**2. Obesity and height**

Investigations that reported an increased risk for VTE caused by obesity have been criticized because they failed to control for hospital confinement or other risk factors.33 High proportions of patients with VTE have been found to be obese,13,34 but the importance of the association is diminished because of the high proportion of obesity in the general population.35 Some investigations showed an increased risk ratio for DVT or PE in obese women,21,36,38 but data in men were less compelling. The Nurses' Health Study showed that the age-adjusted risk ratio for PE women with a body mass index (BMI, calculated as weight in kilograms divided by the square of height in meters) 29.0 kg/m2 or higher was 3.2 compared with the leanest category of less than 21.0 kg/m2.36 The Framingham Heart Study showed that metropolitan relative weight was significantly and independently associated with PE among women, but not men.39 However, the Study of Men Born in 1913 showed that men in the highest decile of waist circumference (>100 cm) had an adjusted relative risk for VTE of 3.92 compared with men with a waist circumference less than 100 cm.40 Among 1272 outpatients (men and women), the odds ratio for DVT, comparing obese (BMI> 30 kg/m2) with nonobese patients, was 2.39.41 Others showed a similar odds ratio for DVT of 2.26 compared with nonobese patients.37 BMI correlated linearly with the development of PE in women.42 On the other hand, the Olmsted County, Minnesota casecontrol study found no evidence that current BMI was an independent risk factor for VTE in men or women.33,43 Others did not show obesity to be a risk for VTE in men.21,38 Analysis of the huge database of the National Hospital Discharge Survey44 showed compelling evidence that obesity is a risk factor for VTE.45 Among patients hospitalized in short-term hospitals throughout the United States, in whom obesity was coded among the discharge diagnoses but not defined, 91,000 of 12,015,000 (0.8%) had PE.45 Among hospitalized patients who were not diagnosed with obesity, PE was diagnosed in 2,366,000 of 691,000,000 (0.3%). DVT was diagnosed in 243,000 of 12,015,000 (2.0%) of patients diagnosed with obesity, and in 5,524,000 of 691,000,000 (0.8%) who were not diagnosed with obesity. The relative risk of PE, comparing obese patients with nonobese patients, was 2.18 and for DVT it was 2.50.45 The relative risks for PE and DVT were age dependent. Obesity had the greatest effect on patients less than 40 years of age, in whom the relative risk for PE in obese patients was 5.19 and the relative risk for DVT was 5.20.45 The higher relative risk of obesity in younger patients may have reflected that younger patients uncommonly have multiple

Risk Factors of Deep Vein Thrombosis 7

estrogen, compared with users of oral contraceptives that contained less than 50 mg was 1.7.65 No difference in the risk of VTE was found with various levels of low doses of 20, 30, 40, and 50 mg/d.66 With doses of estrogen of 50 mg/d, the rate of VTE was 7.0/ 10,000 contraceptive users/y and with more than 50 mg/d, the rate of VTE was 10.0/10,000 oral contraceptive users/y.65 However, some found no appreciable difference in the relative risk of VTE in relation to low or higher estrogen doses.67 Reports of the risk of VTE in relation to the duration of use of oral contraceptives are inconsistent. Some showed relative risks increased as the duration of use of estrogen-containing oral contraceptives increased.68 The relative risks were 0.7 in women who used oral contraceptives for less than 1 year, 1.4 for those who used oral contraceptives for 1 to 4 years and 1.8 in those who used it for 5 years or longer.68 Others showed the opposite effect, with a decreasing relative risk with duration of use.66 The relative risk for DVT or PE was 5.1 with use for less than 1 year, 2.5 with use for 1 to 5 years, and 2.1 with use for longer than 5 years.66 Some showed the risk to be unaffected by the duration of use.67 A synergistic effect of oral contraceptives with obesity has been shown.69,71 The odds ratio of DVT in obese women (BMI \_30 kg/m2) who were users of oral contraceptives ranged from 5.2 to 7.8 compared with obese women who did not use oral contraceptives37,69,71 and among women with a BMI 35 kg/m2 or higher, the odds ratio was 3.1 compared with

Tamoxifen is a selective estrogen-receptor modulator used for treatment of breast cancer and for prevention of breast cancer in high-risk patients.72,74 Among women with breast cancer currently being treated with tamoxifen, compared with previous users or those who never used it, the odds ratio was 7.1.74 Others found a lower odds ratio of 2.7.43 The odds ratio for VTE in women at high risk of breast cancer who received tamoxifen to prevent

There is a 2- to 3-fold increased risk of VTE with the use of hormone replacement therapy in postmenopausal women.75,76 Among postmenopausal women who had coronary artery disease and received estrogen plus progestin, the relative hazard of VTE was 2.7 compared with nonusers.77 Review showed that the risk of VTE is highest in the first year of hormone replacement therapy.78 The risk of VTE is increased for oral estrogen alone, oral estrogen combined with progestin, and probably for transdermal hormone replacement therapy.78

