In December 2011 we reported on three current studies concerning the risk of venous thromboembolism (VTE) when taking contraceptives containing drospirenone (e.g. YASMIN; a-t 2011; 42: 109). In one of the studies, a cohort study (1) conducted with the cooperation of the American Food and Drug Administration (FDA), the VTE risk from the intravaginal ring, NUVARING* (a-t 2003; 34: 17-8) and the hormone patch EVRA (a-t 2003; 34: 75-6) is calculated from the data of four major US health insurance companies, and compared with the risk from a group of oral contraceptives commonly used in the USA. The comparison preparations contain 20 µg to 30 µg ethinylestradiol (EE) with levonorgestrel (e.g. LEIOS), norethisterone (e.g. EVE 20) or norgestimate (e.g. CILEST). The study encompasses not quite 24,000 women years of use of the vaginal ring, among which 25 VTE have been identified. In contrast to the comparison contraceptives (617,265 women years, 389 VTE) the ring significantly increases the risk of venous thromboembolism (relative risk [RR] 1.56; 95% confidence interval [CI] 1.02-2.37) (1). Consequently there are data available for the first time on the VTE risk from a product that has been on the market in Germany for almost nine years. Since the ring contains the progestin etonogestrel, the active metabolite of desogestrel (in MARVELON, amongst other products), which is known to increase the risk of thromboembolic complications in third generation contraceptives, the result is admittedly not unexpected.

The hormone patch EVRA (67,865 women years, 67 VTE) also increases the risk of VTE according to this study (RR 1.55; 95% CI 1.17-2.07) (1). The transdermal contraceptive contains the progestin norelgestromin, the active metabolite of norgestimate, which in many metabolic and coagulation properties resembles the third generation progestins, desogestrel and gestodene (in FEMOVAN and other products) (2). Until now there have only been data concerning the patch from post-marketing studies sponsored by the US parent company Johnson & Johnson. Results from them are inconsistent: compared with oral contraceptives with the parent substance norgestimate, the thrombogenicity of which is however similarly unclear,** it significantly increased the risk of VTE in one of two case-control studies that were both updated several times (3,4) (Odds ratio [OR] 2.0; 95% CI 1.2-3.3) (3). Compared with second generation pills with levonorgestrel, the risk tends to increase according to one of two jointly published studies (OR 2.0; 95% CI 0.9-4.1) (3; p = 0.07) (5,6). Following an extended evaluation of this case-control study which was only published on the Internet, the difference becomes significant if users are included in the analysis who have known risk factors for venous thromboembolism (OR 1.9; 95% CI 1.1-3.3) (6). According to assessment by the FDA, the result of the current study supports the suspected increased risk of VTE when using the hormone patch (1).

All data known to date concern the American patch that contains 0.75 mg EE. The European product contains 20% less estrogen but is bioequivalent with the higher dose patch (10).*** The safety considerations can therefore be transferred, in our opinion, to the patch available in Germany.

In the light of the discussions concerning the risk of thromboembolism from contraceptives containing drospirenone, in the Bulletin zur Arzneimittelsicherheit (bulletin on drug safety) jointly issued by the German Federal Institute for Drugs and Medical Devices (BfArM) and the Paul-Ehrlich Institute, staff of the BfArM called for giving preference to combined oral contraceptives with a known low risk of VTE, particularly in first-time users and young women (under 30) (12). In practice this means applying a reserve status not only to third generation pills containing desogestrel, drospirenone or gestodene, the vaginal ring and the hormone patch EVRA, but also to all combined hormonal contraceptives, the thrombogenicity of which is unknown. According to a summary by the BfArM there are at the moment official recommendations in the form of guidelines, in five European countries, for choosing in the first instance a second generation pill, e.g. with levonorgestrel, as a combined oral contraceptive (12).

According to a current epidemiological study, the hormone patch EVRA and the intravaginal ring NUVARING increase the risk of venous thromboembolism compared with oral contraceptives commonly used in the USA, for example with levonorgestrel (e.g. LEIOS).

For the patch, these data support indications from previous studies. For the vaginal ring there were hitherto no relevant data.

No advantages are confirmed for either of these two contraceptives. The drug of choice is therefore a low dose oral contraceptive with a second generation progestin such as levonorgestrel.

1

FDA: Combined Hormonal Contraceptives (CHCs) and the Risk of Cardiovascular Disease Endpoints; final report; published online on 27 Oct. 2011
http://www.fda.gov/downloads/Drugs/DrugSafety/UCM277384.pdf

2

FDA: Medical Review ORTHO EVRA, as at 2001; can be found on
http://www.accessdata.fda.gov/drugsatfda_docs/nda/2001/021-180_Ortho.cfm

3

DORE, D.D. et al.: Contraception 2010; 81: 408-13

4

JICK, S.S. et al.: Contraception 2010; 81: 452-3

5

JICK, S.S. et al.: Contraception 2010; 81: 16-21

6

Johnson & Johnson: study summary, undated; can be found on
http://www.clinicaltrials.gov/ct2/show/NCT00511784

7

Janssen-Cilag: SPC CILEST, as at Nov. 2010

8

Janssen-Cilag: SPC PRAMINO, as at May 2010

9

LIDEGAARD, Ø. et al.: BMJ 2011; 343: d6423 (15 pages)

10

Prescrire Int. 2007; 16: 107

11

EMA: Europ. Assessment report (EPAR) EVRA, as at 2010
http://www.ema.europa.eu/ema/index.jsp?curl=pages/medicines/human/medicines/000410/human_med_000775.jsp&mid=WC0b01ac058001d124

12

PANTKE, E. et al.: Bulletin zur Arzneimittelsicherheit (bulletin on drug safety) 2011; No 4: 3-7, can be found on http://www.bfarm.de

*

Newly available at an almost identical price also as CIRCLET from Varipharm, a subsidiary of the NUVARING supplier, MSD.

**

A third generation progestin the risk of venous thromboembolism from which is "not known" according to the summary of product characteristics (SPC; 7,8). In a large Danish cohort study there was no evidence of any increased risk compared with pills containing levonorgestrel (rate ratio 1.18; 95% CI 0.86-1.62) (9).

***

A slow release patch with 0.75 mg EE was originally also licensed in Europe: the decisive clinical studies on which marketing authorisation was based also in the USA were conducted with this patch. Even before it was introduced onto the market however the production site changed. To achieve bioequivalence, the estrogen content of the new patch had to be reduced to 0.6 mg EE (10). For both preparations, release of the active substance was indicated for the risk assessment as 20 µg EE per 24 hours (a-t 2005; 36: 114), based on serum concentrations measured. Since 2009, the summary of product characteristics has however instead alleged that the daily amount of EE released is 34 µg. This is the amount of estrogen released from the patch onto the skin. The qualitative and quantitative composition of the patch has not changed (11).