Venous thromboembolism: disease burden, outcomes and risk factors
J. A. HEIT
Divisions of Cardiovascular Diseases (Section of Vascular Diseases) and Hematology (Section of Hematology Research), Department of Internal Medicine; and the Divisions of Hematopathology and Laboratory Genetics, Department of Laboratory Medicine and Pathology; Mayo Clinic and Mayo Foundation, Rochester, MN, USA
John A. Heit, Hematology Research, Stabile 660, Mayo Clinic, 200 First Street SW, Rochester, MN 55905, USA. Tel.: +1 507 284 4634; fax: +1 507 266 9302; e-mail: email@example.com
Summary. The epidemiology of venous thromboembolism (VTE) in the community has important implications for VTE prevention and management. This review describes the disease burden (incidence), outcomes (survival, recurrence and complications) and risk factors for deep vein thrombosis and pulmonary embolism occurring in the community. Recent comprehensive studies of the epidemiology of VTE that reported the racial demography and included the full spectrum of disease occurring within a well-defined geographic area over time, separated by event type, incident vs. recurrent event and level of diagnostic certainty, were reviewed. Studies of VTE outcomes had to include a relevant duration of follow-up. VTE incidence among whites of European origin exceeded 1 per 1000; the incidence among persons of African and Asian origin may be higher and lower, respectively. VTE incidence over recent time remains unchanged. Survival after VTE is worse than expected, especially for pulmonary embolism. Thirty percent of patients develop VTE recurrence and venous stasis syndrome. Exposures can identify populations at risk but have a low predictive value for the individual. An acquired or familial thrombophilia may predict the subset of exposed persons who actually develop symptomatic VTE. In conclusion, VTE is a common, lethal disease that recurs frequently and causes serious long-term complications. To improve survival and prevent complications, VTE occurrence must be reduced. Better individual risk stratification is needed in order to modify exposures and target primary and secondary prophylaxis to the person who would benefit most.
This review focuses on the implications for venous thromboembolism (VTE) prevention and management that can be drawn from the epidemiology of deep vein thrombosis of the leg, pelvis or arm, and its complication, pulmonary embolism. The epidemiology of thrombosis affecting other venous circulations (e.g. cerebral sinus, mesenteric, renal, hepatic, portal) is beyond the scope of this review.
Venous thromboembolism is now recognized as a complex (multifactorial) disease, involving both environmental exposures (e.g. clinical risk factors) as well as genetic and environmental interactions. Most epidemiology studies to date have addressed populations of primarily European origin, and the data discussed in this review mainly relate to those populations. Where available, data from populations with origins from other continents are presented. Also where available, population-based studies of VTE epidemiology (e.g. incidence, survival, recurrence, complications, and risk factors) are presented as it is data from these populations that are most generalizable to the reader's individual patients.
The incidence of deep vein thrombosis and pulmonary embolism
The average annual incidence rates of VTE among whites of predominantly European origin (e.g. Olmsted County, Minnesota) during the 25-year period, 1966–1990 (age- and sex-adjusted to the 1980 US white population) and during the more recent 7-year period, 1991–1997 (similarly adjusted, but to the 2000 US white population) were 117 and 117.7 per 100 000 person-years, respectively . Based on the more recent rates, 249 000 incident VTE cases occur annually among US whites. The incidence appears to be similar or higher among African-Americans and lower among Asian- and Native-Americans [2–6]. Assuming that the 1991–1997 age- and sex-specific VTE incidence among Blacks (Black or African American alone) is comparable to whites, then, taking into account the different age and sex distribution of Black-Americans, the overall age- and sex-adjusted VTE incidence was 77.6 per 100 000. Based on this incidence, 27 000 incident VTE cases occur annually among US Blacks, for a total of over 275 000 new VTE cases per year in the US.