Antithrombin is a serine protease inhibitor of thrombin and also inhibits factors IXa, Xa, XIa, and XIIa. Thrombin is irreversibly bound by antithrombin and prevents thrombin's action on fi brinogen, on factors V, VIII, and XIII, and on platelets.79 This anticoagulant is synthesized in the liver and endothelial cells, and has a half-life of 2.8 days.80 Antithrombin deficiency has a prevalence of 1 : 5000 with more than 100 genetic mutations and an autosomal dominant inheritance pattern.81 Homozygotes typically die *in utero* whereas heterozygotes typically have an antithrombin level that is 40 to 70% of normal.

similarly obese nonusers of oral contraceptives.71

**7. Hormonal replacement therapy** 

**8. Congenital hypercoagulable disorders** 

**8.1 Antithrombin deficiency** 

breast cancer was 2.1.73 Others found a hazard ratio of 1.63.72

**6. Tamoxifen** 

confounding- associated risk factors, which make the risk of obesity inapparent. Previous investigators used several indices of obesity including a BMI greater than 35 kg/m2 as well as BMI 30 to 35 kg/m2,46 BMI 29 kg/m2 or greater,36 weight more than 20% of median recommended weight for height,13 and for men, waist circumference 100 cm or greater.40 It is likely that all patients diagnosed with obesity in the National Hospital Discharge Survey database were obese, irrespective of the criteria used. However, some obese patients may not have had a listed discharge diagnosis of obesity, and they would have been included in the nonobese group. This situation would have tended to reduce the relative risk of obesity in VTE. Various abnormalities of hemostasis have been described in obesity, in particular increased plasminogen activator inhibitor-1 (PAI-1).47,48 Other abnormalities of coagulation have been reported as well,48 including increased platelet activation,39 increased levels of plasma fibrinogen, factor VII, factor VIII, and von Willebrand factor.49 Fibrinogen, factor VIIc, and PAI-1 correlated with BMI.50 Regarding height, in the study of Swedish men, those taller than 179 cm (5' 10") had a 1.5 times higher risk of VTE than men shorter than 172 cm.51 The Physicians' Health Study of male physicians also showed that taller men had a significantly increased risk of VTE.52

#### **3. Air travel**

The possibility of VTE after travel is not unique to air travel.53,54 Prolonged periods in cramped quarters, irrespective of travel, can lead to PE.55 The term economy class syndrome was introduced in 1988,56 but has since been replaced with flight-related DVT in recognition that all travelers are at risk, irrespective of the class of travel57 Rates of development of PE with air travel lasting 12 to 18 hours have been calculated as 2.6 PE/million travelers.58 With air travel of 8 hours or longer, 1.65/million passengers had acute PE on arrival.59 With 6 to 8 hours of air travel the rate of acute PE on arrival was 0.25/million and among those who traveled for 6 hours or less none developed acute PE on arrival.59 The trend showing increasing rates of PE with duration of travel is compelling, but the incidence of DVT was about 3000 times higher in a prospective investigation.60 In a prospective investigation of travelers who traveled for 10 hours or longer, 4 of 878 (0.5%) developed PE and 5 of 878 (0.6%) developed DVT.60

#### **4. Varicose veins**

Varicose veins were found by some to be an agedependent risk factor for VTE.43 Among patients aged 45 years the odds ratio for VTE was 4.2.43 In patients aged 60 years the odds ratio was 1.9 and at aged 75 years, varicose veins were not associated with an increased risk of VTE.43 However, others did not find varicose veins to be a risk factor for DVT61 or PE found at autopsy.21

#### **5. Oral contraceptives**

Although the risk of VTE is higher among users of oral estrogen-containing contraceptives than nonusers, 62,63 the absolute risk is low.64 An absolute risk of VTE of less than 1/10,000 patients/y increased to only 3 to 4/10,000 patients/y during the time oral contraceptives were used.64 The relative risk for VTE in women using oral contraceptives containing 50 mg of estrogen, compared with users of oral contraceptives that contained less than 50 mg was 1.5.65 The relative risk for VTE in women using oral contraceptives containing more than 50 mg of

