Hormonal contraception and risk of cardiovascular disease. An international perspective.

The most frequent major adverse effect of hormonal contraception is an increased risk of cardiovascular disease. The effect on the risk of venous thromboembolism (VTE), ischemic and hemorrhagic stroke, and myocardial infarction (MI) differs and is strongly influenced by smoking and the presence of other cardiovascular risks factors, such as hypertension and diabetes mellitus. The incidence of each disease rises with age and there are differences in risk among hormonal contraceptive preparations. This article provides a framework within which to assess the balance of risks among types of hormonal contraceptives according to individual circumstances. Data on cardiovascular disease mortality rates in women of reproductive age in different countries of the world were compiled from nationally reported statistics and supplemented where possible with reported disease incidence rates. Risks associated with current use of hormonal contraception were compiled from the most recent publications on the cardiovascular effects of steroid hormone contraception. These were combined to estimate the total cardiovascular incidence and mortality according to baseline cardiovascular risk and individual characteristics. Mortality rates for cardiovascular diseases are very low in women of reproductive age. Myocardial infarction mortality rates rise from < 0.4 per 100,000 woman-years at age 15-24 years to the range 2 to 7 per 100,000 woman-years at age 35-44 years. Stroke mortality rates similarly rise steeply with age and are between 3 and 5 times higher than those for MI. VTE mortality rates rise less steeply with age and are approximately one-tenth the MI mortality rates at age 35-44 years. The adverse effect of oral contraceptives (OC) on the risk of VTE is the most important contributor to the total number of cardiovascular cases attributable to OC use. The increased risk of stroke and MI dominate the patterns of mortality in OC users and smokers. The additional risks attributable to smoking are greater than the additional risks attributable to OC use. The risk attributable to OC use in women < 35 years of age is small, even if they smoke, but there are substantially increased risks in older women who both smoke and use OC. The additional mortality attributable to OC use can be reduced by screening users, as this results in lower relative risks of ischemic stroke and MI. Differences between OC types in the relative risk of VTE contribute little to the total cardiovascular mortality associated with OC use, even though the total number of cardiovascular events is increased. A potential reduction in the risk of MI with desogestrel and gestodene compared with levonorgestrel-containing OC would have little difference on overall mortality rates in women in their 20s and 30s, but may result in a net reduction in OC-attributable mortality in women aged 40-44 years who smoke. An overall quantification of the risks for different types of oral contraceptive users is necessary for an informed choice of contraceptive method, and any assessment of the balance of cardiovascular risks is complex. The model provides a tool to assess, at the level of the individual, the risks associated with use of different OC according to personal circumstances. It is important to consider the user's age and smoking status when determining OC attributable risks.