILIA

Common inherited hypercoagulable states — Inherited thrombophilia is a genetic tendency to venous thromboembolism. The most frequent causes of an inherited (primary) hypercoagulable state are the factor V Leiden mutation and the prothrombin gene mutation, which together account for 50 to 60 percent of cases. Defects in protein S, protein C, and antithrombin (formerly known as antithrombin III) account for most of the remaining cases (table 1) [16-19].

Each of these is discussed separately, as follows:

●Factor V Leiden mutation (see "Factor V Leiden and activated protein C resistance")

●Prothrombin gene mutation (see "Prothrombin G20210A")

●Protein S deficiency (see "Protein S deficiency")

●Protein C deficiency (see "Protein C deficiency")

●Antithrombin deficiency (see "Antithrombin deficiency")

Deep vein thrombosis — The thrombotic risk associated with the inherited thrombophilias has been assessed in two ways: evaluation of patients with DVT (table 2) and evaluation of families with thrombophilia. In a Spanish study of 2132 consecutive unselected patients with venous thromboembolism, for example, 12.9 percent had an anticoagulant protein deficiency (7.3 percent with protein S, 3.2 percent with protein C, and 0.5 percent with antithrombin). An additional 4.1 percent had elevated levels of antiphospholipid antibodies (aPL) [16].

Similar findings were noted in a series of 277 Dutch outpatients with deep vein thrombosis: 8.3 percent had an isolated deficiency of antithrombin, protein C, protein S, or plasminogen compared with 2.2 percent of controls [20]. The prevalence of a protein deficiency was only modestly greater in "high risk" patients with recurrent, familial, or juvenile onset deep vein thrombosis (9, 16, and 12 percent respectively). A higher frequency of inherited thrombophilia has also been noted in patients with thrombosis of visceral or cerebral vessels [21].

The overall 8 to 13 percent prevalence of an isolated anticoagulant protein deficiency in patients with deep vein thrombosis does not include the contribution of factor V Leiden or the prothrombin gene mutation, now considered to be the most common causes of inherited thrombophilia.

●The Physicians' Health Study and the Leiden Thrombophilia Study found a 12 to 19 percent prevalence of heterozygosity for the factor V Leiden mutation in patients with a first DVT (or pulmonary embolism in the Physicians' Health Study) compared with 3 to 6 percent in controls [22,23]. The prevalence reached 26 percent in the Physicians' Health Study in 31 men over the age of 60 with no identifiable precipitating factors [22]. (See "Factor V Leiden and activated protein C resistance".)

●The prevalence of the prothrombin gene mutation is approximately 6 to 8 percent in patients with deep vein thrombosis compared with 2 to 2.5 percent in controls [17,24]. (See "Prothrombin G20210A".)

Thus, the total prevalence of an inherited thrombophilia in subjects with a deep vein thrombosis ranges from 24 to 37 percent overall compared with about 10 percent in controls.

Thrombotic risk in families — The absolute risk of thrombosis among patients with inherited thrombophilia was evaluated in a report of 150 pedigrees, which compared the risk for thrombosis in individuals with inherited thrombophilia due to factor V Leiden or to antithrombin, protein C, or protein S deficiency (table 3) [12]. The lifetime probability of developing thrombosis compared with those with no defect was 8.5 times higher for carriers of protein S deficiency, 8.1 for antithrombin deficiency, 7.3 for protein C deficiency, and 2.2 for factor V Leiden. The thrombosis risk in affected women was also increased during pregnancy and the use of oral contraceptives (see 'Pregnancy' below).

Multiple inherited thrombotic defects — A second thrombotic defect can occur among patients with any of the causes of familial thrombophilia, particularly factor V Leiden [25-28]. (See "Factor V Leiden and activated protein C resistance", section on 'Venous thromboembolism'.) In one series, factor V Leiden was present in 4 of 14 patients with protein S deficiency and 6 of 15 with protein C deficiency [26]. Carriers of two (or more) defects seem to be at a higher risk for thrombosis than their relatives with a single defect. In one review of four studies, approximately 75 percent of the family members who were carriers of two defects had experienced thrombosis compared with 10 to 30 percent of the carriers of a single defect [25].

Inherited thrombophilia also may interact to increase the risk of venous thrombosis in patients with acquired causes of hypercoagulability, such as oral contraceptive use (table 2) [29] and pregnancy (see 'Combined acquired plus inherited risk factors' below).

The clinical characteristics and diagnosis of the five major causes of inherited thrombophilia: the factor V Leiden mutation, antithrombin deficiency, protein S and protein C deficiency, and the prothrombin gene mutation, are discussed in detail separately. (See "Factor V Leiden and activated protein C resistance" and "Antithrombin deficiency" and "Protein C deficiency" and "Protein S deficiency" and "Prothrombin G20210A".)

Other alleged inherited thrombophilias — The remainder of this section will briefly review other alleged causes of inherited thrombophilia. As will be seen, it remains unclear whether some of these disorders are actually associated with an increased risk of venous thrombosis.

Heparin cofactor II deficiency — Heparin cofactor II (antithrombin is heparin cofactor I) is a heparin-dependent glycoprotein that acts as a thrombin inhibitor. Several families have been described with quantitative deficiency in this protein inherited as an autosomal dominant trait. Heterozygous individuals have plasma heparin cofactor II concentrations that are approximately 50 percent of normal values.

It is not clear if heparin cofactor II deficiency is a significant risk factor for thrombosis [30,31]. In one series of 305 patients with juvenile thromboembolic episodes, two patients had heparin cofactor II deficiency [31]. However, each of these patients had a second defect: the factor V Leiden mutation and protein C deficiency.

Plasminogen and plasminogen activator inhibitor-1 (PAI-1) — Plasminogen deficiency has not been shown to be a risk factor for venous thrombosis. A 4G polymorphism in the PAI-1 promoter (designated 4G/5G) may be a weak risk factor when combined with other thrombophilic defects; we however recommend against testing for the 4G/5G PAI-1 promoter polymorphism or measuring plasma PAI-1 levels. (See "Plasminogen deficiency" and "Thrombotic and hemorrhagic disorders due to abnormal fibrinolysis".)

Dysfibrinogenemia — Patients with dysfibrinogenemia have structural defects that cause alterations in the conversion of fibrinogen to fibrin. Approximately 300 abnormal fibrinogens have been reported, and more than 80 structural defects have been identified [32,33]. The most common structural defects involve the fibrinopeptides and their cleavage sites, and the second most common involves the gamma-chain polymerization region. Approximately one-half of the mutants are clinically silent, whereas hemorrhage and thrombosis occur in almost equal numbers of symptomatic patients. (See "Disorders of fibrinogen".)

Factor XII deficiency — Factor XII, Hageman factor, is the zymogen of a serine protease that initiates the contact activation reactions and intrinsic blood coagulation in vitro. Severe factor XII deficiency (factor XII activity less than 1 percent of normal) is inherited as an autosomal recessive trait; affected patients have marked prolongation in the activated partial thromboplastin time (aPTT) but do not exhibit a bleeding diathesis [34]. To the contrary, venous thromboembolism and myocardial infarction have been reported in a number of factor XII-deficient patients [35], including John Hageman, the initial patient described with this abnormality [36]. This thrombophilic tendency has been attributed to reduced plasma fibrinolytic activity [37].

The frequency with which severe factor XII deficiency leads to thrombosis is uncertain. One review found that 8 percent had a history of thromboembolism [35]. However, interpretation of this type of data is difficult, since patients with complications are more likely to be reported than those who are asymptomatic.

This led to cross-sectional analyses of thromboembolic events in larger numbers of unselected families with factor XII deficiency. In a study of 14 Swiss families with factor XII deficiency, 2 of 18 homozygous or doubly heterozygous patients had sustained deep venous thrombosis; however, each episode occurred at a time when other predisposing thrombotic risk factors were present [38]. Only one of 45 heterozygotes in these families had a possible history of venous thrombosis. However, other groups have found a more pronounced association with thrombosis [39,40]. Thus, it remains unproven if factor XII deficiency is associated with an increased risk of thrombosis.

Other genetic variants — A number of other genetic variants associated with an increased risk for a first episode of VTE have been found by candidate and genome-wide screens with odds ratios generally <1.5 [41-45]. Further study is required before these data can be useful clinically.

Similarly, adults with supernumerary sex chromosome aneuploidies (ie, more than two sex chromosomes) are at increased risk of VTE compared with patients who have two sex chromosomes (1.3 versus 0.25 percent per person year; 10-year risk difference 8.8 percent) [46]. Further research is needed to understand the clinical implications of this association.

ACQUIRED RISK FACTORS

Overview — Acquired risk factors or predisposing conditions for thrombosis include a prior thrombotic event, recent major surgery, presence of a central venous catheter, trauma, immobilization, malignancy, pregnancy, the use of oral contraceptives or heparin, myeloproliferative disorders, antiphospholipid syndrome (APS), and a number of other major medical illnesses (table 1) [2,47-51]. (See "Prevention of venous thromboembolic disease in acutely ill hospitalized medical adults", section on 'Assess risk'.)

While upper extremity DVT is less common than lower extremity DVT, some of the same risk factors are present [52,53]. As an example, in a study of 512 patients with acute symptomatic upper extremity DVT with or without pulmonary embolism, 38 percent had cancer; 45 percent had central venous catheter-related DVT [52]. (See "Catheter-related upper extremity venous thrombosis in adults".)

Multiple acquired risk factors — Many patients with an episode of VTE have more than one acquired risk factor for thrombosis. This was shown in a population-based study of the incidence of VTE in residents of Worcester, MA during 1999. The six most prevalent pre-existing medical characteristics of patients in this study were [6]:

●More than 48 hours of immobility in the preceding month – 45 percent

●Hospital admission in the past three months – 39 percent

●Surgery in the past three months – 34 percent

●Malignancy in the past three months – 34 percent

●Infection in the past three months – 34 percent

●Current hospitalization – 26 percent

Only 11 percent of the 587 episodes of VTE had none of these six characteristics present, while 36 and 53 percent had 1 to 2 and ≥3 risk factors, respectively.

Triggers of hospitalization for VTE were evaluated in a case-crossover study of 16,781 participants in the Health and Retirement Study, a nationally representative sample of older Americans. Exposures to triggering events during the 90-day period prior to 399 hospitalizations for VTE were compared with exposures occurring during four 90-day comparison periods that did not result in hospitalization for VTE. Results included [54]:

●Infections, especially for those associated with a previous hospital or skilled nursing facility stay, were the most common trigger of hospitalization for VTE, occurring in 52.4 percent of the risk periods before hospitalization, as compared with 28 percent for the four comparison periods.

●Other significant triggers included use of erythropoiesis-stimulating agents, blood transfusion, major surgery, fractures, immobility, and chemotherapy.

●These triggers, when combined, accounted for 70 percent of exposures before VTE hospitalization, as opposed to 35 percent in the comparison periods.

Combined acquired plus inherited risk factors — As noted above, patients with VTE may have multiple inherited defects or multiple acquired risk factors for VTE. In addition, they may also have combinations of both inherited and acquired thrombophilic defects. This was best shown in the MEGA case-control study in 4311 consecutive subjects with a first episode of VTE and 5768 controls, in which the following observations were made [51]:

●Subjects with self-reported major medical illnesses (ie, liver or kidney disease, rheumatoid arthritis, multiple sclerosis, heart failure, hemorrhagic stroke, arterial thrombosis) had increased venous thrombotic risk, with odds ratios in the range of 1.5 to 4.9.

●The combination of a major medical illness, as defined above, plus immobilization increased the odds ratio for development of venous thrombosis to 10.9 (95% CI 4.2-28).

●The combination of a major medical illness along with immobilization plus a thrombophilic defect increased the risk of venous thrombosis even further, with odds ratios of 80, 35, 88, 84, and 53 for increased factor VIII levels, increased factor IX levels, increased von Willebrand factor levels, presence of factor V Leiden, or the presence of non-O blood groups, respectively.

Regardless of whether the patient has a genetic and/or an environmental risk factor for VTE, the presence of a positive family history of VTE has been found to be a strong additional risk factor for VTE [55].

Previous thromboembolism — Previous thrombotic episodes are a major risk factor for recurrent VTE. In an outpatient prospective cohort study, the risk of recurrence after an acute episode of venous thrombosis was 18, 25, and 30 percent at two, five, and eight years, respectively [56]. In a community-based epidemiologic study, a previous history of VTE conferred a relative risk (RR) of 7.9 for VTE recurrence [57].

The magnitude of this risk is highly dependent upon patient-specific factors. Those with an episode of thrombosis in the absence of known risk factors (ie, idiopathic VTE), or in association with permanent risk factors (eg, cancer) have a higher rate of recurrence than those with time-limited, reversible risk factors (eg, recent major surgery, immobilization) [56,58,59]. In one study, for example, the influence of these risk factors on the two-year cumulative incidence of VTE recurrence was as follows [60]:

●Patients with surgery in the six weeks prior to a first VTE – zero percent recurrence; a much lower risk for recurrence in patients with surgery associated VTE (hazard ratio [HR] 0.36) was also noted in another study [56].

●Pregnancy-associated VTE – zero percent recurrence.

●Patients with nonsurgical risk factors (eg, oral contraceptive use, fracture or application of plaster cast, immobilization) – 8.8 percent.

●Patients with an unprovoked VTE in whom there was no identifiable clinical risk factor – 19.4 percent.

The presence or absence of these risk factors (eg, pregnancy, surgery) becomes important in the therapeutic approach to patients with a first or recurrent episode of VTE. (See "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)", section on 'Summary and recommendations'.)

In addition, the site of the first VTE appears to be a predictor of the site and perhaps frequency of future episodes of VTE:

●Patients with a PE during the first episode of VTE are much more likely to have a PE during recurrence. One study followed 436 patients recovering from a first episode of idiopathic VTE [61]. When compared with subjects with DVT and no symptoms of PE, those with symptomatic PE had a significantly higher risk of VTE recurrence (RR 2.2) and a higher risk of developing symptomatic PE at recurrence (RR 4.0). In another report of 165 patients presenting with PE, approximately 60 percent of the recurrences were a second episode of PE [58].

●Patients with a DVT during the first episode of VTE are highly likely to have a DVT during recurrence. This was illustrated in a study of 267 patients with a first episode of idiopathic DVT; 34 of the 42 recurrences were DVTs, with the most common site of recurrence being in the contralateral leg [59].

Malignancy — Patients with cancer often have a hypercoagulable state due to the production of substances with procoagulant activity (eg, tissue factor and cancer procoagulant). Clinical venous thromboembolism occurs in approximately 15 percent of such patients and is a common cause of serious clinical outcomes, including major hemorrhage and death, especially in older patients as well as in those malignancies associated with advanced disease at the time of diagnosis (eg, pancreatic cancer) and myeloproliferative neoplasms (MPNs) [62-67]. The risk of VTE in such patients appears to be highest during the initial hospitalization, time of diagnosis, onset of chemotherapy, as well as at the time of disease progression [68]. MPNs have a higher rate of both venous (10 percent) and arterial thrombosis (3 percent) [67]. (See "Risk and prevention of venous thromboembolism in adults with cancer", section on 'Hospitalized medical patients'.)

The risk of thrombosis may be further increased in patients with malignancy and a central venous catheter, in which situation the prevalence of a venous thrombotic event may be as high as 12 percent [69-72]. The use of certain drugs may also increase the risk of VTE in these patients. (See "Cancer-associated hypercoagulable state: Causes and mechanisms" and "Multiple myeloma: Prevention of venous thromboembolism in patients receiving immunomodulatory drugs (thalidomide, lenalidomide, and pomalidomide)" and "Cancer-associated hypercoagulable state: Causes and mechanisms", section on 'Therapy-related factors'.)

Approximately 20 percent of patients with symptomatic deep venous thrombosis have a known active malignancy [48,73]. In a retrospective study of over 63,000 patients admitted to Danish non-psychiatric hospitals from 1977 through 1992 for a diagnosis of VTE, 18 percent had received a diagnosis of cancer (other than non-melanoma skin cancer) prior to the thromboembolic event [74]. (See "Risk and prevention of venous thromboembolism in adults with cancer", section on 'Incidence and risk factors'.)

