Oral Contraceptives and the Risk for Venous Thromboembolism

Peter Kovacs, MD, PhD


October 09, 2009

The Venous Thrombotic Risk of Oral Contraceptives, Effects of Oestrogen Dose and Progestogen Type: Results of the MEGA Case-Control Study

van Hylckama Vlieg A, Helmerhorst FM, Vandenbroucke JP, Doggen CJ, Rosendaal FR
BMJ. 2009;339:b2921


Oral contraceptive pills (OCP) are among the most common forms of contraception and are used by approximately one fourth of women of reproductive age.[1] Combination pills contain estrogen and progestogen. They work by suppressing follicle-stimulating hormone and luteinizing hormone production, altering tubal motility, and inducing unfavorable cervical mucous and endometrial changes. When used properly, failure rates are below 1%. In addition to providing effective contraception, their use is accompanied by numerous noncontraceptive health benefits (eg, reduced menstrual blood loss, less dysmenorrhea, less acne/ hirsutism, fewer ovarian cysts, fewer premenstrual symptoms, and fewer cases of ovarian and endometrial cancer). However, combination pills also cause undesirable side effects. Soon after their introduction, an increased risk for thromboembolic events and cardiovascular risk was observed (mainly among older women who smoked). In an effort to reduce this risk, newer formulations with lower estrogen dose and new progestogen components were introduced.

Estrogen influences hemostasis by increasing the levels of clotting factors (VII, VIII, X, fibrinogen) and plasminogen, lowering antithrombin III and protein S levels, and altering activated protein C (APC) resistance. APC induces decreased factor V activity. With increased APC resistance, this inhibition is not in effect and the coagulation cascade proceeds. The net effect of combination pills is a procoagulant effect. The overall hemostatic effect is partly caused by estrogen (found in all pills but at different doses) and partly caused by the type of progestogen. Various progestogens induce various changes. Third generation pills that contain desogestrel were found to induce a greater increase in factor II and VII levels.[2] Tans and colleagues[3] reported increased APC resistance with desogestrel-containing pills, and others have confirmed this finding.[4]

In addition to differences among the pills, other factors may also be responsible for venous thromboembolism (VTE) risk. Several known risk factors associated with VTE are age, weight, family history, and smoking. Not all studies control for the known risk factors, and unknown factors may remain unbalanced between cases and controls. In general, one would associate newer preparations with an improved risk profile; therefore, new users and higher-risk patients are more likely to be prescribed the new third-generation pills. This also would influence the number of VTE events.


This large case-control study compared VTE risk (confirmed by Doppler, spiral computed tomography, ventilation perfusion scan, and angiography) associated with levonorgestrel-containing pills (2nd-generation) and 3rd-generation pills (desogestrel, gestodene), as well as with newer combinations (drospirenone, cyproterone acetate). The study included 1524 cases and 1760 controls. Relative risks were adjusted for age, estrogen dose, body mass index, family history, and smoking status. Overall OCP use was associated with a fivefold increased risk for VTE. The risk was lower with levonorgestrel-containing pills (odds ratio [OR] 3.6; 95% CI 2.9-4.6) when compared with pills containing desogestrel (OR 7.3; 95% CI 5.3-10), gestodene (OR 5.6; 95% CI 3.7-8.4), drospirenone (OR 6.3; 95% CI 2.9-13.7), or cyproterone acetate (OR 6.8; 95% CI 4.7-10). A trend for lower risk with lower ethinyl estradiol doses was present. The risk for VTE was highest in the first 3 months of use (OR 12.6; 95% CI 7.1-22.4), but even at 1 year a fivefold increased risk was detected. Among those who used the pill for at least 2 years, the risk was still elevated with desogestrel- (OR 1.9; 95% CI 1.3-2.9) and gestodene- (OR 1.5; 95% CI 0.9-2.6) containing pills when compared with levonorgestrel-containing pills.


This study confirms findings of previous studies that also reported an increased risk for VTE with third-generation OCPs compared with second-generation pills. The overall risk for VTE is 1 in 10,000 in the general population. The risk is elevated to 4-6 in 10,000 with OCP use, whereas with pregnancy, the risk is 8-10 in 10,000. These data also show that the risk for VTE is lower with any type of pill when compared with the risk during pregnancy. This is an important comparison because the pill is primarily used to prevent pregnancy. Various pills may be associated with various VTE risk profiles, but other benefits (contraceptive and noncontraceptive) must be considered as well to get an overall picture. Not all pills are tolerated well by all patients, and pills are associated with different bleeding/side effect profiles. Certain pills have a more pronounced effect on premenstrual symptoms or androgen complaints (drospirenone-containing pills).[5] Obviously one should use the pill with the best safety parameters, but the decision has to be based on an overall profile. If the patient discontinues the pill because of side effects (or lack of beneficial effects) and becomes pregnant, her risk for VTE increases or she is exposed to a different set of risks if the pregnancy is terminated. Although VTE risk does need to be considered when a pill is prescribed, the overall response to the pill and satisfaction with the pill is important as well, because we want the patient to comply with contraceptive use.



Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: