Early epidemiologic studies22,23 and 24 indicated a four-fold to eight-fold increase in the risk of venous thromboembolism (VTE) among oral contraceptive users. As the estrogen content of COCs declined, reported risks of VTE fell to approximately three-fold, but the increased risk of VTE remains the greatest health threat that COCs pose.25,26,26a

A large World Health Organization (WHO) international multicenter hospital-based case-control study found an increased risk of VTE with low-dose COC use (Europe: odds ratio [OR] = 4.24, 95% CI 3.07–5.87; developing countries: OR = 3.02, 95% CI 2.28–4.00).27 Further analysis of the data from this study,28 as well as others,26 has suggested an additional two-fold increase in VTE risk among users of COCs containing desogestrel and gestodene, compared to users of COCs containing LN. The additional risk of VTE observed with the use of desogestrel- or gestodene-containing COCs is best explained by selection and prescribing bias.26 Women at highest risk of VTE, new starters, and women who have had complications on COCs in the past are most likely to have received COCs containing desogestrel or gestodene. In contrast, women who had been using COCs without complication for years comprise a population at low risk of complications, and were more likely to remain on older formulations. Whether selection and prescribing bias completely account for this observed effect is unknown.

Genetic predisposition also plays a role in modifying the risk of VTE associated with oral contraceptive use. The factor V
Leiden mutation is a point mutation that results in resistance to the anticoagulant effects of activated protein C. The factor V Leiden mutation is estimated to affect 4.4% of Europeans and 0.6% of Asians, and is extremely rare (or nonexistent) in populations from Africa and Southeast Asia.29 Women with the factor V Leiden mutation have an eight-fold increased risk of VTE (RR = 7.9, 95% CI 3.2–19.4), compared to women without the mutation.30 When a woman with the factor V Leiden mutation uses COCs, her baseline risk of VTE increases four-fold, and her overall risk of VTE is 35 times greater (RR = 34.7, 95% CI 7.8–154) than for women without the mutation who are not using COCs.30 Screening the general population for the factor V Leiden mutation is currently not recommended; however, women who are known to have the mutation or a strong family history of thromboembolic disease should avoid COCs. Other thrombophilic disorders, such as protein C or S deficiency, also increase the risk of VTE with COC use.