After years of controversial debate, it is now generally acknowledged that third generation oral contraceptives containing desogestrel (e.g. MARVELON) or gestodene (e.g. FEMOVAN) as progestins increase the risk of thromboembolism compared with oral contraceptives containing levonorgestrel (LEIOS, generics). Since a decision by the European Medicines Agency in 2001, the Summaries of Product Characteristics have had to include a warning to that effect (a-t 2001; 32: 104) (1). The thrombogenicity of medicinal products containing the relatively new progestin drospirenone (in PETIBELLE, YASMIN, YAZ) has so far been debatable. The procoagulatory effects of pills containing drospirenone are similar to those of combination pills with desogestrel and, in a non-controlled surveillance study (2), resulted in a very high absolute rate of venous thromboembolism. However, an observational study financed by the manufacturer Schering and conducted by an industry-oriented research body claims that the risk does not differ from that of second generation pills containing levonorgestrel (a-t 2007; 38: 95-6) (3). Two observational studies from the Netherlands and Denmark, which were independent of the manufacturer and were published simultaneously in August 2009 (4,5), now indicate that there is an increased thromboembolism risk with drospirenone contraceptives compared with second generation contraceptive pills.

In the Dutch case-control study, the data from six anticoagulation clinics and for a total of 3,300 non-pregnant women aged between 18 and 50 were analysed (4). The Danish cohort study, which is based on registery data, includes all non-pregnant Danish women aged between 15 and 49 who have no previous history of cardiovascular or malignant disease. This study covering the years 1995 to 2005 is a record of more than 130,000 women-years of the use of drospirenone contraceptives and includes 103 venous thromboembolisms. After adjustment, there is a significant relative risk increase to 1.64 (95% confidence interval [CI] 1.27-2.10) compared with levonorgestrel-containing contraceptives, which is similar to that found with third generation pills (desogestrel 1.82; 95% CI 1.49-2.22; gestodene 1.86; 95% CI 1.59-2.18) (5). The results of the much smaller Dutch case-control study point in the same direction. However, in this study the risk increase of 1.7 for drospirenone compared with levonorgestrel-containing oral contraceptives is not significant (95% CI 0.7-3.9) (4).

The German Federal Institute for Drugs and Medical Devices (BfArM) took the opportunity of these two studies to reassess the risk of thromboembolism with drospirenone-containing contraceptives together with the regulatory authorities of the other EU countries (6). A decision is awaited.

  According to two current, manufacturer-independent observational studies, the risk of thromboembolism for drospirenone-containing contraceptives (e.g. YAZ) is equivalent to that of 3rd generation contraceptives. We advise against prescribing.

  The safest combined oral contraceptives are levonorgestrel-containing pills with a low oestrogen dose (LEIOS, generics). They should be preferred.

1

EMEA: Combined oral contraceptives and venous thromboembolism, 28. Sept. 2001; http://www.emea.europa.eu/pdfs/human/regaffair/0220101en.pdf

2

PEARCE, H.M. et al.: Brit. J. Clin. Pharmacol. 2005; 60: 98-102

3

DINGER, J.C. et al.: Contraception 2007; 75: 344-54

4

van HYLCKAMA VLIEG, A. et al.: BMJ 2009; 339: b2921, 8 pages

5

LIDEGAARD, Ø. et al.: BMJ 2009; 339: b2890, 8 pages

6

BfArM: Risikoinformation vom 27. Aug. 2009; http://www.bfarm.de/cln_ 028/nn_828832/DE/Pharmakovigilanz/risikoinfo/kok-vte.html__nnn=true