Danish Analysis Updates Stroke and MI Risks of Contemporary Contraceptives

Combined estrogen-progestin as well as progestin-only contraceptive products are associated with a slightly increased risk of ischemic stroke and MI, a contemporary Danish study of more than 2 million women suggests.

Compared with nonusers, there were 21 extra ischemic strokes and 10 extra MIs per 100,000 person-years in current users of combined estrogen-progestin contraceptives. Extrapolating that further, the risk would equate to one additional ischemic stroke for every 4,760 women on combined oral contraceptives for 1 year compared with nonusers and one additional MI for every 10,000 women per year.

“Considering the low baseline risks of ischemic stroke and myocardial infarction among women of reproductive age, the increased relative arterial thrombotic risk observed with hormonal contraception use translated to low absolute excess risks,” say Harman Yonis, MD (Nordsjællands Hospital, Hillerød, Denmark), and colleagues in the study, which was published recently in the BMJ.

The greatest risks were associated with estrogen-containing contraceptives, with no indication that duration of use was a factor in the risk of experiencing an arterial thrombotic event. There also were no differences by age in the risk of an event for users of combined oral contraceptives. The only commercially available contraceptive option that was not associated with any increased risk of stroke or MI was the long-acting levonorgestrel-releasing intrauterine device (IUD).

The study adds to a Cochrane meta-analysis of 24 studies and a Danish observational study from 2012, both of which indicated increased risk of stroke and MI with use of various combined oral contraceptives.

Commenting for TCTMD, Michael C. Honigberg, MD (Massachusetts General Hospital, Boston, and the Broad Institute of MIT and Harvard, Cambridge), said while a small rise in risk associated with systemic estrogens is not surprising,  the signal for increased risk with some of the progestin-only contraceptive options is somewhat unexpected.

“We often think in practice that maybe those would be a little bit safer for patients in whom we're worried about their cardiovascular risk,” he said. “The authors acknowledged there could be some residual confounding from preferentially prescribing some of the progestin-only options to patients who might have been higher risk, but in a methodologically rigorous study like this one, a signal for excess risk is still interesting.”

Doses and Individual Contraceptive Products

The nationwide prospective cohort study tracked outcomes in 2,025,691 women aged 15-49 years living in Denmark between 1996-2021.

Looking at the increased risk of events by dose, current use of combined oral contraceptives containing 20 μg of ethinyl estradiol was associated with a 1.9-fold increased risk for ischemic stroke and a 1.6-fold increased risk for MI compared with nonuse, while those containing 30-40 μg were associated with 2.0-fold and 2.1-fold increased risks, respectively.

For progestin-only pill users, the rate of extra ischemic strokes per 100,00 person-years was 15 and the rate of extra MIs was 4, the researchers found. The standardized ischemic stroke rate per 100,000 person-years was 39 with use of combined oral contraception, and the standardized MI rate was 18 per 100,000 person-years.

It validates the way that a lot of us are already practicing. We typically recommend those IUDs that secrete a progestin. Michael C. Honigberg

Nonoral formulations of combined contraceptives also were associated with increased risk compared with no contraceptive use, with a rate of extra ischemic strokes per 100,00 person-years of 28 for the vaginal ring and 17 for the vaginal patch. The vaginal ring also was associated with increased risk of MI.

Among all the contraceptives included in the study, the highest risks were associated with combined oral pills containing third- or fourth-generation progestins, which conferred a risk equivalent to 34 additional ischemic strokes and 14 additional MIs per 100,00 person-years.

Yonis and colleagues say while the data come from high-quality national registries, they cannot dismiss the potential for residual confounding, especially since individual exposure to contraceptive products and the reason or preference for use of particular hormonal agents could not be controlled for in the observational study.

Public Health Considerations

An important takeaway that Honigberg said aligns with current practice is that IUDs are highly safe and should be encouraged in younger patients who are at increased cardiovascular risk or who have established cardiovascular disease.

“It validates the way that a lot of us are already practicing. We typically recommend those IUDs that secrete a progestin like the one that they examined in this paper,” he added. “The hormonal release is very local, there's very low systemic absorption of the progestin, and they are less likely to cause significant bleeding, which can be an issue in our cardiovascular patients if they're on antiplatelet therapy or anticoagulation.”

In an accompanying editorial, Therese Johansson, PhD (Royal Institute of Technology, Stockholm, Sweden), notes that the findings regarding increased risk of stroke and MI with combined nonoral devices like the vaginal ring and the patch add to what is currently known about the risks of contemporary hormonal contraceptives and highlight the need for targeted public health interventions.

“Educational campaigns should focus on increasing awareness of the potential risks associated with various contraceptive methods, thereby enabling women to make informed choices,” Johansson writes. She further notes that while the Nordic women in the study had access to subsidized contraception, that is not the case for many women throughout the world.

Honigberg stressed, too, that the researchers had a long exclusion list, which meant that the overall population studied was probably healthier than the average woman on contraceptives.

“It does beg the question that if we’re seeing this signal with a small magnitude of increased risk in a pretty healthy population overall, the magnitude of risk increase might actually be different in a population that's at higher overall cardiovascular risk like you would find in the US,” he said.