The VTE incidence over the more recent 7-year period, 1991–1997 (117.7 per 100 000), has not changed significantly compared to the 10-year period, 1981–1990 (116.7 per 100 000; Fig. 1). The observed decline in incidence from 1975 to 1977 likely reflects the dramatic increase in utilization of objective diagnostic imaging for VTE (e.g. venography, nuclear lung scan) .
Venous thromboembolism is predominantly a disease of older age [1,8]. In the absence of a central venous catheter  or thrombophilia , VTE is rare prior to late adolescence [1,11]. The age- and sex-adjusted VTE incidence rate for persons age 15 years or older is 149–192 per 100 000 [1,8]. Incidence rates increase markedly with age for both men and women and for both deep vein thrombosis and pulmonary embolism [1,8]. The overall age-adjusted incidence rate is higher for men (130 per 100 000) then women (110 per 100 000; male:female sex ratio is 1.2:1) . Incidence rates are somewhat higher in women during the childbearing years, while incidence rates after age 45 years are generally higher in men. Pulmonary embolism accounts for an increasing proportion of VTE with increasing age for both genders .
Survival after deep vein thrombosis and pulmonary embolism
Survival after VTE is worse than expected, and survival after pulmonary embolism is much worse than after deep vein thrombosis alone [12–14]. The risk of early death among patients with symptomatic pulmonary embolism is 18-fold higher compared to patients with deep vein thrombosis alone . Pulmonary embolism is an independent predictor of reduced survival for up to 3 months after onset. For almost one-quarter of pulmonary embolism patients, the initial clinical presentation is sudden death. Independent predictors of reduced early survival after VTE include increasing age, male gender, lower body mass index, confinement to a hospital or nursing home at VTE onset, congestive heart failure, chronic lung disease, serious neurological disease, and active malignancy [8,12,13]. Additional clinical predictors of poor early survival after pulmonary embolism include syncope and arterial hypotension . Evidence of right heart failure based on clinical examination, plasma markers (e.g. cardiac troponin T, brain natriuretic peptide) [16,17] or echocardiography  predicts poor survival among normotensive pulmonary embolism patients. Pulmonary embolism patients with these characteristics should receive aggressive anticoagulation therapy, and possibly thrombolytic therapy in selected cases [18,19]. Survival over time may be improving for those pulmonary embolism patients living sufficiently long to be diagnosed and treated [12,20,21].
Venous thromboembolism recurrence
Venous thromboembolism recurs frequently; about 30% of patients develop recurrence within the next 10 years . The hazard of recurrence varies with the time since the incident event and is highest within the first 6–12 months. However, even at 10 years after an incident deep vein thrombosis or pulmonary embolism, VTE patients are still at risk for recurrence. Moreover, while active therapeutic anticoagulation is effective in preventing recurrence [23–25], the duration of anticoagulation does not affect the risk of recurrence once primary therapy for the incident event is stopped [26–28]. These data suggest that for a subset of patients, VTE is a chronic disease with episodic recurrence; indefinite secondary prophylaxis may be warranted for this patient subset [23–25,28,29]. Independent predictors of recurrence include male gender [22,30,31], increasing patient age and body mass index, neurological disease with extremity paresis, and active malignancy [8,22,32–35]. Additional predictors include ‘idiopathic’ VTE [24,26,35], a lupus anticoagulant or antiphospholipid antibody [24,36], antithrombin, protein C or protein S deficiency , and possibly persistent residual deep vein thrombosis . Prolonged secondary prophylaxis with anticoagulation therapy should be considered for patients with these characteristics.