The majority of cancers associated with thromboembolic events are clinically evident and have been previously diagnosed at the time of the event. In the Danish study, 78 percent of the cancers were diagnosed before the event [74]. The five most common sites for cancer diagnosed at the time of VTE included:

●Lung – 17 percent

●Pancreas – 10 percent

●Colon and rectum – 8 percent

●Kidney – 8 percent

●Prostate – 7 percent

However, thromboembolism can precede the diagnosis of malignancy [75-77]. In one report, for example, 250 consecutive patients with symptomatic DVT were evaluated, 105 of whom had an identified cause or risk factor for the thrombosis [75]. Malignancy was identified at the time of the thrombotic event in five (3.3 percent) of 153 patients with no other identifiable risk factor. During a two-year follow-up, there was an increased incidence of cancer in the patients with idiopathic thrombosis compared with the 105 patients with secondary thrombosis (8 versus 2 percent). The incidence of cancer was considerably higher (17 percent) among the 35 patients with recurrent idiopathic venous thrombosis. It should be noted that this study was performed prior to the implementation of widespread age-appropriate cancer screening.

Other studies have noted a less prominent association of venous thromboembolism with previously undiagnosed cancer. A Danish study evaluated almost 27,000 patients with VTE; the standardized incidence ratio for cancer was only 1.3 compared with those without DVT or pulmonary embolism [76].

When compared with a group of patients with cancer who did not have VTE, matched in terms of type of cancer, age, sex, and year of diagnosis, the group with VTE tended to have a higher prevalence of distant metastasis (44 versus 35 percent) and a lower one-year survival (12 versus 36 percent). Similar results were obtained in those patients in whom cancer was diagnosed within one year after an episode of VTE.

The role of screening for malignancy in patients with venous thromboembolism is discussed separately. (See "Evaluating adult patients with established venous thromboembolism for acquired and inherited risk factors", section on 'Evaluation for occult malignancy'.)

Surgery — Thrombotic risk is greatly increased during surgery, particularly orthopedic, major vascular, neurosurgery, and cancer surgery [78-81]. Risk factors in this group include older age, previous venous thromboembolism, the coexistence of malignancy or medical illness (eg, cardiac disease), thrombophilia, and longer surgical, anesthesia, and immobilization times [78,80,82-86]. Without prophylaxis, there is a markedly increased risk of both venous thrombosis and pulmonary embolism. (See "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients".)

According to the older 2004 American College of Chest Physicians (ACCP) Anticoagulation Guidelines, patients undergoing surgical procedures were divided into the following risk categories (table 4) [87]:

●Low risk patients are under the age of 40, have none of the risk factors listed above, will require general anesthesia for less than 30 minutes, and are undergoing minor elective, abdominal, or thoracic surgery. Without prophylaxis their risk of proximal vein thrombosis is less than 1.0 percent, and risk of fatal pulmonary embolism is less than 0.01 percent.

●Moderate risk patients are over the age of 40, will require general anesthesia for more than 30 minutes, and have one or more of the above risk factors. Without prophylaxis, their risk of proximal vein thrombosis is 2 to 10 percent, and their risk of fatal pulmonary embolism is 0.1 to 0.7 percent.

●High risk patients include those over the age of 40 who are having surgery for malignancy or an orthopedic procedure of the lower extremity lasting more than 30 minutes, and those who have an inhibitor deficiency state or other risk factors. The risk of proximal vein thrombosis and fatal pulmonary embolism in this group is 10 to 20 percent and 1.0 to 5.0 percent, respectively.

These risk groups were redefined in the 2012 ACCP Guidelines. (See "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients", section on 'Assess risk for thrombosis'.)

Thromboprophylaxis significantly reduces the incidence of symptomatic DVT or pulmonary embolism in the immediate postoperative period. However, there is a continued risk of DVT following hospital discharge in those who have had total knee or hip arthroplasty after the usual 7 to 10 days of thromboprophylaxis [88-91].

In patients so treated, who did not develop a symptomatic venous thromboembolism during hospitalization, the incidence of symptomatic non-fatal venous thromboembolism (VTE) or fatal pulmonary embolism in the subsequent three months has been estimated at 2.2 percent (95% CI 1.4-3.0) and 0.05 percent (95% CI 0-0.12), respectively [92]. The estimated 30-day risk of symptomatic non-fatal VTE was higher for those undergoing total hip replacement than total knee replacement (2.5 versus 1.4 percent), in agreement with other studies [82,93].

Trauma

Major trauma — The risk of thrombosis is increased in all forms of major injury [94-97]. In one study of 716 patients admitted to a regional trauma unit, DVT in the lower extremities was found in 58 percent of patients with adequate venographic studies; 18 percent had proximal vein thrombosis [94]. Thrombi were detected in 54 percent of patients with major head injuries, 61 percent of patients with pelvic fracture, 77 percent of patients with tibial fracture, and 80 percent of those with femoral fracture.

The mechanisms of activation of the coagulation system following surgery or trauma are incompletely understood, but may include decreased venous blood flow in the lower extremities, diminished fibrinolysis, immobilization (see 'Immobilization' below), the release or exposure of tissue factor, and depletion of endogenous anticoagulants such as antithrombin [98]. In addition, the femoral vein in the operated leg may kink after hip replacement surgery, thereby increasing the risk of proximal venous thrombosis in the absence of calf vein thrombosis [99].

Other VTE risk factors in trauma patients may include spinal cord injury, lower extremity or pelvic fracture, need for a surgical procedure, insertion of a femoral venous line or repair of a major vein, increasing age, delayed initiation of thromboprophylaxis, or prolonged immobility either in the hospital or entrapment at the scene of the trauma [94,100-103]. (See "Overview of inpatient management of the adult trauma patient", section on 'Thromboprophylaxis' and "Venous thromboembolism risk and prevention in the severely injured trauma patient", section on 'Thromboprophylaxis'.)

Minor injuries — A large population-based study investigated the VTE risk following a minor injury (ie, one not requiring surgery, a plaster cast, hospitalization, or extended bed rest at home for at least four days) [104]. A minor injury occurring in the preceding three to four weeks was associated with a three- to fivefold increase in DVT risk. In carriers of factor V Leiden, this risk was increased 50-fold.

Intravenous drug use — Direct trauma, irritation, and infection may be responsible for the high prevalence of DVT noted in young drug users who inject these agents directly into their femoral veins [105-107].

Pregnancy — Pregnancy is associated with an increased risk of thrombosis that may be due in part to obstruction of venous return by the enlarged uterus, as well as the hypercoagulable state associated with pregnancy. Estimates of the age-adjusted incidence of VTE range from 5 to 50 times higher in pregnant versus non-pregnant women. This subject is discussed in depth separately. (See "Deep vein thrombosis in pregnancy: Epidemiology, pathogenesis, and diagnosis" and "Use of anticoagulants during pregnancy and postpartum".)

Inherited thrombophilia — The risk of thrombosis is further accentuated in those women who have inherited thrombophilia, as illustrated by the following observations [108-111]:

●In one series of 60 women with an inherited deficiency of a naturally occurring anticoagulant (antithrombin, protein C, or protein S), the risk of venous thrombosis during pregnancy or the postpartum period was increased eightfold (4.1 versus 0.5 percent in nondeficient women) [109].

●The thrombotic risk for a woman with factor V Leiden during pregnancy or the puerperium has been estimated at approximately 1 in 400 to 500 compared with 1 in 1400 in the general population [108]. (See "Factor V Leiden and activated protein C resistance", section on 'Fetal loss and obstetric complications'.)

●A case-control study looked for factor V Leiden and the prothrombin gene mutation in 190 women with a first history of venous thromboembolism during pregnancy or the postpartum period and 190 controls without a history of thrombosis [111]. Compared with the controls, the women with a positive history had a higher prevalence of factor V Leiden (30 versus 6 percent), the prothrombin gene mutation (9 versus 2 percent), or both (7 versus zero percent). Women with both mutations had a higher probability of thrombosis during pregnancy than those with only the factor V Leiden or only the prothrombin mutation (5.2 versus 0.3 and 0.4 percent, respectively).

The prevalence of selected thrombophilias in women with no history of thrombosis and the probability of thromboembolism during pregnancy in these carriers have been described. The risk of thromboembolism during pregnancy is also higher among women with inherited thrombophilia who have a past history or family history of thromboembolic events. These data have been derived from essentially White populations; there is a paucity of data on the prevalence and clinical significance of these states in individuals other than White patients [112]. (See "Inherited thrombophilias in pregnancy" and "Deep vein thrombosis and pulmonary embolism in pregnancy: Prevention".)