confounding- associated risk factors, which make the risk of obesity inapparent. Previous investigators used several indices of obesity including a BMI greater than 35 kg/m2 as well as BMI 30 to 35 kg/m2,46 BMI 29 kg/m2 or greater,36 weight more than 20% of median recommended weight for height,13 and for men, waist circumference 100 cm or greater.40 It is likely that all patients diagnosed with obesity in the National Hospital Discharge Survey database were obese, irrespective of the criteria used. However, some obese patients may not have had a listed discharge diagnosis of obesity, and they would have been included in the nonobese group. This situation would have tended to reduce the relative risk of obesity in VTE. Various abnormalities of hemostasis have been described in obesity, in particular increased plasminogen activator inhibitor-1 (PAI-1).47,48 Other abnormalities of coagulation have been reported as well,48 including increased platelet activation,39 increased levels of plasma fibrinogen, factor VII, factor VIII, and von Willebrand factor.49 Fibrinogen, factor VIIc, and PAI-1 correlated with BMI.50 Regarding height, in the study of Swedish men, those taller than 179 cm (5' 10") had a 1.5 times higher risk of VTE than men shorter than 172 cm.51 The Physicians' Health Study of male physicians also showed that taller men had a

The possibility of VTE after travel is not unique to air travel.53,54 Prolonged periods in cramped quarters, irrespective of travel, can lead to PE.55 The term economy class syndrome was introduced in 1988,56 but has since been replaced with flight-related DVT in recognition that all travelers are at risk, irrespective of the class of travel57 Rates of development of PE with air travel lasting 12 to 18 hours have been calculated as 2.6 PE/million travelers.58 With air travel of 8 hours or longer, 1.65/million passengers had acute PE on arrival.59 With 6 to 8 hours of air travel the rate of acute PE on arrival was 0.25/million and among those who traveled for 6 hours or less none developed acute PE on arrival.59 The trend showing increasing rates of PE with duration of travel is compelling, but the incidence of DVT was about 3000 times higher in a prospective investigation.60 In a prospective investigation of travelers who traveled for 10 hours or longer, 4 of 878 (0.5%)

Varicose veins were found by some to be an agedependent risk factor for VTE.43 Among patients aged 45 years the odds ratio for VTE was 4.2.43 In patients aged 60 years the odds ratio was 1.9 and at aged 75 years, varicose veins were not associated with an increased risk of VTE.43 However, others did not find varicose veins to be a risk factor for DVT61 or PE

Although the risk of VTE is higher among users of oral estrogen-containing contraceptives than nonusers, 62,63 the absolute risk is low.64 An absolute risk of VTE of less than 1/10,000 patients/y increased to only 3 to 4/10,000 patients/y during the time oral contraceptives were used.64 The relative risk for VTE in women using oral contraceptives containing 50 mg of estrogen, compared with users of oral contraceptives that contained less than 50 mg was 1.5.65 The relative risk for VTE in women using oral contraceptives containing more than 50 mg of

significantly increased risk of VTE.52

developed PE and 5 of 878 (0.6%) developed DVT.60

**3. Air travel** 

**4. Varicose veins** 

found at autopsy.21

**5. Oral contraceptives** 

estrogen, compared with users of oral contraceptives that contained less than 50 mg was 1.7.65 No difference in the risk of VTE was found with various levels of low doses of 20, 30, 40, and 50 mg/d.66 With doses of estrogen of 50 mg/d, the rate of VTE was 7.0/ 10,000 contraceptive users/y and with more than 50 mg/d, the rate of VTE was 10.0/10,000 oral contraceptive users/y.65 However, some found no appreciable difference in the relative risk of VTE in relation to low or higher estrogen doses.67 Reports of the risk of VTE in relation to the duration of use of oral contraceptives are inconsistent. Some showed relative risks increased as the duration of use of estrogen-containing oral contraceptives increased.68 The relative risks were 0.7 in women who used oral contraceptives for less than 1 year, 1.4 for those who used oral contraceptives for 1 to 4 years and 1.8 in those who used it for 5 years or longer.68 Others showed the opposite effect, with a decreasing relative risk with duration of use.66 The relative risk for DVT or PE was 5.1 with use for less than 1 year, 2.5 with use for 1 to 5 years, and 2.1 with use for longer than 5 years.66 Some showed the risk to be unaffected by the duration of use.67 A synergistic effect of oral contraceptives with obesity has been shown.69,71 The odds ratio of DVT in obese women (BMI \_30 kg/m2) who were users of oral contraceptives ranged from 5.2 to 7.8 compared with obese women who did not use oral contraceptives37,69,71 and among women with a BMI 35 kg/m2 or higher, the odds ratio was 3.1 compared with similarly obese nonusers of oral contraceptives.71

#### **6. Tamoxifen**

Tamoxifen is a selective estrogen-receptor modulator used for treatment of breast cancer and for prevention of breast cancer in high-risk patients.72,74 Among women with breast cancer currently being treated with tamoxifen, compared with previous users or those who never used it, the odds ratio was 7.1.74 Others found a lower odds ratio of 2.7.43 The odds ratio for VTE in women at high risk of breast cancer who received tamoxifen to prevent breast cancer was 2.1.73 Others found a hazard ratio of 1.63.72
