Cyproterone acetate with ethinylestradiol as a risk factor for venous thromboembolism: an epidemiological evaluation.

OBJECTIVE

To review the incidence of venous thromboembolism (VTE) in association with combined cyproterone acetate and ethinylestradiol (CPA/EE) use, compared to the incidence of VTE in association with conventional oral contraceptive (OC) use, in women aged 14 to 49 years.

METHODS

A review consistent with the guidelines of "best evidence synthesis" was conducted on 6 peer-reviewed, controlled epidemiological studies reporting both outcome numerators and exposure denominators. Relative risks as the estimate of the association between CPA/EE and the clinical outcome of VTE were set aside in favour of absolute incidence rates, rate differences, or attributable risks.

RESULTS

The range of absolute incidence rates of VTE among CPA/EE users varied from 1.2 to 9.9 per 10 000 women-years (WYs). All other marketed conventional OCs were the reference. In 6 studies conducted in 3 European countries, CPA/EE exhibited an attributable risk in women not higher than 0.04% when compared to approved conventional OCs. Except for 1 study exhibiting a small statistically significant benefit, there was no difference between the incidence rates of VTE among CPA/EE users and those among conventional OC users. VTE as an adverse event for CPA/EE was either rare or very rare, according to the criteria set by the World Health Organization/ Council for International Organizations of Medical Sciences. No deaths were reported in the studies reviewed.

CONCLUSIONS

The evidence in the worldwide literature does not demonstrate an excess of health-threatening harm for CPA/EE. A best-evidence synthesis, using sound epidemiological methods, showed clearly that the risks of VTE among CPA/EE users do not exceed the risks of VTE among conventional OC users.