Hormonal Contraception and Cardiovascular Risk: Context Matters

Hormonal contraception generates anxious headlines about clots and strokes, and the honest picture is more measured than the headlines and more nuanced than blanket reassurance. For most healthy women, the cardiovascular risk is low. For specific women, with specific factors, it rises enough to change the choice. The whole issue turns on context, and matching the method to the woman’s risk profile is the entire task.

The Mechanism

Estrogen is the primary driver of most cardiovascular risk associated with combined hormonal contraceptives, including combined oral contraceptives, the patch, and the vaginal ring. Synthetic estrogens, chiefly ethinyl estradiol used in most combined preparations, exert effects on the coagulation system that shift the blood toward a more clot-prone state. Specifically, estrogen increases hepatic production of several procoagulant factors, including fibrinogen, factors VII, VIII, and X, and von Willebrand factor, while simultaneously reducing levels of protein S, a natural anticoagulant. The net result is a measurable increase in thrombotic potential that can be detected in laboratory studies and that translates into elevated clinical risk in susceptible individuals.

For venous thromboembolism, which includes deep vein thrombosis and pulmonary embolism, this shift in coagulation balance is the primary mechanism. The elevated clotting tendency raises the probability that a clot will form in a vein, particularly in circumstances where blood flow is sluggish, such as during prolonged immobility on a long flight or during the postoperative period. Women with inherited thrombophilias, such as Factor V Leiden or prothrombin gene mutation, are particularly susceptible because their baseline coagulation system already favors clot formation, and the estrogen-driven shift amplifies a risk that was already elevated.

For arterial events, including ischemic stroke and myocardial infarction, the pathway is somewhat different and depends heavily on the condition of the arterial wall. Uncontrolled hypertension damages arterial walls and makes them more prone to plaque rupture and subsequent occlusive clot formation; adding an estrogen-driven procoagulant state raises the risk of arterial thrombotic events on that already-compromised background. Smoking damages arterial endothelium through oxidative stress and accelerates atherosclerotic plaque development, creating a similar substrate for arterial events that the procoagulant shift can then precipitate. The arterial risk from combined contraceptives is therefore dominated by the interaction with these background vascular conditions rather than being a primary effect of estrogen on a healthy artery.

The migraine with aura interaction represents a third distinct pathway. Migraine with aura is independently associated with higher stroke risk through mechanisms that include cortical spreading depression, platelet activation, endothelial dysfunction, and in some patients, the presence of patent foramen ovale. When estrogen-containing contraception adds its procoagulant effects to this already-elevated and multifactorial vascular risk, the compounding is more than additive. Current international guidelines categorize this combination as a contraindication, not merely a caution, because the evidence consistently shows that the risks outweigh the benefits.

Progestin-only methods, including the progestin-only pill, hormonal intrauterine devices, implants, and injectable progestins, do not carry the same venous or arterial risk profile. They lack the estrogen-driven coagulation shift, do not raise clotting factors in the same way, and do not appear to compound the stroke risk associated with migraine with aura. The venous thromboembolism risk of progestin-only methods is substantially lower than that of combined methods, which is why they represent the preferred hormonal option for women whose specific profile makes estrogen problematic.

What the Evidence Shows

The evidence on combined oral contraceptives and venous thromboembolism is among the most extensively studied topics in pharmacoepidemiology, with data accumulating over more than five decades. Large Danish registry studies, which have followed millions of women with complete records over many years, have been particularly informative. A foundational 2011 analysis by Lidegaard and colleagues, published in the New England Journal of Medicine, followed more than 1.6 million Danish women aged 15 to 49 for more than a decade and provided some of the most rigorous quantification of these risks.

In that Danish analysis, the background rate of venous thromboembolism in women not using hormonal contraception was approximately 2 to 3 events per 10,000 woman-years. Combined oral contraceptive users had rates ranging from approximately 6 to 9 per 10,000 woman-years depending on the type of progestin, representing a roughly threefold to sixfold relative increase over background. For context, pregnancy itself carries a venous thromboembolism risk of roughly 5 to 20 per 10,000 woman-years, and the postpartum period, particularly the first six weeks after delivery, carries even higher risk. These comparisons are not arguments to dismiss the contraceptive risk, but they are essential to absolute risk framing.

The same Danish data found that the type of progestin in combined preparations influenced the venous thromboembolism risk profile. Pills containing newer progestins such as desogestrel and gestodene carried approximately twice the venous risk of pills containing levonorgestrel, the older first-line progestin, though the mechanisms for this difference are still debated and arterial risks were less clearly differentiated between progestin types.

For stroke risk, the compounding with migraine with aura is the signal with the greatest clinical consequence. Case-control studies, including a frequently cited analysis by Chang and colleagues published in BMJ in 1999 and confirmed in subsequent work, found that women with migraine with aura who used oral contraceptives faced a relative risk of ischemic stroke approximately 6 to 8 times higher than women without either factor. Adding smoking raised the relative risk further, with some estimates exceeding thirteen. For the individual young woman with none of these compounding factors, absolute stroke risk remains in a low range. For the woman with all three, the absolute risk shifts into a range that clinically justifies the guideline classification as a contraindication.

The World Health Organization Medical Eligibility Criteria for Contraceptive Use, which is one of the most widely used international frameworks for contraceptive decision-making, classifies combinations of methods and medical conditions on a four-tier scale. Smoking at age 35 or older, uncontrolled hypertension with systolic blood pressure 160 or above, migraine with aura, personal history of venous thromboembolism, and personal history of ischemic stroke are all classified as Category 4 for combined estrogen-containing methods, meaning the risk is judged unacceptable and the method is contraindicated. Category 3, meaning risks generally outweigh benefits, includes migraine without aura in women over 35, adequately controlled hypertension, and several other conditions. Understanding which category applies to her specific profile is the basis for an individualized recommendation.

