Women’s Issues and Venous Thromboembolism
Surgeon General's Workshop on Deep Vein Thrombosis
Suman Rathbun, MD, MS
University of Oklahoma Health Sciences Center
Graphic. The risk of venous thromboembolism (VTE) throughout a woman's life has been found to vary chronologically. Hormonal exposure in large part underlies this risk. Early, a woman may be exposed to oral contraceptive therapy (OCPs), later pregnancy, and finally with menopause, a woman may choose to receive hormone replacement therapy (HRT). This varying risk may be related to the duration of estrogen exposure. In addition, the risk of VTE in women is inherently influenced by age, race, weight, and inherited predisposition to clotting (thrombophilia). Externally, this risk is influenced by the formulation, dose and duration of estrogen and progestin administration. Taken together, these factors may pose significant risk of VTE throughout a woman's life and should be thoughtfully considered as an important women's health issue.
Graphic showing Risk of VTE with OCPs. Over the past 10 years, there have been numerous evaluations of the risk of VTE associated with the use of oral contraceptive preparations. This risk has been further stratified by presence of thrombophilia, age, weight, estrogen dose and route of administration, and progestin preparation. The best estimates are that among all users of OCPs, the risk of VTE is increased 2 to 8-fold compared with non-users. While this elevation in risk with OCP use warrants discussion, it must be put in the context of the very low overall risk of VTE with OCP use. For example, women of fertile age have a risk of VTE of 1 per 10,000 persons per year that may be increase to 2 to 8 per 10,000 person-years with the use of OCPs. Compared with the risk of pregnancy that is associated with an even greater risk of VTE and the social and economic consequences thereafter, the low risk of VTE with effective OCP contraceptive use must be noted.
The biological effects of OCPs on clotting factors include increased levels of clotting proteins, and decreased levels of clotting inhibitors. OCP use also leads to an acquired activated protein C resistance thereby preventing the breakdown of clot. Recently it has been found that this resistance may be more pronounced with use of the newer third-generation OCPs compared with second-generation OCPs, and less pronounced in those using progestin-only OCPs.
Investigation of the temporal relationship between OCP use and VTE risk has revealed that the risk is increased in first time users of OCPs and is especially elevated during the first six months of use revealing those who may have inherited clotting tendencies. With discontinuation of use, the risk returns to that of non-users within one to three months. Duration of use or lifetime cumulative use does not seem to alter the VTE risk associated with OCP use, nor does re-exposure after a period of non-use.
It is known that the incidence of VTE is rare before puberty, and increases with age. The risk in women over 40 may be more than three times greater than in those less than 25 years of age. However, there have been few quantitative studies that document the magnitude of interaction between age and OCP use. Reasonable assessment dictates that OCP use confers a higher risk of VTE with increasing age.
Similarly, while it is well known that smoking increases the risk of cardiovascular events in women who use OCPs, the risk of VTE in OCP users that smoke has not been well elucidated, but has been shown to be increased approximately two-fold.
The evaluation of elevated body mass index (BMI) in conjunction with OCP use has shown that the risk of VTE is increased five to ten-fold, and especially significant with BMI greater than 30kg/m2. This risk is further increased in the presence of smoking.
Graphic showing OCP formulation. After the introduction of OCPs in the 1960s, it became readily apparent that use was associated with an elevated risk of cardiovascular disease. This recognition prompted evaluation of the formulation of OCPs, scrutinizing both the dose of estrogen used as well as the formulation of the progestin component . Subsequently, the form of estrogen was changed from mestranol to ethinyl estradiol, and the estrogen dose was lowered from 100 micrograms (first-generation) to 50 micrograms (second-generation) with preparations available containing 30 micrograms or less. This reduction was effective in lowering the risk of both cardiovascular disease and VTE. The progestin component was also reviewed, and studies were performed to evaluate any differences in risk with levonorgestrel found in second-generation OCPs compared with desogestrel or gestodene found in the newer third-generation formulations. Early studies showed that the risk of venous thrombosis associated with third generation OCPs was significantly higher by about two-fold compared with second generation OCPs. While initially disputed as due to confounding by age, two recent meta-analyses confirmed the increased risk with third-generation OCPs. Newer fourth-generation preparations containing cyproterone acetate are also being evaluated, and preliminary studies reveal a four-fold increased risk in VTE compared with second generation OCPs. To date, no studies have been published that confirm the increased risk of VTE with the fourth-generation progestin drospirenone. In addition, there have been no apparent differences found between monophasic OCPs giving a constant daily dose of estrogen and progestin compared with multiphasic OCPs that vary the dose of the components through the cycle.
