Oral Contraceptives and Ischemic Stroke Risk
A 27-year-old, previously healthy woman with no history of migraine presented to the emergency department with sudden-onset left hemiparesis. Her only stroke risk factor was oral contraceptive use for the past several months. She awoke in normal health but soon developed left-sided weakness. On arrival to the emergency department, she had a blood pressure of 126/73 and a heart rate of 88. National Institutes of Health Stroke Scale was 5 for a left facial droop, hemiparesis, and dysarthria. Her computed tomography head was normal. She received intravenous tPA (tissue-type plasminogen activator) at her local hospital and was transferred to a comprehensive stroke center. On arrival, she had improved to a National Institutes of Health Stroke Scale of 2 for left facial droop and left arm drift. Magnetic resonance imaging of the brain revealed diffusion restriction within the right insula, putamen, and caudate (Figure [A] and [B]). Magnetic resonance angiography of the head and neck showed no intracranial or extracranial stenosis (Figure [C] and [D]). Fat-saturated imaging was considered but not done because she had widely patent arteries and no symptoms suggestive of dissection, such as headache or neck pain. She underwent an extensive workup, including transesophageal echocardiogram with bubble study, cardiac monitoring for paroxysmal atrial fibrillation, hemoglobin A1c, lipid panel, lupus anticoagulant, anticardiolipin antibodies, antinuclear antibodies, antithrombin-3, β-2 glycoprotein, homocysteine, protein C and S, and genetic testing for prothrombin gene mutation and Factor V Leiden. All testing was negative for a potential cause of stroke. She was diagnosed with cryptogenic stroke and discharged on aspirin and statin. Her oral contraceptive pills (OCPs) were stopped.
This case illustrates a common scenario where the only potential risk factor for stroke in a young woman is OCP use. Data linking OCPs to stroke is mixed, but most physicians err on the side of caution and discontinue OCPs. This discussion focuses on the potential role of OCPs in stroke and the management of patients with stroke who use OCPs.
Approximately, 10.7 million women aged 15 to 44 years in the United States use OCPs.1 OCPs are associated with increased procoagulant factors, including fibrinogen, prothrombin, and factors VII and VIII, and a decrease in levels of antithrombin and tissue factor pathway inhibitor. Futhermore, OCP users often have acquired resistance for activated protein C.2 Studies in the 1970s linked increased ischemic stroke to OCPs containing high-dose estrogens, but these formulations are no longer used. Several meta-analyses of case–control and cohort studies attempt to clarify the relationship between modern low-dose combination estrogen–progestin OCPs and stroke. The data agree that modern combination OCPs do not increase hemorrhagic stroke risk, and progestin-only OCPs do not increase any stroke-type risk. Whether or not combined OCPs increase risk of ischemic stroke is less straightforward.
The first OCPs were progestin only, and although less widely used today, are still available. Progestin-only pills are sometimes referred to as minipills. Although they less reliably suppress ovulation, they have fewer side effects then combination pills and can be used by women who cannot tolerate estrogens.3,4 Combination estrogen–progestin pills are more prevalent than progestin-only pills because of higher efficacy.5 Early combination pills contained >80 to 100 µg of the estrogen analog ethinyl estradiol; these high-dose formulations are uncommon today.6 Current combined OCP formulations are low dose and contain <50 μg (typically 30–35 μg) ethinyl estradiol7 (Table).
Hormonal contraception options include the weekly transdermal patch and monthly vaginal ring, which contain ethinyl estradiol and progestin and carry the same risk of unintended pregnancy as combined estrogen–progestin OCPs.5 In addition, there are several long-acting reversible contraceptive methods that do not use estradiol, including the nonhormonal copper intrauterine device (IUD), progestin-releasing intrauterine devices, progestin implants, and progestin injections (Table).