Drugs — A number of drugs have been associated with an increased risk of venous thrombosis, while statin use may be associated with a reduced risk of recurrence after a first episode [113]

Oral and transdermal contraceptives — Because of their widespread use, oral contraceptives (OCs) are the most important cause of thrombosis in young women [114-116]. The risk of thrombosis increases within the first 6 to 12 months of the initiation of therapy [117] and is unaffected by duration of use; the risk is considered by most experts to return to previous levels within one to three months of cessation. An increased risk for VTE has also been found in women using contraceptive transdermal patches (Xulane). The use of OCs in older women to alleviate menopausal symptoms is also associated with a substantially increased risk of VTE, especially in those with inherited thrombophilia or a family history of VTE [118].

This subject is discussed in detail separately. (See "Combined estrogen-progestin contraception: Side effects and health concerns", section on 'Venous thromboembolism' and "Contraception: Transdermal contraceptive patches", section on 'Risk of venous thrombotic events' and "Transgender women: Evaluation and management", section on 'Estrogen therapy'.)

Hormone replacement therapy — Observational studies, the HERS trials, and a meta-analysis published before the Women's Health Initiative (WHI), evaluated the association between hormone replacement therapy (HRT) and venous thromboembolism (VTE), and suggested that HRT caused an approximately twofold increase in VTE risk, which appeared to be greatest in the first year of treatment [119-122]. The most definitive data come from the WHI and are consistent with previous estimates [123].

The risk for VTE following the use of HRT appears to be further increased in older women, as well as in those with obesity, have an underlying thrombophilia (eg, factor V Leiden, prothrombin gene mutation), or a past history of venous thromboembolism [118,123-127].

This subject is discussed in detail separately. (See "Menopausal hormone therapy and cardiovascular risk", section on 'Venous thromboembolism'.)

Testosterone — Postmarketing reports of VTE in men taking testosterone has led to the US Food and Drug Administration (FDA) requiring manufacturers to place a general warning in the labeling of all approved testosterone products regarding this risk [128]. (See "Testosterone treatment of male hypogonadism", section on 'Venous thromboembolism'.)

Tamoxifen — A number of studies, including the large Breast Cancer Prevention Trials, have demonstrated that tamoxifen use is associated with an increased rate of venous thromboembolic events and that there is a significant additional procoagulant effect when tamoxifen is added to chemotherapy [129,130]. As an example, the role of tamoxifen in thromboembolic events was assessed in seven Eastern Cooperative Oncology Group (ECOG) trials, three of which randomly allocated patients to receive or not receive tamoxifen as part of their design [131]. Each trial with a tamoxifen versus no tamoxifen randomization showed a trend toward more venous events in the tamoxifen-containing arm. Using data from all seven studies, premenopausal, but not postmenopausal, women who received chemotherapy and tamoxifen developed more venous thrombosis than those who received chemotherapy alone (3 versus 1 percent). (See "Cancer-associated hypercoagulable state: Causes and mechanisms", section on 'Therapy-related factors'.)

Bevacizumab — Use of the monoclonal antibody bevacizumab has been associated with an increased risk for both arterial and venous thromboembolic events. (See "Toxicity of molecularly targeted antiangiogenic agents: Cardiovascular effects", section on 'Arterial and venous thromboembolism'.)

Glucocorticoids — Glucocorticoid administration is associated with an increased risk of VTE [50,132]. A population-based case-control study compared rates of glucocorticoid use in nearly 39,000 cases of VTE to rates in age- and sex-matched controls without VTE [132]. Compared to former use of glucocorticoids (>3 months prior to VTE), recent use (<3 months) was associated with a 1.2- to 2-fold increased risk of VTE. The risk was highest in first-time users (incidence rate ratio, 3.06; 95% 2.77-3.38) and increased with cumulative doses (2-fold for a prednisolone-equivalent dose of 1 to 2 g compared to <10 mg). This association appeared to be independent of confounders such as severity of the underlying disease and the presence of conditions known to have an increased risk of thrombosis (eg, rheumatoid arthritis).

Antidepressants — One database of nearly 800,000 women in the UK and Scotland reported a higher risk of VTE in women taking antidepressants compared with women who reported neither depression nor antidepressant use (HR 3.9) [133].

Others — In a retrospective matched propensity study of 21,931 patients, tranexamic acid, a medication frequently transfused for bleeding during trauma, was associated with a greater than three-fold increase in the risk of VTE, after adjusting for prespecified confounders [134]. (See "Initial management of moderate to severe hemorrhage in the adult trauma patient".)

Immobilization — Venous stasis associated with bed rest or prolonged immobilization (eg, heart failure, stroke, myocardial infarction, leg injury) is an important risk factor for venous thrombosis [135,136]. (See "Prevention and treatment of venous thromboembolism in patients with acute stroke".)

In a trial in patients who required immobilization in a cast after leg injury, DVT on venography occurred in 19 percent of patients treated with placebo compared with 9 percent of those treated with low molecular weight heparin [136]. Most of the thromboses were distal; two patients with proximal DVT had a pulmonary embolism, both in the placebo group. A subsequent meta-analysis of the incidence of asymptomatic DVT in patients with immobilization of the lower extremities found similar results when comparing the prophylactic use of low molecular weight heparin versus placebo or no treatment (17.1 versus 9.6 percent; RR 0.58, 95% CI 0.39-0.86) [137,138].

Prolonged sitting — VTE has also been associated with prolonged sitting, such as might occur during attendance at a theater, prolonged immobility or crouching (eg, carpet fitters) at work, or sitting at a computer for long periods [139-143]. The latter condition has been called "seated immobility thromboembolism" or "e-thrombosis."

Extended travel — Prolonged travel, especially by air, appears to confer a small increase in risk of VTE, a phenomenon that has been termed the "economy class syndrome" (figure 1). This subject, including possible preventive measures, is discussed in detail separately.

Antiphospholipid antibodies — The antiphospholipid syndrome (APS) is characterized by the presence of antibodies directed against plasma proteins bound to anionic phospholipids. Patients may present with venous or arterial thrombosis, recurrent fetal loss, and/or thrombocytopenia. The disorder may be primary or associated with systemic lupus erythematosus and other rheumatic diseases. In one large series, elevated levels of aPL antibodies were present in 4.1 percent of 2132 consecutive patients presenting with deep vein thrombosis [16]. (See "Clinical manifestations of antiphospholipid syndrome", section on 'Thrombotic events'.)

Renal diseases

Chronic renal disease — Autopsy series have suggested that VTE events are relatively common in patients with end stage renal disease, and epidemiologic studies have reported an increased risk for VTE in those receiving dialysis [144-147]. As an example, using the Healthcare Cost and Utilization Project's Nationwide Inpatient sample, the annual frequency of pulmonary embolism in adults with end-stage kidney disease, chronic kidney disease, and normal renal function was 527, 204, and 66 per 100,000 persons [148].

The LITE study of 19,071 middle-aged and older non-dialysis, non-renal transplant patients demonstrated a relative VTE risk of 1.7 (95% CI 1.2-2.5) for patients with stage 3/4 chronic renal disease (ie, estimated glomerular filtration rate (eGFR) 15 to 59 mL/min per 1.73 m²), compared with patients with normal renal function (ie, eGFR >90 mL/min per 1.73 m²) [147]. Data from the PREVEND study have shown that the major risk factor for VTE in subjects with stages 1 to 3 chronic renal disease is the presence of albuminuria (urinary albumin ≥30 mg/24 hours) [149]. As an example, the adjusted HR for those with an eGFR of 30 to 60 mL/min and albuminuria was 4.1 (95% CI 1.5-11) when compared with those having an eGFR >90 mL/min and no albuminuria.

Of importance, pooled individual data from the HUNT2, PREVEND, and Tromsø studies have shown that both the estimated glomerular filtration rate and the urinary albumin-creatinine ratio are independently associated with an increased risk for VTE in the general population, even across the normal ranges for these two tests [150]. The reason for increased risk is unknown but may relate to elevated levels of factor VIII and von Willebrand factor in this population [151].

Nephrotic syndrome — Patients with the nephrotic syndrome have an increased prevalence (10 to 40 percent) of both arterial and venous thromboemboli, particularly deep vein and renal vein thrombosis. (See "Hypercoagulability in nephrotic syndrome", section on 'Pathogenesis' and "Hypercoagulability in nephrotic syndrome", section on 'Epidemiology' and "Venous thrombosis and thromboembolism (VTE) in children: Risk factors, clinical manifestations, and diagnosis", section on 'Other venous thrombosis'.)

