Oral contraceptives and venous thromboembolism: old questions revisited.

©FSRH J Fam Plann Reprod Health Care 2009: 35(4) Background Two studies recently published in the British Medical Journal, a retrospective cohort study using information from the Danish registries1 and the MEGA case-control study for which cases were identified via specialised anticoagulation clinics in The Netherlands,2 revisit the question of whether so-called ‘third generation’ combined oral contraceptives (COCs) containing the progestogens, desogestrel and gestodene, increase the risk of venous thromboembolism (VTE) more than so-called ‘second generation’ COCs as represented by products containing levonorgestrel. It now seems that this old dispute is being extended to other progestogens. These studies found an increased relative risk for desogestrel and gestodene, as well as for cyproterone acetate and drospirenone, compared to levonorgestrel. In addition, the MEGA casecontrol study found an increased relative risk for lynestrenol and norgestimate. For the sake of brevity this commentary focuses on drospirenone, which is currently the most widely used progestogen. Many of the following arguments, however, apply also to other progestogens. An increased risk of VTE has been linked to COCs since the 1960s, and has been associated with the estrogen dose. In 1995, World Health Organization (WHO) investigators alerted the public that desogestrel and gestodene carry a higher risk of VTE than levonorgestrel. In the ensuing years several studies were published, some of which supported and some of which conflicted with the WHO results. The heated scientific debate led to a widespread ‘Pill-scare’ propagated in the media, and ultimately occupied the law courts. In 2002, the High Court of Justice in London came to the conclusion that “there is not as a matter of probability any increased relative risk of VTE carried by any of the third generation oral contraceptives ... as compared with second generation products containing levonorgestrel”. A legal decision, however, cannot be a substitute for scientific judgment. Even 14 years after the publication of the WHO study, whether the increased relative risk for desogestrel and gestodene is real or the result of bias and confounding remains controversial. In addition to the above-mentioned Danish cohort study and the MEGA case-control study, three other large studies3–5 have investigated the VTE risk of drospirenone in comparison to levonorgestrel or other progestogens. Even a cursory comparison shows that the results of all of these studies, although using different methodologies, are not fundamentally different.