Is an EMA review on hormonal contraception and thrombosis needed?

BMJ 2013; 346 doi: https://doi.org/10.1136/bmj.f1464 (Published 07 March 2013) Cite this as: BMJ 2013;346:f1464
  1. Frans M Helmerhorst, professor in clinical epidemiology of fertility ,
  2. Frits R Rosendaal, professor in clinical epidemiology
  1. 1Leiden University Medical Center, 2300 RC Leiden, Netherlands
  1. F.M.Helmerhorst{at}LUMC.nl

Sufficient evidence exists to recommend the second generation pill with the lowest tolerable oestrogen dose for all indications

Four recently reported deaths in women using the Diane-35 contraceptive and a lawsuit against the French drug authority (L’Agence Nationale de Sécurité du Médicament) after it banned Diane-35 led the authority to request that the European Medicines Agency (EMA) review the safety of combined oral contraceptives.1 2 Of particular concern were third and fourth generation drugs, including Diane-35 and its generics. This review was granted on 7 February 2013.3 4 The Dutch College for the Evaluation of Medicines (Dutch “EMA”) decided that a new study on Diane-35 was in order.

Most oral contraceptives are combination preparations, containing a progestogen, to prevent ovulation, and an oestrogen to prevent breakthrough bleeding. Since the introduction of the pill, the oestrogen dose, in the form of ethinylestradiol, has been reduced (heavy v light pills) and the type of progestogen has changed several times (indicating the generation). The categorisation is imprecise and incomplete. For example, cyproterone acetate, the progestogen in Diane-35, does not belong to a generation. Furthermore, the categorisation assumes that all side effects of oral contraceptives are class effects. In our recent network meta-analysis of all combined oral contraceptives (unpublished data), we found that the risk of venous thrombosis depended on the dose of oestrogen and the type of progestogen, even within generations.

Many studies have shown that oral contraceptive users have an increased risk of venous thrombosis (deep vein thrombosis, pulmonary embolism) and arterial thrombosis.5 6 Venous thrombosis is more common than arterial thrombosis, but in young women the incidence of these side effects is low. Even the “safest” oral contraceptive increases the risk of venous thrombosis, however, and the risk is twice as high for oral contraceptives containing a third generation progestogen, drospirenone (sometimes called fourth generation), or cyproterone acetate.5 This knowledge is not new—the increased risk for pills containing third generation progestogens, cyproterone acetate, and drospirenone has been known since 1995, 2001,7 and 2003, respectively.8

The EMA’s public report at the beginning of the review states that Diane “works by blocking the effects of a class of hormones called androgens,”3 and that this is responsible for its supposed benefits on acne and hirsutism. However, as early as 2004 (and in three updates) a systematic review concluded that all types of monophasic combined oral contraceptives are effective against acne.9

All combined oral contraceptives are equally effective in preventing pregnancy. Their side effects (such as weight gain10) and benefits (in terms of acne and hirsutism) are also similar, so the only rational strategy is to use the safest one with regard to venous thrombosis. The common arguments that the risk of thrombosis is low or that the risk of thrombosis during pregnancy is higher than when using oral contraceptives are flawed. Millions of women in Europe use oral contraceptives, so use of the pill with the best safety profile in terms of thrombosis would probably prevent thousands of thrombotic events and hundreds of deaths a year. Because the pill with the safest thrombosis profile is as effective at preventing pregnancy as the less safe ones, the risk of thrombosis in pregnancy is irrelevant in the choice of oral contraceptive. The safest oral contraceptive is one that contains the lowest tolerable dose of ethinylestradiol (lowest dose that prevents breakthrough bleeding—30 μg11) together with the second generation progestogen, levonorgestrel.

Sufficient evidence is already available on which clinicians and regulatory agencies can base their decisions, so lengthy evaluations, let alone new studies, are not needed.

In his 2011 BMJ editorial, Nick Dunn recommended prescribing an oral contraceptive that contains levonorgestrel unless “there is a persistent reason to use another type.”12 Because oral contraceptives containing levonorgestrel and the lowest tolerable dose of oestrogen are also adequate for the treatment of acne or hirsutism, we can see no reason to use another type. Third and fourth generation oral contraceptives are widely overprescribed.


Cite this as: BMJ 2013;346:f1464


  • Competing interests: We have read and understood the BMJ Group policy on declaration of interests and have no relevant interests to declare.

  • Provenance and peer review: Commissioned; not externally peer reviewed.


View Abstract

Sign in

Log in through your institution

Free trial

Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
Sign up for a free trial