The mechanisms responsible for the hypercoagulability in the nephrotic syndrome are not understood [152,153]. Alterations in many blood coagulation parameters have been described, but most are inconsistently reported, of relatively minor magnitude, and correlate poorly with thrombotic events [154]. Among the natural anticoagulant proteins, plasma antithrombin levels are often reduced due, at least in part, to increased urinary excretion [155,156]. The antigenic levels of protein C and protein S are generally increased [157,158] but functionally active protein S may be reduced [157]. Platelet hyperreactivity or increased whole blood viscosity may contribute to the thrombotic diathesis [159,160].

Renal transplantation — The reported prevalence of VTE following renal transplantation (RT) is increased, in the range of 5 to 8 percent, with pulmonary embolism being a common cause of death [161-164]. In one study of 484 RT patients, 34 (7 percent) developed a first VTE at a median time of 6.5 months post-RT (range 1 to 81 months) [161]. Of importance, the incidence of recurrent VTE following cessation of oral anticoagulants (OAC) after a median treatment time of six months (range 3 to 43 months) was 50 percent (14 of 28). This recurrence risk was 10 times that of an age- and sex-matched group of patients with normal renal function who were treated with OAC for a similar period of time for a first episode of idiopathic VTE. RT patients may need treatment with OAC for longer period of time, especially if a hypercoagulable state has been identified [164].

Liver disease — It has often been stated that patients with chronic liver disease (CLD) and an elevated INR are "auto-anticoagulated" and are therefore not at risk for VTE. This notion is unfounded, as shown in a retrospective cohort study of 190 hospitalized patients with CLD, 12 of whom (6.3 percent) developed VTE [165]. This subject is discussed in depth separately. (See "Hemostatic abnormalities in patients with liver disease".)

Cardiovascular diseases — The Longitudinal Investigation of Thromboembolism Etiology, which combined information from two prospective cohort studies, the Atherosclerosis Risk in Communities (ARIC) and the Cardiovascular Health Study (CHS) investigated the relationship between risk factors for arterial disease and occurrence of a first episode of VTE in 19,293 subjects, with over 148,000 person-years of follow-up [166,167]. (See "Overview of established risk factors for cardiovascular disease".)

The results can be summarized as follows:

●Factors associated with an increased risk of a first episode of VTE were obesity (HR 2.7 for a body mass index >40), increased age, male sex, diabetes, and being from a Black population. (See 'Obesity' below and 'Age' below.)

●Hypertension, dyslipidemia, physical inactivity, smoking, and alcohol consumption were not associated with an increased risk of VTE. Specifically, there was no trend of VTE hazard across quartiles of high-density lipoprotein cholesterol or apolipoprotein A-I [168]. The lack of an association between both apolipoproteins and the classical lipoproteins and VTE risk was also noted in the population-based PREVEND cohort study [169]. (See 'Smoking' below.)

●Increased baseline levels of D-dimer were strongly and positively related to the future occurrence of venous thrombosis [170]. (See "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity", section on 'D-dimer'.)

A meta-analysis of cardiovascular risk factors in patients with VTE included 21 case-control and cohort studies, and came to slightly different conclusions. When compared with control subjects, VTE risks were [171]:

●Obesity 2.3 (95% CI 1.7-3.2)

●Hypertension 1.5 (95% CI 1.2-1.8)

●Diabetes mellitus 1.4 (95% CI 1.1-1.8)

●Smoking 1.2 (95% CI 0.95-1.5)

●Hypercholesterolemia 1.2 (95% CI 0.67-2.0)

In a prospective, population-based study of over 27,000 inhabitants of Tromsø in Norway, independent risk factors for VTE included increased age, male sex, obesity, a positive family history of myocardial infarction, and atrial fibrillation [172,173]. In a cohort study of 40,000 women taken from the South Sweden population registry and followed over a 10-year period, women non-smokers who were physically active and who consumed alcohol in moderation were at lower risk of developing VTE [174].

VTE and atherosclerotic disease — In a population-based case control study of 5824 patients with VTE and 58,240 controls, an arterial cardiovascular event within 3 months conferred a short-term increased risk of VTE of 4.22 (95% CI 2.33-7.64) following myocardial infarction and 4.41 (95% CI 2.92-6.65) after stroke [175]. An analysis of the Nationwide Inpatient Sample database reported that among hospitalizations for patients with ST elevation myocardial infarction (STEMI), 1 percent were complicated by VTE (ie, 10 per 1000 admissions) [176]. Patients with STEMI who experienced VTE were older and had a higher proportion of African Americans and women than patients who had a STEMI and did not have VTE.

An association between atherosclerotic disease and spontaneous venous thrombosis was also detected in a study of unselected patients with DVT who did not have symptomatic atherosclerosis [177]. The odds ratio for detecting at least one carotid artery plaque by ultrasonography in 153 subjects with spontaneous DVT was 1.8 (95% CI 1.1-2.9) when compared with 150 controls. The strength of this association did not change in a multivariate analysis including risk factors for atherosclerosis. The prevalence of carotid lesions in 146 patients with secondary DVT (eg, active cancer, recent trauma or surgery, prolonged immobilization) did not differ significantly from that in controls.

A number of studies and a meta-analysis have suggested that subjects with VTE are at increased risk for acute cardiovascular events [178-183]. The following two cohort studies are instructive in this regard.

In a 20-year population-based Danish cohort study, the risk of subsequent myocardial infarction (MI) and stroke was assessed in subjects with VTE and no history of cardiovascular disease (eg, no hypertension, stroke, heart failure, angina, or MI) [178]. The following results were obtained:

●For the 25,199 patients with DVT, relative risks were 1.6 (95% CI 1.4-1.9) for MI and 2.2 (95% CI 1.8-2.6) for stroke in the first year after the thrombotic event.

●Relative risks were higher for the 16,925 patients with PE, being 2.6 (95% CI 2.1-3.1) and 2.9 (95% CI 2.3-3.7), respectively.

●Relative risks for arterial cardiovascular events were raised 20 to 40 percent over those of the 163,566 matched controls during the subsequent 20 years of follow-up, and were similar for both provoked and unprovoked VTE.

In a population-based cohort study of all inhabitants of Groningen, the Netherlands aged 28 to 75 years, and with a median follow-up of 10.7 years, the annual incidence of arterial thromboembolism after an episode of VTE was 2.03 percent, compared with 0.87 percent in those without VTE (adjusted HR 1.40, 95% CI 1.04-1.88) [181]. The risk of arterial thromboembolism was highest in the first year after VTE (3.0 percent) and after an unprovoked event (2.53 percent).

Why this relationship exists is not clear, although VTE and arterial cardiovascular events share similar risk factors, such as obesity, hypertension, smoking, and diabetes/hyperglycemia [2,171,184-188]. (See "Overview of established risk factors for cardiovascular disease".)

Heart failure — Heart failure appears to be a hypercoagulable state [189] that can result in intracardiac thrombi and DVT. The major risk factors for intracardiac thrombi are reduced left ventricular function and atrial fibrillation. (See "Antithrombotic therapy in patients with heart failure".)

The risk of DVT may be greatest in patients with right heart failure (eg, those with peripheral edema). Thromboprophylaxis during immobilization or hospital admission in such patients may be advisable [189,190], especially since heart failure and immobility have been shown to be potent risk factors for in-hospital death as well as death within 30 days in patients with VTE [191]. One study reported a long-term risk of VTE in patients who were hospitalized for heart failure (adjusted HR 3.13) [192].

Cardiovascular risk factors

Obesity — A number of studies have found a significantly increased risk for deep vein thrombosis, and/or pulmonary embolism in subjects with obesity [57,171,187,193-199], and a reduced risk for underweight subjects [197], as well as an increased risk of recurrent VTE once anticoagulation treatment has been withdrawn [200].

●The Atherosclerosis Risk in Communities (ARIC) and the Cardiovascular Health Study (CHS) found an increased risk of a first episode of VTE with obesity (HR 2.7 for a body mass index >40) [166].

●Analysis of data from the National Hospital Discharge Survey indicated an increased relative risk for DVT (RR 2.50, 95% CI 2.49-2.51) and PE (RR 2.21, 95% CI 2.20-2.23) in patients with obesity, with the greatest impact being seen in males and females below the age of 40 [201].