Of equal importance, several baseline characteristics predict either a reduced risk of recurrence or are not predictive of recurrence [22,32,34,35]. For women, pregnancy or the postpartum state, oral contraceptive use, and gynecologic surgery are associated with a reduced risk of recurrence . Recent surgery, trauma or fracture either have no effect  or predict a reduced risk of recurrence . Additional characteristics having no significant effect on recurrence risk include recent immobilization, hormone or tamoxifen therapy, and failed prophylaxis . For these patients, and for patients with thrombosis limited to a calf vein or deep muscular vein (e.g. gastrocnemius, soleus), a shorter duration of oral anticoagulation therapy likely is adequate [27,28,39]. While the incident event type (deep vein thrombosis alone vs. pulmonary embolism) is not a predictor of recurrence, patients with recurrence are significantly more likely to recur with the same event type as the incident event type [40,41]. Because the 7-day case fatality rate is significantly higher for recurrent pulmonary embolism (34%) compared to recurrent deep vein thrombosis alone (4%) , prolonged anticoagulation should be considered for incident pulmonary embolism, especially for patients with chronically reduced cardiopulmonary functional reserve.
Venous stasis syndrome and venous ulcer
The major complications of VTE are venous stasis syndrome (dependent leg swelling and pain, stasis dermatitis) and venous ulcer. The 20-year cumulative incidence of venous stasis syndrome after VTE and after proximal deep vein thrombosis are about 25% and 40%, respectively [34,42]. Risk factors for venous stasis syndrome include the VTE event type (deep vein thrombosis) and location (proximal deep vein thrombosis). The 20-year cumulative incidence of venous ulcer is 3.7% . The risk for venous ulcer is increased 30% per decade of age at the incident VTE . Venous thromboembolism accounts for about 11% of all venous stasis syndrome occurring in the community . Given the pain, impairment, and high costs associated with these complications , prevention of venous stasis syndrome is of paramount importance. Below knee graduated compression elastic stockings reduce the incidence of venous stasis syndrome by about 50% .
Risk factors for venous thromboembolism
In order to improve survival, avoid recurrence, prevent complications, and reduce health care costs, the occurrence of VTE must be reduced. To reduce VTE incidence, persons at risk for VTE must first be identified. Independent risk factors for VTE include patient age, surgery, trauma, hospital or nursing home confinement, active malignant neoplasm with or without concurrent chemotherapy, central vein catheterization or transvenous pacemaker, prior superficial vein thrombosis, varicose veins, and neurological disease with extremity paresis; patients with chronic liver disease have a reduced risk [46,47]. The incidence of VTE increases significantly with age for both idiopathic and secondary VTE, suggesting that the risk associated with advancing age may be due to the biology of aging rather than simply an increased exposure to VTE risk factors with advancing age . Compared to residents in the community, hospitalized residents have over a 150-fold increased incidence of acute VTE . Hospitalization and nursing home residence together account for almost 60% of incident VTE events occurring in the community . Thus, hospital confinement provides an important opportunity to significantly reduce VTE incidence. Of note, hospitalization for medical illness and hospitalization for surgery account for almost equal proportions of VTE (22% and 24%, respectively), emphasizing the need to provide prophylaxis to both of these risk groups. Nursing home residence independently accounts for over one-tenth of all VTE disease in the community .
The risk among surgery patients can be further stratified based on patient age, type of surgery, and the presence of active cancer [51,52]. The incidence of postoperative VTE is increased for surgery patients that are 65 years of age or older . High-risk surgical procedures include neurosurgery, major orthopedic surgery of the leg, thoracic, abdominal or pelvic surgery for malignancy, renal transplantation, and cardiovascular surgery . Obesity [53,54] and poor American Society of Anesthesiology physical status  are risk factors for VTE after total hip arthroplasty. The risk from surgery may be less with neuraxial (spinal or epidural) anesthesia compared to general anesthesia . Risk factors for VTE among patients hospitalized for acute medical illness may include infection, age >75 years, cancer, and prior VTE .
Active cancer accounts for almost 20% of incident VTE events occurring in the community . The risk appears to be higher for patients with pancreatic cancer, lymphoma, malignant brain tumors, cancer of the liver, leukemia, and colorectal and other digestive cancers [57,58]. Cancer patients receiving immunosuppressive or cytotoxic chemotherapy are at even higher risk for VTE , including therapy with l-asparaginase, thalidomide, or tamoxifen. A routine examination for occult malignancy is warranted for patients presenting with idiopathic VTE, especially among patients in whom VTE recurs .
