The dienogest-containing oral contraceptive VALETTE has been on the German market for 15 years. Adequate evaluable data on the risk of venous thromboembolism have not yet become available for the internationally little used combination with ethinylestradiol: Of two case-control studies, one was published (1) but only as a "summary" and the other not at all (a-t 2009; 40: 62-3). Now this second study has been published in full. It was intended to demonstrate the non-inferiority of VALETTE compared with levonorgestrel-containing oral contraceptives (FEMIGOA, etc.), defined as the maximal increase in thromboembolic risk with dienogest to less than double (upper limit of the confidence interval [CI] < 2). The calculated result meets this criterion (odds ratio 1.0; 95% CI 0.6-1.8) (2).

We believe, however, that the safety of the dienogest-containing combination has not been proven: The upper value of the confidence interval 1.8 means that the risk of venous thromboembolic events with dienogest can be up to 80% higher than with levonorgestrel - unacceptably high for establishing non-inferiority. Furthermore, the study sponsored by Bayer Schering - parent company of VALETTE supplier Jenapharm - was conducted by the Center for Epidemiology and Health Research (ZEG Berlin) which is based in Berlin and headed by two former Schering employees, including J. DINGER the lead author of the current publication. Studies from this Center regularly describe results favourable to the manufacturers of oral contraceptives. For example, the EURAS* study (3) on drospirenone-containing contraceptives such as YASMIN, which also found the risk of thromboembolism to be comparable to that of levonorgestrel-containing tablets, emerged from this institute (a-t 2007; 38: 95-6). Following the publication of two manufacturer-independent studies in which the hazard with drospirenone matches that of third-generation contraceptives, the prescribing information for YASMIN and PETIBELLE had to be accordingly amended (a t 2010; 41: 53-4). Of four case-control studies identified in a systematic review (4) of the thromboembolic risk of gestagen monopreparations, three suggested a numerically elevated risk, whereas the fourth - conducted by ZEG - found a reduced risk compared with non-use (5).

Industry-independent data on the thromboembolic risk of hormonal contraceptives with the newer gestagens such as dienogest are urgently needed. The problem is not new. As long ago as 2001 a meta-analysis of epidemiological studies on the risk of thrombosis from combined oral contraceptives had found that the increased risk from third-generation contraceptives such as desogestrel (MARVELON, generics) compared with second generation contraceptives (e.g. in levonorgestrel) as stated in studies (co)financed by pharmaceutical manufacturers was considerably less than in independent studies (a-t 2001; 32: 84).

The dienogest-containing oral contraceptive VALETTE is not inferior to levonorgestrel-containing tablets (FEMIGOA, etc.) in terms of the risk of venous thromboembolism according to a company-sponsored case-control study conducted by an institute with close ties to the industry. The calculated result, however, does not exclude a risk from dienogest which is up to 80% higher.

As long as independent studies are unavailable, we advise against the use of dienogest-containing contraceptives.

1

HEINEMANN, L.A. et al.: LAMSO 2001; 2: 6 pages (journal cannot be found anymore on the internet, -ed.)

2

DINGER, J. et al.: J. Fam. Plann. Reprod. Health Care 2010; 36: 123-9

3

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

4

BERGENDAL, A. et al.: Acta Obstet. Gynecol. Scand. 2009; 88: 261-6

5

HEINEMANN, L.A. et al.: Eur. J. Contracept. Reprod. Health Care 1999; 4: 67-73

*

EURAS = European Active Surveillance Study