Obesity also appears to be a contributing factor for further increasing the risk of VTE in a number of high-risk settings [196]:

●Along with smoking and age in patients with factor V Leiden or the prothrombin gene mutation [202]. (See "Factor V Leiden and activated protein C resistance".)

●In passengers with obesity undertaking long-duration air travel. (See "Assessment of adult patients for air travel", section on 'Venous thromboembolism'.)

●In females with obesity taking oral contraceptives. In one study, females with obesity who used oral contraceptives had a 24-fold higher thrombotic risk than women with a normal body mass index who did not use these agents [196]. (See "Combined estrogen-progestin contraception: Side effects and health concerns", section on 'Obesity'.)

●The increased risk of VTE in individuals with a high BMI is partly mediated by factor VIII-related APC resistance. This risk is more pronounced when other causes of APC resistance are also present (eg, factor V Leiden, blood group non-O carriers) [203]. (See "Factor V Leiden and activated protein C resistance", section on 'Venous thromboembolism'.)

Smoking — Although an initial report from the Leiden group [204], the ARIC and CHS studies [166], a Norwegian study [172], and a meta-analysis [171] found no significant relationship between smoking and VTE, at least six other studies, including an update from the ARIC study, have detected a relationship between the two, with relative risks ranging from 1.3 to 3.3 [187,199,202,205-212].

●In one study, a high number of pack-years resulted in the highest risk of venous thrombosis among young current smokers (OR for ≥20 pack-years of 4.3; 95% CI 2.6-7.1), when compared with young nonsmokers [207]. In addition, women who were current smokers and used oral contraceptives had an 8.8-fold higher risk (95% CI 5.7-13) than nonsmoking women who did not use oral contraceptives.

●In a Danish case-cohort study, the adjusted HR for VTE emerging from the combined exposure to the non-O blood type and heavy smoking was 2.98 (95% CI 1.89-4.69), which exceeded the sum of the individual effects [213].

Age — Many risk factors for venous thrombosis, such as immobilization and malignancy, also correlate with age, although few studies adequately address the confounding variables [214-217]. Two studies worthy of mentioning include:

●The Worcester DVT study, based on previously hospitalized patients, found that annual incidence rates for DVT increased from 17 per 100,000 persons/year for those between the ages of 40 and 49 to 232 per 100,000 persons/year for those between the ages of 70 and 79 [218].

●The ARIC and CHS studies found an increasing incidence of first VTE with age, with a HR of 1.7 (95% CI 1.5-2) for every decade of life after age 55 [166].

Air pollution — Multiple observational studies have demonstrated an association between fine particulate air pollution and distance from a major urban road or freeway and cardiovascular and cardiopulmonary mortality. However, there is conflicting evidence concerning whether air pollution is [219-221], or is not [222,223], causally related to VTE development. (See "Overview of possible risk factors for cardiovascular disease", section on 'Air pollution'.)

Microalbuminuria — An ongoing community-based prospective cohort study (the PREVEND cohort) evaluated the association between microalbuminuria, a known cardiovascular risk factor, and the risk of developing VTE. (See "Moderately increased albuminuria (microalbuminuria) and cardiovascular disease".)

Results included [224]:

●The annual incidences of VTE were 0.12, 0.20, 0.40, and 0.56 percent for those with urinary albumin excretion (UAE) of <15, 15 to 29, 30 to 300, and >300 mg/24 hours, respectively.

●When adjusted for age, cancer, use of oral contraceptives, and atherosclerosis risk factors, HRs associated with UAE of 15 to 29, 30 to 300, and >300 mg/24 hours were 1.40 (95% CI 0.86-2.35), 2.20 (95% CI 1.44-3.36) and 2.82 (95% CI 1.21-6.61) respectively, when compared with those having UAE <15 mg/24 hours.

Hematologic risk factors

Heparin-induced thrombocytopenia — The major clinical problem associated with heparin-induced thrombocytopenia (HIT) is thrombosis, both venous and arterial (ie, heparin-induced thrombocytopenia with thrombosis). Thus, if a patient receiving heparin develops an arterial or venous thromboembolic event, the presence of thrombocytopenia suggests that this is due to HIT rather than failure of anticoagulation. (See "Management of heparin-induced thrombocytopenia".)

The frequency with which initiation of heparin therapy for thromboprophylaxis or treatment was associated with new or recurrent VTE and HIT was explored in a comprehensive literature search. The following observations were made [225]:

●In 3792 patients treated with intravenous or subcutaneous unfractionated heparin (UFH), there were 242 episodes of VTE (6.4 percent). Thirteen percent of the episodes of VTE (0.8 percent of the patients treated with UFH) were associated with serologically-confirmed HIT.

●In 2427 patients treated with low molecular weight (LMW) heparin, there were 144 episodes of VTE (5.9 percent). Only one of the 144 episodes (0.04 percent of patients treated with LMW heparin) was HIT-associated.

In this study, approximately one in every eight episodes of VTE occurring in UFH-treated patients was due to HIT. Accordingly, clinicians should suspect the possibility of HIT if VTE, along with thrombocytopenia, develops soon after the use of UFH [226]. (See "Management of heparin-induced thrombocytopenia".)

Very rarely, patients can develop a HIT-like syndrome in the absence of heparin exposure, referred to as autoimmune HIT or spontaneous HIT. Coronavirus disease 2019 (COVID-19) vaccine-induced immune thrombotic thrombocytopenia is one-such syndrome. (See "COVID-19: Vaccine-induced immune thrombotic thrombocytopenia (VITT)".)

Hyperviscosity — Thrombosis can be a manifestation of diseases associated with hyperviscosity secondary to increased plasma viscosity, an increased number of red or white blood cells, or decreased deformability of cells.

●Increased plasma viscosity can result from hyperfibrinogenemia or hypergammaglobulinemia. Hypergammaglobulinemia associated with the hyperviscosity syndrome is most commonly encountered in patients with Waldenstrom macroglobulinemia or multiple myeloma. Thrombosis in hypergammaglobulinemic states is attributable to abnormal rheology (ie, increased serum viscosity). Presenting symptoms of the hyperviscosity syndrome include bleeding due to platelet dysfunction, visual disturbances, and neurologic defects. (See "Epidemiology, pathogenesis, clinical manifestations, and diagnosis of Waldenström macroglobulinemia", section on 'Hyperviscosity syndrome'.)

●Increased whole blood viscosity plays an important role in the pathogenesis of thrombosis in polycythemia vera, a major complication of this disorder. Acquired qualitative platelet defects have also been implicated in the hypercoagulable state. Common thrombotic complications include cerebrovascular accidents, myocardial infarction, peripheral arterial occlusion, deep venous thrombosis, pulmonary embolism, and portal and hepatic vein thrombosis (Budd-Chiari syndrome). (See "Etiology of the Budd-Chiari syndrome".)

●Increased whole blood viscosity also occurs in occasional patients with myeloid and monocytic leukemias who have markedly elevated white blood counts (generally >100,000/microL). Small vessels in the lungs, brain, and less commonly other organs may be obstructed by high levels of immature leukocytes. (See "Hyperleukocytosis and leukostasis in hematologic malignancies".)

●Increased whole blood viscosity in sickle cell disease is due primarily to decreased deformability of sickled erythrocytes, and may contribute to the occlusion of small blood vessels. Other factors that may play a role in this complication include enhanced adhesion of sickle erythrocytes to vascular endothelium and increased coagulation and platelet activation [227-229].

Myeloproliferative neoplasms and PNH — The chronic myeloproliferative neoplasms, particularly polycythemia vera and essential thrombocythemia, are characterized by thrombotic complications. These complications include both arterial and venous thrombosis and microcirculatory disorders such as erythromelalgia and visual and neurologic symptoms [230]. Hyperviscosity may play a contributing role. (See 'Hyperviscosity' above and "Prognosis and treatment of polycythemia vera and secondary polycythemia".)

A high percent of patients with idiopathic hepatic (eg, Budd-Chiari syndrome) or portal vein thrombosis, but not those with idiopathic lower extremity DVT [231-234], have had in vitro evidence (eg, spontaneous erythroid colony growth in the absence of erythropoietin, presence of the JAK2 mutation, clonal karyotypic abnormalities on bone marrow examination) suggestive of an occult myeloproliferative neoplasm [235-238]. (See "Etiology of the Budd-Chiari syndrome", section on 'Myeloproliferative disorders'.)