Progesterone-only preparations containing injectable medroxyprogesterone are infrequently used. Case-control studies have found little effect on the risk of VTE when used for contraception. Other progesterone-only products include an oral daily tablet and an intrauterine device. Both these preparations carry less risk of VTE compared with combination estrogen/progestin products, and may be an option for women at high risk for VTE.
The transdermal (through the skin) contraceptive patch containing a combination of ethinyl estradiol and the progestin norelgestromin was marketed in 2002. Since, then only one study has been published that evaluated the risk of venous thromboembolism. This case-control study compared the risk of non-fatal VTE in woman receiving the transdermal contraceptive patch with monophasic or triphasic OCPs containing norgestimate and 35 micrograms of ethinyl estradiol. Overall, the risk of non-fatal VTE was no higher in those receiving the transdermal patch compared with those receiving OCPs, although the effect of duration of use could not be evaluated.
|Mutation||No. Studies||Cases||Controls||OR (95% CI)|
|None||7||411/733||529/1797||3.1 (2.17, 4.42)|
|Factor V L||6||69/365||73/1252||3.78 (2.22, 6.42)|
|Prothrombin G||3||30/260||64/760||1.34 (0.81, 2.23)|
|AT3||1||5/60||4/144||3.18 (0.82, 12.29)|
|Protein C||1||17/60||20/144||2.45 (1.18, 5.11)|
|Protein S||2||21/85||13/250||5.31 (2.48, 11.37)|
|Fact V + Pro G||2||5/220||3/621||4.03 (1.01, 16.01)|
Graphic showing VTE Risk During Pregnancy. The reported overall risk of venous thromboembolism during pregnancy is variable, and though to be from 18 per 100,000 women during pregnancy to 1900 per 100,000 women post-partum, but generally the risk is five times more likely in pregnant women compared with non-pregnant women. Moreover, pulmonary embolism is the most common cause of maternal death in developed countries. DVT occurs more commonly on the left side. Acquired resistance to activated protein C as discussed above, and a reduction in protein S as well as venous stasis (pooling of blood in the veins) are the likely physiological predisposing causes of VTE during pregnancy. Only recently has longitudinal data been presented that more descriptively assesses the risk. In a 30 year population-based cohort study in Olmstead County, Minnesota, it was found the relative risk of VTE during pregnancy and in the post-partum period is four-fold compared with those not pregnant.
Other notable findings were a relatively constant rate of VTE during the last 20 weeks of pregnancy, and a much higher risk post-partum. The incidence of VTE increased with age, but younger women age 15 to 19 had the highest risk overall. The reasons for this finding are not clear. Smoking was associated with a higher risk of VTE during pregnancy. Previous studies have found that obesity, immobilization, increased gestational age at delivery, and preeclampsia were associated with a higher risk, but these findings were not confirmed in this study. Only a history of VTE and thrombophilia were associated with higher risk.
The risk of pulmonary embolism, especially post-partum, decreased over the study period. Although this study included mostly white patients, other studies have found that pregnancy-associated VTE is more common in blacks. There is no consistent data confirming an increased risk of VTE after C-section.
|Mutation||No. Studies||Cases||Controls||OR (95% CI)|
|FVL (Homo)||5||29/91||145/1248||34.4 (9.86, 120.05)|
|Fact V L (Hetero)||6||96/226||263/1595||8.32 (5.44, 12.7)|
|Prothrombin G||4||42/61||277/1005||6.8 (2.46, 18.77)|
|MTHFR (Homo)||4||20/128||89/543||0.74 (0.22, 2.48)|
|AT 3||3||8/11||242/815||4.69 (1.30, 16.96)|
|Protein C||3||23/32||232/715||4.76 (2.15, 10.57)|
|Protein S||3||15/28||250/911||3.19 (1.48, 6.89)|
Graphic of prevention during pregnancy (risk groups and recommendations). The use of medical prevention with unfractionated heparin (UFH) or low molecular weight heparin (LMWH) in women with a history of VTE or known thrombophilia is controversial. There are no large randomized trials, and recommendations are based on small studies. Two studies have evaluated the risk of recurrent thrombosis during pregnancy in women with a history of VTE and found it to be between 2 and 6% in women not receiving any pharmacological prophylaxis. Women with known hereditary thrombophilia were at higher risk for recurrent VTE. Consensus panel recommendations support use of prophylaxis with UFH or LMWH for women with a single VTE and known thrombophilia, and therapeutic dose UFH or LMWH in women with a history of recurrent VTE. In women with a history of VTE associated with a risk factor that is no longer present, watchful waiting without pharmacological treatment is appropriate during pregnancy with medical prophylaxis post-partum for 6 weeks. For pregnant women with known antiphospholipid antibodies and one or more fetal losses, a combination of heparin and low dose aspirin is recommended. All women with a history of VTE, should wear graduated compression stockings during pregnancy and post-partum.
