Hormonal Contraceptives Tied to Strokes, Heart Attacks in Real-World Data

—
Absolute risk remains low, Danish study found, and IUDs stood out for their relative safety

by
Nicole Lou, Senior Staff Writer, MedPage Today
February 12, 2025

Various hormonal contraceptives were linked to incident cardiovascular disease in an analysis of Denmark’s nationwide records.

From a pool of over 2 million women of reproductive age, users of most methods of hormonal contraception — including progestin-only products and non-oral contraceptives — had small but significantly increased risks of ischemic stroke and in some cases myocardial infarction (MI) when compared with non-users. Only the progestin-only intrauterine device (IUD) group appeared more reassuring with regards to risk of ischemic stroke and MI, according to researchers led by Harman Yonis, MD, of Nordsjaellands Hospital and University of Copenhagen in Denmark, reporting in The BMJ.

“Our findings may suggest a dose-related association between progestin-only products and arterial thrombotic risk with the highest risk associated with implant use, then pills. Intrauterine devices, which cause the smallest increase in serum progestin level, were not associated with increased risk,” Yonis and colleagues wrote.

Additionally, the study supported the older literature in showing a prothrombotic effect to exogenous estrogens. The highest risk estimates for stroke and MI were linked to the estrogen-containing vaginal ring and patch — which both continuously release estrogen, bypassing first-pass metabolism, the authors explained.

“It is important to note that the absolute risk remains low. Nonetheless, these side effects are serious and given that approximately 248 million women use hormonal contraceptives daily, the results carry important implications,” commented Therese Johansson, PhD, of KTH Royal Institute of Technology in Stockholm, Sweden, in an accompanying editorial.

It had been controversial whether the established oral contraceptives carry arterial thrombotic risks. Whether the supposed risk extended to newer-formulation progestin-only pills and non-oral hormonal contraceptives had been even less clear.

Johansson noted the advancements of those hormonal contraceptives designed to contain less estrogen and provide more convenient delivery in recent years.

“Policymakers should prioritize making safer alternatives, such as the levonorgestrel-releasing intrauterine system for women with cardiovascular risk factors, both affordable and accessible, particularly in low-resource settings where cardiovascular risks are frequently underdiagnosed and untreated,” she suggested.

The IUDs also showed no excess risk of venous thromboembolism in a separate study, also released this week, that instead pointed to combined hormonal contraceptives, particularly those containing third-generation progestins, as being associated with higher risk.

For the present analysis on stroke and MI, Yonis’ group used Denmark’s prospective registers to identify women age 15-49 years who lived in the country from 1996 to 2021. Participants had a weighted median age of 34, a body mass index of 23, and 20% had university education.

Study eligibility criteria had included no history of arterial or venous thrombosis, antipsychotics use, cancer (except non-melanoma skin cancer), thrombophilia, liver disease, kidney disease, polycystic ovary syndrome, endometriosis, infertility treatment, hormone therapy use, oophorectomy, and hysterectomy.

Compared with non-users, increased risks of ischemic stroke and MI were seen with users of the following types of hormonal contraception:

• Combined oral estrogen-progestin contraception: increased risks of ischemic stroke (21 extra ischemic strokes per 100,000 person-years, adjusted rate ratio [RR] 2.0, 95% CI 1.9-2.2) and MI (10 extra MIs per 100,000 person-years, adjusted RR 2.0, 95% CI 1.7-2.2)
• Progestin-only pills: increased risks of ischemic stroke (15 extra ischemic strokes per 100,000 person-years, adjusted RR 1.6, 95% CI 1.3-2.0) and MI (four extra MIs per 100,000 person-years, adjusted RR 1.5, 95% CI 1.1-2.1)
• Combined vaginal ring: increased risks of ischemic stroke (28 extra ischemic strokes per 100,000 person-years, adjusted RR 2.4, 95% CI 1.5-3.7) and MI (41 extra MIs per 100,000 person-years, adjusted RR 3.8, 95% CI 2.0-7.3)
• Combined hormonal patch: increased risk of ischemic stroke (adjusted RR 3.4, 95% CI 1.3-9.1) but not MI
• Progestin-only implant: increased risk of ischemic stroke (adjusted RR 2.1, 95% CI 1.2-3.8); MI risk not calculated due to few events
• Progestin-only injection: numerical trend of more ischemic strokes (adjusted RR 1.8, 95% CI 0.8-4.4), MI risk not calculated due to few events

No increased risk was seen with the progestin-only IUD for ischemic stroke (adjusted RR 1.1, 95% CI 1.0-1.3) and MI (adjusted RR 1.1, 95% CI 0.9-1.3).

A prescription database provided information on prescriptions filled for the various hormonal contraceptives. These data were used as proxy for an estimate of actual contraceptive usage, leaving room in the study for misclassification of exposure to the contraceptives.

Additionally, Yonis and colleagues cautioned that their real-life observational study cannot exclude residual confounding and selection bias.