Paroxysmal nocturnal hemoglobinuria (PNH) is a rare acquired clonal disorder of bone marrow stem cells. Affected patients generally have chronic intravascular hemolysis with episodes of gross hemoglobinuria accompanied by leukopenia and thrombocytopenia. PNH is also associated with an approximately 40 percent prevalence (in the United States and Europe) of venous thrombosis in the intraabdominal venous network (mesenteric, hepatic, portal, splenic, and renal veins) and cerebral vessels, as opposed to deep vein thrombosis or pulmonary embolism [239]. (See "Clinical manifestations and diagnosis of paroxysmal nocturnal hemoglobinuria".)

The mechanism for the increased tendency to thrombosis in PNH is not completely understood. (See "Pathogenesis of paroxysmal nocturnal hemoglobinuria" and "Clinical manifestations and diagnosis of paroxysmal nocturnal hemoglobinuria".)

Gastrointestinal

Inflammatory bowel disease — Venous thromboembolism is a known complication of inflammatory bowel disease (IBD, Crohn's disease, ulcerative colitis). In one study, for example, the absolute risks for development of VTE were, as follows [240]:

●Hospitalized patients with IBD flare – 37.5/1000 person-years compared to 13.9/1000 person-years for controls (HR 3.2, 95% CI 1.7-6.3)

●Ambulatory patients with IBD flare – 9.0/1000 person-years compared to 0.6/100 person-years for controls (HR 8.4, 95% CI 5.5-12.8).

This subject is discussed further separately. (See "Clinical manifestations, diagnosis, and prognosis of ulcerative colitis in adults", section on 'Extraintestinal manifestations' and "Clinical manifestations, diagnosis, and prognosis of Crohn disease in adults".)

Seasonal variation — Hospital admission rates for both pulmonary embolism and deep venous thrombosis are highest during the winter months and lowest during the summer months [241,242]. This was demonstrated in a retrospective study of records linking discharge data and diagnostic test results complied over a four-year period (figure 2) [241]. Possible explanations for this observation include vasoconstriction due to reduced activity or cold weather, superimposed infection, or seasonal variation in thrombogenic factors, including fibrinogen and factor VII activity [243] and active sun exposure [242].

MISCELLANEOUS

COVID-19 — Hospitalized patients with COVID-19 have an increased risk of developing VTE, the details of which are discussed separately. (See "COVID-19: Hypercoagulability".)

In addition, VTE that occurs in association with COVID-19 vaccination is also discussed separately. (See "COVID-19: Vaccine-induced immune thrombotic thrombocytopenia (VITT)" and "COVID-19: Vaccines", section on 'Thrombosis with thrombocytopenia'.)

Hyperhomocysteinemia — Homocystinuria is an inborn error of metabolism that is associated with very high levels of homocysteine and arterial and venous thrombosis. However, mild hyperhomocysteinemia is a clinically distinct disorder. In this section, we discuss the risk of VTE with respect to mild hyperhomocysteinemia.

Hyperhomocysteinemia may be both a genetic and acquired abnormality. The most common genetic defect is homozygosity for a thermolabile mutant of the enzyme methylenetetrahydrofolate reductase (MTHFR). Plasma homocysteine concentrations can also be increased by deficiencies in vitamin B6, B12, or folic acid.

While older studies reported a two-fold increased risk for a first episode of VTE, studies since then have shown that the thrombotic risk associated with mild hyperhomocysteinemia is lower than previously thought. A large epidemiologic study reported no increased risk when confounding variables were included in the analysis [244]; another retrospective study reported a relative risk of only 1.6 [245]. Furthermore, reducing levels of homocysteine with B vitamin supplementation has not resulted in a reduction in the incidence of recurrent VTE [246]. Thus, we recommend against measuring homocysteine levels and never test for MTHFR mutations in patients with VTE. (See "Overview of homocysteine".)

Others — Several other conditions have been reported to have an association with an increased risk of VTE including the following [69,70,247-262]:

●Polycystic ovary syndrome (see "Clinical manifestations of polycystic ovary syndrome in adults", section on 'Venous thromboembolism')

●Central venous lines and peripherally inserted central catheters (see "Central venous catheters: Overview of complications and prevention in adults" and "Central venous catheters: Overview of complications and prevention in adults", section on 'Catheter malfunction')

●Rheumatoid arthritis (see "Overview of pleuropulmonary diseases associated with rheumatoid arthritis", section on 'Venous thromboembolic disease')

●Ovarian hyperstimulation syndrome (see "Pathogenesis, clinical manifestations, and diagnosis of ovarian hyperstimulation syndrome", section on 'Clinical manifestations')

●Active tuberculosis (see "Clinical manifestations and complications of pulmonary tuberculosis", section on 'Venous thromboembolism')

●Asthma (the contribution of glucocorticoid use is unknown)

●Sepsis (see "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients")

●Chronic psoriasis (see "Psoriasis: Epidemiology, clinical manifestations, and diagnosis")

●Superficial vein thrombosis (See "Superficial vein thrombosis and phlebitis of the lower extremity veins".)

●Obstructive sleep apnea

●Klinefelter Syndrome (see "Clinical features, diagnosis, and management of Klinefelter syndrome")

●Antineutrophil cytoplasmic antibodies-associated vasculitis (see "Granulomatosis with polyangiitis and microscopic polyangiitis: Respiratory tract involvement")

●Anemia (in acutely ill hospitalized patients) (see "Diagnostic approach to anemia in adults")

The higher rate of VTE in African Americans compared with White Americans is thought to be mostly explained by the higher prevalence of risk factors in African Americans [263].

ANATOMIC RISK FACTORS FOR DEEP VENOUS THROMBOSIS — The following section will briefly review several anatomic risk factors for the development for deep venous thrombosis (DVT) in the upper or lower extremities; these are not known to be inherited.

Varicose veins — In a Taiwanese claims database study, compared with propensity-matched control patients without varicose veins, patients with varicose veins had an increased rate of DVT (6.6 versus 1.2 per 1000 person-years; HR 5.30) and pulmonary embolism (PE) (0.48 versus 0.28 per 1000 person-years; hazard ratio [HR] 1.73) [264]. Further studies are required to confirm this association and ensure that it is not due to confounding variables such as smoking or obesity.

Paget-Schroetter syndrome — Paget-Schroetter syndrome, also referred to as spontaneous upper extremity venous thrombosis, is usually due to an underlying compressive anomaly at the thoracic outlet. This is frequently due to compression of the vein either between the first rib and a hypertrophied scalene or subclavius tendon or between these tendons themselves. Compression between the clavicle and a cervical rib as well as partial occlusion of the vein by a congenital web have also been reported. This subject is discussed separately. (See "Primary (spontaneous) upper extremity deep vein thrombosis" and "Brachial plexus syndromes", section on 'Thoracic outlet syndrome'.)

May-Thurner syndrome — Hemodynamically significant compression of the left common iliac vein between the overlying right common iliac artery and the underlying vertebral body (May-Thurner syndrome, iliac vein compression syndrome) is a common anatomic pattern in normal subjects, which has been associated with unprovoked left iliofemoral DVT or chronic venous insufficiency [265,266]. However, visualization of a clot this high in the pelvis may be difficult to detect using ultrasonography, and, if DVT is strongly suspected, further testing should be performed using contrast venography, magnetic resonance imaging, or intravascular ultrasound imaging [267]. (See "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity", section on 'Alternative imaging'.)

May-Thurner syndrome is most commonly seen in women between the ages of 20 and 50 [268,269] and may be more common in those with reduced left common iliac vein diameters and/or severe degrees of iliac vein compression [270,271]. Episodes of DVT may be recurrent and/or poorly responsive to treatment with anticoagulation alone [272,273], and may require catheter-directed thrombolysis, venous angioplasty and/or intravascular stenting, especially in those with limb-threatening thrombosis [267-269].

It has been suggested that compression of the left common iliac vein by the right common iliac artery is the explanation for the predominance of left-sided over right-sided DVT in the general population, which was left sided in 55.9 percent (95% CI 54.0-57.8) in one study of 2576 subjects with unilateral DVT [274].

Inferior vena cava abnormalities — Congenital venous malformations of the inferior vena cava (IVC), including agenesis, hypoplasia, or malformation, may lead to DVT [275-278]. In such cases, DVT primarily occurs in young patients and may be bilateral or recurrent, a clinical picture similar to that of the inherited thrombophilias.

