The risk of deep venous thrombosis with oral contraceptives containing drospirenone

BMJ 2011; 342 doi: http://dx.doi.org/10.1136/bmj.d2519 (Published 21 April 2011) Cite this as: BMJ 2011;342:d2519
  1. Nick Dunn, senior lecturer in medical education
  1. 1Faculty of Medicine, University of Southampton, Southampton SO17 1BJ, UK
  1. nick.dunn{at}soton.ac.uk

Data are inconclusive, but alternatives may be preferable unless specifically contraindicated

Drospirenone is a synthetic progestogen, derived from the aldosterone blocker spironolactone. It is structurally related to progesterone and has antiandrogenic and anti-mineralocorticoid properties. It has been marketed in oral contraceptive preparations, combined with 30 mg of ethinylestradiol, since about 2000. The contraceptive efficacy of such pills is undoubted, but their adverse effect profile is the subject of controversy, particularly the potential to cause venous thromboembolism. Two linked studies add to this debate (doi:10.1136/bmj.d2139; doi:10.1136/bmj.d2151).1 2 Both are observational database studies and produce remarkably similar results, which support another two studies published in the BMJ in 2009.3 4 All of these studies suggest that drospirenone increases the risk of venous thromboembolism compared with the progestogen, levonorgestrel. However, at least two other studies disagree.5 6

The two new papers analyse data taken from the UK General Practice Research Database (GPRD) and Pharmetrics, a medical claims company in the United States. They both show a twofold to threefold increased risk of venous thromboembolism in women taking oral contraceptives containing drospirenone rather than levonorgestrel. This compares with 1.5-fold to twofold increased risk in the earlier BMJ studies.3 4

Although the new papers were well planned and executed, they have weaknesses. The GPRD study had comparatively low numbers (61 cases and 215 controls), and there were 27 cases of deep venous thrombosis and 34 cases of pulmonary embolism (a case mix that does not reflect clinical practice).1 Data were missing (for example on obesity, a risk factor for venous thromboembolism, and thus a potential confounder) and had to be imputed into the mathematical models; in addition, some subgroup analyses are not justified because of the low case numbers. In the Pharmetrics study, data on obesity were incomplete—height and weight measurements were not available,2 and the data on duration of exposure to all contraceptives may not be reliable. This is important because first time users with short duration of exposure are at increased risk of venous thromboembolism compared with those who have used the drug for a long time.1

Faults can be found with any observational study, and those that use routinely collected data are more prone to error than those that use data collected specifically for the study. Nonetheless, the concurrence in risk estimates between the papers is compelling. A random effects meta-analysis of results from studies with suitable data1 2 4 5 found an increased risk (although not statistically significant) for drospirenone preparations compared with levonorgestrel preparations (odds ratio 1.75, 95% confidence interval 0.84 to 2.67). There is also some supporting biological evidence that drospirenone can precipitate unfavourable changes in blood clotting factors.7

Prescribing practice should reflect available evidence on benefits and risks. Drospirenone contraceptives are said to improve acne, but there is little evidence to support their superiority over other low dose oral contraceptives.8 They may be therapeutic for women who have premenstrual dysphoric disorder—a severe form of premenstrual tension that affects 3-8% of menstruating women, but the causes of this condition are multifactorial, and other drugs are available (for example, antidepressants).

When prescribing oral contraceptives, the patient’s individual risk-benefit profile should be carefully considered, because such patients are often young, healthy, and may take the chosen pill for a long time. Of note, none of the published studies shows a significantly lower risk of venous thromboembolism for drospirenone than for levonorgestrel. Although the evidence for increased risk from drospirenone is inconclusive and the absolute risk of venous thromboembolism for women on low dose oral contraceptives is small (about 20-30/100 000 women years of use), it seems sensible to prescribe an oral contraceptive with a well known favourable safety profile (one that contains levonorgestrel) unless there is a persistent reason to use another type. The number of patients with such a specific indication for drospirenone is likely to be small.


Cite this as: BMJ 2011;342:d2519


  • Research, doi:10.1136/bmj.d2139
  • Research, doi:10.1136/bmj.d2151
  • Competing interests: The author has completed the Unified Competing Interest form at http://www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declares: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.

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