Hormonal contraceptives with ethinyl estradiol were associated with increased thrombotic stroke and MI
Andrews, J.
Annals of Internal Medicine 157(8): JC4
2012
ISSN/ISBN: 0003-4819 DOI: 10.7326/0003-4819-157-8-201210160-02009Document Number: 563701
Are hormonal contraceptives associated with increased risk for thrombotic stroke or myocardial infarction (MI)? DesignPopulation-based cohort study with linkage of 4 national registries between 1995 and 2009. Denmark. 1 626 158 women 15 to 49 years of age. Exclusion criteria included past venous or arterial thrombotic event; past gynecologic, breast, lung, abdominal, or hematologic cancer; or past hysterectomy, sterilization, bilateral oophorectomy, or 2 unilateral oophorectomies. Data were censored during pregnancies.Risk factorsFilled prescriptions for hormonal contraceptives. Analyses were adjusted for age; education (level and duration); calendar year; and use of medications for hypertension, heart disease, diabetes, and hyperlipidemia.OutcomesThrombotic stroke and MI, identified through diagnostic codes in the National Registry of Patients and the Register of Causes of Death.Main results3311 (0.2%) women had a thrombotic stroke, and 1725 (0.1%) had an MI during 14 251 063 woman-years of follow-up. Thrombotic stroke. Incidence in nonusers was 24 per 100 000 woman-years (2260 events). Levonorgestrel intrauterine devices (IUDs) were associated with reduced risk (adjusted relative risk [aRR] 0.73, 95% CI 0.54 to 0.98); progestin-only pills or implants were not associated with increased risk (aRR range 0.44 to 1.4). Two of three 20- g ethinyl estradiol formulations, six of seven 30- to 40- g formulations, and one of two 50- g formulations were associated with increased risk (aRR range 1.5 to 2.3) (Table). The vaginal ring was associated with increased risk (aRR 2.5, CI 1.4 to 4.4). MI. Incidence in nonusers was 13 per 100 000 woman-years (1228 events). Progestin-only interventions were not associated with increased risk (aRR range 0 to 2.1). One 20- g ethinyl estradiol formulation, five 30- to 40- g formulations, and both 50- g formulations were associated with increased risk (aRR range 1.6 to 4.3) (Table). The vaginal ring was not associated with increased risk. Data were insufficient to evaluate hormonal contraceptives given in a patch. Oral hormonal contraceptives containing ethinyl estradiol were associated with increased risk for thrombotic stroke and myocardial infarction.Association between prescription for hormonal contraceptives and thrombotic stroke or MI*OutcomesHormonal contraceptivesRR (95% CI)Thrombotic strokeEthinyl estradiol, 20 g, plus progestin 1.6 (1.4 to 1.9)Ethinyl estradiol, 30 to 40 g, plus progestin 1.8 (1.6 to 1.9)Ethinyl estradiol, 50 g, plus progestin2.0 (1.5 to 2.7)MIEthinyl estradiol, 20 g, plus progestin 1.4 (1.1 to 1.8)Ethinyl estradiol, 30 to 40 g, plus progestin 1.9 (1.7 to 2.1)Ethinyl estradiol, 50 g, plus progestin3.7 (2.8 to 5.0)*MI = myocardial infarction; RR = risk ratio; CI defined in Glossary. Adjusted for progestin type, age, and calendar year. Risk > 1.0 indicates increased risk associated with hormonal contraceptive use. P = 0.24 for dose trend in thrombotic stroke; P< 0.001 for dose trend in MI. Progestin = desogestrel, gestodene, or drospirenone. Progestin = norethindrone, levonorgestrel, norgestimate, desogestrel, gestodene, drospirenone, or cyproterone acetate.Progestin = norethindrone or levonorgestrel.