●In a series of 97 consecutive patients presenting with confirmed DVT of the lower extremities, 5 of the 31 with thrombotic occlusion of the iliac veins had an anomaly of the IVC; thrombosis was bilateral in one and recurrent in two. Average age was 25 and 53 years in the groups with or without IVC anomalies, respectively [279].

In one report of 10 cases of IVC agenesis-associated DVT (8 men, two women; mean age 25 years), DVT followed intense and unusual physical activity in eight, was bilateral in six, and was localized to iliofemoral veins in nine [280]. None had symptoms of pulmonary embolism.

Diagnostic testing — Diagnostic testing and screening for the bona fide inherited thrombophilic disorders (Factor V Leiden and prothrombin G20210A mutations, deficiencies of antithrombin, protein C and protein S), including the timing of such tests as well as the possible interferences from heparin and warfarin therapy, are discussed separately (table 5). (See "Evaluating adult patients with established venous thromboembolism for acquired and inherited risk factors" and "Screening for inherited thrombophilia in asymptomatic adults".)

ELEVATED CLOTTING FACTORS AND CHEMOKINES — Elevated levels of some clotting factors and chemokines, including factors VIII, IX, XI, thrombin-activatable fibrinolysis inhibitor (TAFI), and interleukin 8 have been associated with an increased thrombotic risk (table 6). These measurements in patients with venous thrombosis have generally been made following a thrombotic event, so that a post-thrombotic phenomenon cannot be entirely excluded.

The basis for the elevated levels of these factors has not yet been proven to be genetic, although one gain-of-function mutation in factor IX (factor IX Padua) has been described. Factor IX Padua has a specific activity 5 to 10 times as high as that of recombinant wild-type factor IX, and was associated with an episode of proximal DVT in a previously healthy 23-year-old male [281].

Factor VIII — Elevated plasma factor VIII coagulant activity (VIII:C) is now accepted as an independent marker of increased thrombotic risk [282-287]. In a population-based case control study performed in the Netherlands, patients with factor VIII:C levels greater than 150 percent of normal had an adjusted odds ratio of 4.8 for a first episode of venous thrombosis event compared with individuals with levels under 100 percent [282]. In both this study and an English series of patients referred for evaluation of unexplained thrombosis, the prevalence of elevated factor VIII:C levels was approximately 25 percent [283]. The risk appears to be additive to that of oral contraceptives [288] in contrast to the supra-additive risk observed for the hereditary thrombophilias (Factor V Leiden and prothrombin G20210A mutations, deficiencies of antithrombin, protein C, and protein S).

Asymptomatic subjects with elevated factor VIII:C levels and a positive family history of VTE or arterial events before the age of 50 appear to have a high annual incidence of first VTE and arterial vascular events [289].

An elevated factor VIII level may also be a strong thrombotic risk factor in the Black population. In a United Kingdom study of 100 Black patients of African or Caribbean origin with venous thrombosis, only 9 percent had an underlying genetic cause (eg, protein C or S deficiency, antithrombin deficiency), while 34 percent had VIII:C levels higher than the 90th percentile for normal Black subjects (odds ratio 4.6) [290].

High factor VIII levels persist over time and are not associated with increased levels of acute phase reactants (eg, C-reactive protein, fibrinogen) [283,291,292]. Thus, elevated factor VIII levels appear to be constitutional with a heritable contribution [293-295]; the genetic determinant responsible for elevated factor VIII levels has not yet been identified [296-299].

Other plasma components — In a single, large, population-based case control study performed in the Netherlands, a two to threefold increased risk for a first episode of venous thrombosis was found for altered levels of a number of plasma components, coagulant factors, anticoagulant factors, and inflammatory chemokines (table 6). Others have since been added to this list, as follows:

Elevated levels

●Plasma factor IX antigen [281,300-302]

●Plasma factor XI antigen [303,304]

●Thrombin activatable fibrinolysis inhibitor (TAFI) [305-309]

●Plasminogen activator inhibitor-1 [309]

●Interleukin 8 [310,311]

●Factor VII levels [295]

●Plasma fibronectin levels [312]

●Von Willebrand factor [295,313-315]

●Fibrinogen [292,316,317]. (See "Disorders of fibrinogen".)

●Altered fibrin clot structure and function [318]. (See "Disorders of fibrinogen".)

Reduced levels

●Tissue factor pathway inhibitor [319]

●Plasma fibrinolytic activity [309,320]

●Thrombomodulin [321]

While inherited deficiencies of antithrombin, protein C, and protein S are risk factors for VTE, subjects with borderline low values for these proteins also appear to be at increased risk for VTE, with a dose response effect on the risk of VTE for all three of these anticoagulant proteins [322]. In addition, the risk of unprovoked VTE in carriers of factor V Leiden or prothrombin G20210A appears to be increased two to threefold when levels of antithrombin or protein S are borderline low. (See 'Multiple inherited thrombotic defects' above.)

Assays for some of these proteins are not widely available and the clinical utility of these measurements, or that of a shortened activated partial thromboplastin time [323] or an elevated endogenous thrombin potential [324,325], which might reflect such procoagulant imbalances, is currently uncertain. (See "Clinical use of coagulation tests", section on 'Shortened PT and/or aPTT'.)

Single nucleotide polymorphisms — Single nucleotide polymorphisms (SNPs, variations of the nucleotide sequence in the genome occurring in at least 1 percent of the population) and other genetic variations are currently being investigated as potential genetic contributors to the risk of a number of conditions, including sickle cell anemia and VTE [41,321,326-330]. The majority of SNPs are missense (69 percent) with another 24 percent located in transcription factor binding sites or in untranslated regions of mRNA, which could affect mRNA expression or stability [41]. (See "Genetic testing".)

The databases of the Leiden Thrombophilia Study and the Multiple Environmental and Genetic Assessment of Risk Factors for Venous Thrombosis study (MEGA-1 and MEGA-2) were used to identify SNPs associated with an increased risk for a first DVT [41]. A number of such SNPs were identified, including two within the factor XI gene locus, a contribution explained at least in part by an association with elevated factor XI levels [42,331]. A risk score based upon 31 such SNPs yielded odds ratios for development of venous thrombosis ranging from 0.37 for those with no risk alleles to 7.48 for those with six or more risk alleles [43]. Interestingly, the genetic risk score based on the five most strongly associated SNPs, which have an association with the functionality of coagulation proteins (factor V Leiden, prothrombin G20210A, ABO blood group, one SNP in the fibrinogen gamma gene, and one in the factor XI gene) performed similarly to the one based on all 31 SNPs. (See "Red blood cell antigens and antibodies", section on 'ABO blood group system'.)

Non-O blood type — As noted above, the ABO blood group of the patient is a known risk factor for VTE, in that persons with the non-O blood group (ie, groups A, B, and AB) have been shown to have a significantly higher VTE risk than those with blood group O in several studies (odds ratios of 1.79 and 1.84) [213,332,333]. This may be related to associated differences in VWF and factor VIII levels between subjects with O and non-O blood groups. (See "Pathophysiology of von Willebrand disease", section on 'Clearance and control of plasma VWF levels'.)

STAB2 variants — Using gene-based collapsing analysis, rare variants have been identified in the STAB2 gene that increase the risk for VTE [334]. STAB2 encodes Stabilin-2, an endothelial cell surface scavenger receptor. The Stablin-2 variants associated with VTE exhibited lower cell surface expression, which may lead to reduced clearance and higher levels of von Willebrand factor in the blood.

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Beyond the Basics topics (see "Patient education: Deep vein thrombosis (DVT) (Beyond the Basics)" and "Patient education: Pulmonary embolism (Beyond the Basics)" and "Patient education: Antiphospholipid syndrome (Beyond the Basics)")

SUMMARY — A risk factor can now be identified in many patients with venous thromboembolism (VTE).

They are divided into two groups: hereditary and acquired (table 1). There is often more than one factor at play in a given patient, which may include both hereditary and acquired factors (table 2).

●Hereditary – The most frequent hereditary causes of VTE are the factor V Leiden and prothrombin gene mutations, which account for 50 to 60 percent of cases.

●Acquired – The major acquired risk factors for VTE include prior thromboembolism, recent major surgery (table 4), trauma, immobilization, antiphospholipid antibodies, malignancy, pregnancy, oral contraceptives, and myeloproliferative disorders.