The treatment of acute VTE during pregnancy with either UFH or LMWH has shown to be effective during pregnancy (64). Either intravenous UFH or adjusted-dose LMWH for 5 days followed by adjusted-dose UFH or LMWH for the remainder of pregnancy is recommended. While LMWH may be cleared through the kidneys more quickly in pregnancy, there are no consistent guidelines regarding monitoring of clotting factor (anti-factor Xa) levels during pregnancy. Warfarin may be substituted post-partum and continued for at least 6 weeks. The use of UFH and LMWH has been shown to be safe for use in pregnancy, both for prevention and treatment of VTE. Major bleeding is rare and occurs with similar incidence to that in non-pregnant women. Heparin-induced thrombocytopenia (low platelets) and osteoporosis occur more commonly after prolonged UFH use, but are much less common with LMWH use. Anticoagulation should be discontinued 24 hours prior to elective induction of labor. Heparin and LMWH do not cross the placenta and are not secreted into breast milk. Warfarin use has also been found to be safe for nursing mothers and their infants. For pregnant women with contraindications to blood-thinning therapy, there is anecdotal evidence that the use of retrievable inferior vena cava filters is safe.
Graphic showing Risk of VTE with HRT. The benefits and risks of hormone replacement therapy (HRT) have been the topic of rigorous meta-analyses, and large randomized controlled trials, yet still is under great debate. Most oral preparations contain conjugated estrogens derived from pregnant mare urine or micronized estradiol and are given in combination with progesterone acetate to negate the increased risk of endometrial cancer in those without hysterectomy. In addition, phytoestrogens have recently been marketed as a natural source of estrogen replacement.
Most agree that HRT improves the negative symptoms of menopause including hot flashes, depression, and sleep disturbance, and increases bone density, however, the overall cardiovascular benefit of HRT is suspect. Further, these benefits must be balanced against the increase risk of stroke, breast cancer and venous thromboembolism. Recent large prospective trials, the Heart and Estrogen/progestin replacement study (HERS) and Women's Health Initiative (WHI), have confirmed the 2 to 4-fold increase in VTE risk with HRT. The risk is greater with a larger dose of estrogen administered >1.25mg/day, and seems to be greater within the first year returning to normal within months of discontinuation of HRT.
There is emerging evidence that the timing of HRT after menopause may influence the overall risk profile. One recent presentation of the Danish Osteoporosis Prevention Study (DOPS) reported no increased risk of VTE on ten-year follow-up in women aged 45-58 years with HRT administration. The participants in the large HERS and WHI trials were significantly older with too few younger women enrolled to confirm this possible lower risk in younger women.
|Mutation||Estrogen||Cases||Controls||OR (95% CI)|
|FVL||None||16||18||2.6 (1.3, 5.4)|
|Oral||11||3||16.4 (4.3, 62.2)|
|Transdermal||8||6||4.6 (1.6, 13.8)|
|Protein G||None||15||7||6.4 (2.5, 16.5)|
|Transdermal||6||7||3.3 (1.1, 10.2)|
Graphic of Prevention with HRT. Prevention of VTE should be aimed at avoidance of HRT in high-risk groups including those with a history of VTE. In women with genetic thrombophilia who use HRT, the risk of VTE is elevated compared with non-users, however, the overall incidence remains low. Use of estrogen-only or transdermal estrogen may confer less risk of VTE. If required, low-dose estrogen in younger woman for a short duration for relief of post-menopausal symptoms likely is associated with the lowest risk of VTE. To date, no studies have shown an increase in post-operative risk in woman using HRT who undergo elective surgery compared with non-users. Importantly, both mechanical and pharmacological preventive measures should be used appropriately in these patients.