A central venous catheter or transvenous pacemaker now accounts for 9% of incident VTE occurring in the community . Prior superficial vein thrombosis is an independent risk factor for subsequent deep vein thrombosis or pulmonary embolism remote from the episode of superficial thrombophlebitis . The risk of deep vein thrombosis imparted by varicose veins is uncertain and appears to vary by patient age . Long haul (>6 h) air travel is associated with a slightly increased risk for VTE that is preventable with elastic stockings . Coenzyme A reductase inhibitor (statin) therapy may provide a 20%–50% risk reduction for VTE . However, the risk associated with atherosclerosis, or other risk factors for atherosclerosis, remains uncertain [62–64]. Body mass index, current or past tobacco smoking, chronic obstructive pulmonary disease, and renal failure are not independent risk factors for VTE . The risk associated with congestive heart failure, independent of hospitalization, is low [46,47].
Among women, additional risk factors for VTE include oral contraceptive use and hormone therapy , pregnancy and the postpartum period [47,66], and therapy with the selective estrogen receptor modulator, raloxifene. First- and third-generation oral contraceptives convey higher risk than second-generation oral contraceptives [65,67]. Hormone therapy is associated with a two- to fourfold increased risk of venous thrombosis , but the risk may vary by type of estrogen . About 1 in 2000 women will develop venous thrombosis during pregnancy . The risk during the postpartum period is about fivefold higher than the risk during pregnancy. Prior superficial vein thrombosis is an independent risk factor for venous thrombosis during pregnancy or postpartum .
Other conditions associated with VTE include heparin-induced thrombocytopenia, myeloproliferative disorders (especially polycythemia rubra vera and essential thrombocythemia), intravascular coagulation and fibrinolysis/disseminated intravascular coagulation, nephrotic syndrome, paroxysmal nocturnal hemoglobinuria, thromboangiitis obliterans (Buerger's disease), thrombotic thrombocytopenic purpura, Bechet's syndrome, systemic lupus erythematosis, inflammatory bowel disease, homocystinuria, and possibly hyperhomocysteinemia [71,72].
The genetic epidemiology of venous thromboembolism
Recent family based studies indicate that VTE is highly heritable and follows a complex mode of inheritance involving environmental interaction [73–75]. Inherited reductions in plasma natural anticoagulants (e.g. antithrombin, protein C, or protein S) have long been recognized as uncommon but potent risk factors for VTE [76,77]. More recent discoveries of impaired downregulation of the procoagulant system (e.g. activated protein C resistance, Factor V Leiden) [78–80], increased plasma concentrations of procoagulant factors [e.g. factors I (fibrinogen), II (prothrombin), VIII, IX, and XI] [81–85] or activity , and activation innate immunity  have added new paradigms to the list of inherited or acquired disorders predisposing to thrombosis (thrombophilia). Prospective cohort studies indicate that VTE patients have increased basal fibrin formation [88,89]. These plasma hemostasis-related factors or markers of coagulation activation both correlate with increased thrombotic risk and are highly heritable [90–94]. These inherited thrombophilias interact with such clinical risk factors (e.g. environmental risk factors) as oral contraceptives [95,96], pregnancy , hormone therapy , and surgery  to increase the risk of incident VTE. Similarly, genetic interaction increases the risk of incident  and recurrent VTE [101–105]. While the clinical utility of diagnostic testing for an inherited or acquired thrombophilia remains controversial, such studies hold the potential for further stratifying individual patients into high- and low-risk for incident and recurrent VTE, targeting prophylaxis to those who would benefit most, and ultimately, reducing the occurrence of VTE.
Funded, in part, by grants from the National Institutes of Health (HL-60279, HL-66216, AR-30582) and Centers for Disease Control and Prevention (TS326), the US Public Health Service, and by the Mayo Foundation.