Contraception and Stroke Risk
The first meta-analysis addressing ischemic stroke risk in high-dose estrogen-containing OCP users evaluated 16 studies from 1960 to 1999 and found an increased relative risk of stroke of 2.75 (95% confidence interval, 2.24–3.38) across estrogen dose, blood pressure, smoking status, and age.8 Although concerning, this analysis does not reflect reduced risk of lower-dose ethinyl estradiol formulations used today.8 This same meta-analysis did not find any increased risk with progestin-only OCPs.9 A later meta-analysis including studies from 1980 to 2002 evaluated only low-dose combination OCPs and found a lower odds ratio of 2.12 (95% confidence interval, 1.56–2.86) for ischemic stroke.9 More recently, a study from 1995 to 2009 failed to find any increased relative risk for low-dose ethinyl estradiol formulations, estrogen-containing hormonal contraceptive patches, or vaginal rings.9
Importantly, the ischemic stroke risk of OCPs is additive with other stroke risk factors. When the initial, high-dose OCP meta-analysis controlled for smoking and hypertension, the relative risk of stroke was significantly smaller.8 Studies evaluating low-dose combination OCPs that controlled for smoking and hypertension also found significantly lower relative risk of ischemic stroke, 1.93 (95% confidence interval, 1.35–2.74).8 Normotensive nonsmokers using low-dose combination OCPs had an annual stroke risk of 8.5 per 100 000 patients, compared with 4.4 per 100 000 in nonusers.8 A more recent large-scale study following a Swedish cohort from 1991 to 2004 adjusted for smoking status, hypertension, diabetes mellitus, alcohol use, obesity, education, and level of physical activity and found no significant association between stroke and OCP use, duration of OCP use, or estrogen dose.9 Age, an additional stroke risk factor, likely also increases risk in OCP users.8
Migraine, OCPs, and Stroke Risk
Data linking stroke and OCP use in patients with migraine is limited and conflicting. The International Headache Society reports that patients with migraine (with or without aura) have an odds ratio of 3 for ischemic stroke; however, that risk increases to 6 in patients who have migraine with aura.1 Furthermore, the odds ratio for ischemic stroke in patients with migraine (with or without aura) using combination OCPs increases to 5 to 17.1 In contrast, both the World Health Organization Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception and a 2000 meta-analysis failed to find a significant effect of migraine (these studies did not differentiate migraine with or without aura) on ischemic risk in combined OCP users.1,8 The presence of additional stroke risk factors further increases risk of stroke with migraine.1
International Headache Society guidelines find no contraindications to combined OCP use in patients who have migraine without aura or lack other stroke risk factors (diabetes, hyperlipidemia, hypertension, smoking, and obesity).1 Use of nonestrogen contraception, such as an intrauterine device, progestin injections, or progestin implant, is recommended for women with multiple risk factors for stroke. These guidelines warn of a potential increased risk of stroke in patients who have migraine with aura, but there are no specific guidelines to not use OCPs in these patients.1 The American College of Obstetricians and Gynecologists guidelines further recommend, clinicians consider progestin-only (pill, IUD, injection, and implant) or nonhormonal (copper IUD, surgery, and barrier) contraceptive methods to women with migraine with focal neurological signs (for example, hemiplegic migraine), women ≥35 years of age, and smokers.4
Stroke Risk Associated With Lack of Contraception
Despite increased stroke risk with combined OCPs, these risks are less than the stroke risk associated with pregnancy, and clinicians must ensure women have access to adequate contraception.9 The absolute risk of stroke in otherwise healthy, nonpregnant young women is low at 21 per 100 000 patients; the risk of stroke rises during each pregnancy to 34 strokes per 100 000 deliveries.9 Importantly, multiple studies demonstrate overall mortality rates are not elevated with OCP use and may, in fact, be decreased in ever users compared with never users.10
Replacing combination OCPs with the second most prevalent method of birth control, the male condom, would result in ≈687 000 more unintended pregnancies, 26 more maternal strokes, and 33 more maternal deaths yearly in the United States, in addition to the psychosocial and economic impact of these unwanted pregnancies.8 It is imperative that clinicians consider the risks of pregnancy when counseling patients regarding stroke risk and OCP use. Careful follow-up and communication regarding contraceptive choices can prevent the morbidity and mortality associated with both ischemic stroke and unintended pregnancy.
Combination OCPs have a lower risk of stroke than earlier formulations containing high-dose estrogens. Stroke risk in OCP users is impacted by several confounding issues, including OCP formulation, age, hypertension, smoking, and the presence of migraine with aura. It is important to consider the constellation of stroke risk factors in concert with OCP formulation to determine individual stroke risk and the best contraceptive to reduce that risk in an individual woman. The young patient presented earlier used a low-dose combination OCP, did not smoke, was not hypertensive, and had no history of migraine with aura. Therefore, her stroke was likely not caused by her OCP; however, she changed her contraceptive method to a nonhormonal copper IUD.
Modern combined oral contraceptive pills have a lower dose of estrogens and less risk of stroke than older contraceptive formulations
The risk of ischemic stroke in patients using combined oral contraceptives is increased in patients with additional stroke risk factors, including smoking, hypertension, and migraine with aura.
The risk of ischemic stroke because of combined oral contraceptive pills is less than the risk associated with pregnancy.
- Received December 18, 2017.
- Revision received January 30, 2018.
- Accepted February 15, 2018.
- © 2018 American Heart Association, Inc.
- Bousser MG,
- Conard J,
- Kittner S,
- de Lignières B,
- MacGregor EA,
- Massiou H,
- et al
- Gialeraki A,
- Valsami S,
- Pittaras T,
- Panayiotakopoulos G,
- Politou M
- Bushnell C,
- McCullough LD,
- Awad IA,
- Chireau MV,
- Fedder WN,
- Furie KL,
- et al
- Hannaford PC,
- Iversen L,
- Macfarlane TV,
- Elliott AM,
- Angus V,
- Lee AJ