Estrogen dose matters for the magnitude of risk. Earlier high-dose formulations contained 50 micrograms or more of ethinyl estradiol and carried substantially higher absolute risks than current low-dose preparations, which typically contain 20 to 35 micrograms. The shift to lower doses has reduced absolute risks while preserving efficacy, but the relative risk increase over background and the compounding interactions with specific risk factors persist with current doses. Lower dose does not eliminate the mechanism; it attenuates the magnitude.

Progestin-Only Options: The Evidence Profile

For women in whom estrogen-containing contraception is contraindicated — those with migraine with aura, antiphospholipid antibody syndrome, uncontrolled hypertension, a personal history of deep vein thrombosis, or age over 35 with smoking — progestin-only methods are not a compromise. They are the clinically appropriate primary option, supported by an evidence profile that is distinct from combined oral contraceptives in the ways that matter for cardiovascular risk.

The progestin-only pill contains no estrogen and, as a result, does not carry the estrogen-driven upregulation of coagulation factors VII, VIII, X, fibrinogen, and von Willebrand factor that drives the VTE and arterial risk in combined methods. A systematic review by Dragoman and colleagues in the journal Contraception (2018, examining twelve studies) found no statistically significant increase in VTE risk for progestin-only pills compared to non-use. The trade-off is a strict three-hour daily efficacy window for the traditional progestin-only formulation, which some women find operationally difficult compared to the combined pill. Newer formulations with drospirenone, approved in some countries, extend this window to twenty-four hours. 4 / Promising

The levonorgestrel intrauterine system (LNG-IUS) delivers levonorgestrel directly to the uterine cavity with minimal systemic absorption. Serum levonorgestrel concentrations with the 52 mg LNG-IUS are approximately 150 pg/mL — far below the levels achieved with oral progestin-only pills. The Lidegaard group data, which documented elevated VTE risk with certain combined oral contraceptives, found VTE rates with the LNG-IUS that were comparable to those with the copper IUD, which carries no systemic hormonal exposure at all. For women with significant cardiovascular risk profiles, the LNG-IUS represents a highly effective option with negligible systemic hormonal load.

The etonogestrel subdermal implant achieves greater than 99% contraceptive efficacy, does not contain estrogen, and does not appear to increase thrombotic risk based on current evidence. Available pharmacovigilance and cohort data do not show a VTE signal comparable to combined methods.

Depot medroxyprogesterone acetate (DMPA, injectable) has shown modest increases in blood pressure in some studies, and for women with significant cardiovascular risk profiles, current guidelines generally place it in a lower-preference category compared to the LNG-IUS or implant, though it remains an option where other methods are not acceptable or accessible.

For women with antiphospholipid antibody syndrome — a high-thrombosis-risk autoimmune condition — the copper IUD is frequently the recommended first-line option, as all hormonal methods carry theoretical procoagulant considerations in this specific population. This is a case where the contraception conversation requires direct input from a rheumatologist or specialist familiar with the thrombosis risk profile.

What to Do This Week

1. Before any contraception discussion, know your relevant risk factors: whether you smoke and your age, your blood pressure and whether it is well-controlled, whether your migraines include aura, and whether you have any personal or family history of blood clots. These are the factors that change which methods are appropriate for you specifically. Arriving at the conversation with this information allows for a genuinely individualized discussion rather than a generic prescription visit.

2. If you have migraine with aura, state this explicitly when discussing contraception. Do not assume your clinician has linked your migraine history to the contraception decision. Ask directly whether estrogen-containing methods are appropriate given your migraine type, and ask about progestin-only or non-hormonal alternatives. Both categories are highly effective and do not carry the compounded stroke risk.

3. If you are over 35 and smoke, and currently use estrogen-containing contraception, raise this with your clinician at your next visit. The combination is classified as an unacceptable risk in major international guidelines, and either smoking cessation, a method switch, or both is warranted. This is not a minor concern to defer.

4. If your blood pressure is elevated or you do not know your current blood pressure, find out. Know your blood pressure number before a contraception conversation if you can; the specific value matters for the risk categorization and changes the recommendation for estrogen-containing methods.

5. Treat the choice as an individualized medical decision, not a yes-or-no verdict on hormonal contraception as a category. Most women can safely use hormonal methods, including combined estrogen-containing preparations. Those with specific risk factors have effective alternatives that carry a more favorable risk profile for their particular physiology. The goal is matching the method to the actual risk profile, which requires that the risk profile be assessed.

Two additional points on method selection are worth stating plainly. First, progestin-only pills require more precise daily timing than combined pills; missing the three-hour window reduces efficacy. Long-acting progestin-only options, including the hormonal IUD and the subdermal implant, remove that adherence requirement entirely and are among the most effective available methods with a very low absolute cardiovascular risk profile. For women whose risk profile makes estrogen problematic, these long-acting options deserve serious consideration rather than being treated as a fallback. Second, non-hormonal options, including the copper IUD, which is highly effective and carries no hormonal cardiovascular effects, are a legitimate first-line option for women who prefer to avoid all hormonal methods regardless of cardiovascular risk.

The cardiovascular risk of hormonal contraception is low for most women and meaningfully higher for some, depending on specific factors including smoking, blood pressure, and migraine with aura. Matching the method to a woman’s actual risk profile, rather than reacting to headlines or relying on generic reassurance, is how the decision gets made on evidence. That matching requires the clinician to ask and the woman to know enough to answer accurately.