Abstract
Objective This article estimates the interaction between types of combined hormonal contraception
(CHC) and factor V Leiden (FVL) mutation on the risk of venous thrombosis event (VTE).
Subjects and Methods All premenopausal women with first incident VTE who were referred to our unit (Paris,
France) between 2000 and 2009 were included in this case-only study. Differences in
interactions by progestin type were assessed on a multiplicative scale, assuming the
independence of genotype and prescription of type of CHC.
Results Among 2,613 women with VTE, 15.9% had a FVL and 69% used CHC. The interaction between
CHC use and presence of FVL on VTE risk was statistically significant (1.37, 1.06–1.77
95% confidence interval [CI]). This interaction appeared higher for drospirenone 1.99
(1.18–3.38 95% CI) (n = 98) or cyproterone acetate users 1.71 (1.20–2.45 95% CI) (n = 326), but not significant for 1st or 2nd or norgestimate CHC users. The results
were similar when excluding women with a family history of VTE or with provoked VTE.
In this sub-group of women, these interactions appeared higher for third generation,
cyproterone acetate and drospirenone CHC users as compared with 1st or 2nd or norgestimate
CHC users (odds ratio [OR], 1.68 [1.04–2.70; 95% CI], 2.91 [1.71–4.95 95% CI], 3.22
[1.54–6.73 95% CI], respectively).
Conclusion Our results show that the interaction between FVL and CHC use differ by progestin
type, which is higher in CHC containing third-generation progestin, drospirenone or
cyproterone acetate, compared with second generation. Further studies are needed to
assess the cost-effectiveness of biological thrombophilia screening (FVL) when such
prescription of CHC is planned.
Keywords
venous thrombosis - combined contraception - factor V Leiden mutation - epidemiology
