Clinical manifestations

Presentation and course

Ischemic stroke and intracranial hemorrhage. Patients with ischemic stroke or transient ischemic attack typically present with acute-onset, focal neurologic symptoms, such as weakness, numbness, aphasia, dysarthria, vision loss, or gait disturbance. The goal of acute treatment is to preserve the ischemic penumbra with thrombolysis or thrombectomy. Patients with intracranial hemorrhage typically present with focal neurologic signs, as is the case for ischemic stroke. Patients with intracranial hemorrhage may also have a depressed level of consciousness or other symptoms related to increased intracranial pressure. Some patients fully recover after ischemic stroke and intracranial hemorrhage, but others are left with substantial disability.

Cerebral venous thrombosis. Symptoms of cerebral venous thrombosis may present acutely or subacutely. Patients often report headache. Venous infarction, hemorrhage, and seizures can occur, presumably due to impaired venous drainage. In severe cases, intracranial pressure increases. Cerebral venous thrombosis is treated with urgent systemic anticoagulation, even in the setting of associated intracranial hemorrhage. Catheter thrombolysis or thrombectomy can be considered in patients with clinical and radiographic deterioration despite anticoagulation, although no strong evidence supports a role for these interventions.

Reversible cerebral vasoconstriction syndrome. Classically, patients with reversible cerebral vasoconstriction syndrome present with a severe, acute-onset (“thunderclap”) headache. Patients can also have focal neurologic symptoms related to acute ischemic stroke or intracranial hemorrhage. Seizures at the time of presentation can also be seen.

Prognosis and complications

Ischemic stroke associated with hormonal contraception. A 2021 retrospective analysis of 179 female patients in Switzerland aged 18 to 50 with acute ischemic stroke from 2003 to 2015 found that patients on systemic hormonal contraception compared to nonusers had a more favorable 12-month outcome, defined as an mRS of 2 or less (08). Contraceptive users had a lower risk of stroke or transient ischemic attack recurrence at 12 months (OR 0.04; 95% CI: 0.00–0.55). All patients in the contraceptive group discontinued hormonal contraception in the first 12 months after the index stroke.

A more favorable outcome among women with ischemic stroke associated with hormonal contraception is not surprising because these patients tend to be younger and healthier than women with stroke who do not use contraception. Furthermore, women with stroke associated with hormonal contraception have an easily modifiable risk factor: withdrawal of hormonal contraception.

Intracranial hemorrhage associated with hormonal contraception. Data on the prognosis of patients with intracranial hemorrhage associated with hormonal contraception are lacking. In general, the 30-day mortality rate after intracranial hemorrhage can range from 30% to 50% and is largely affected by clinical features and underlying etiology, but this rate is difficult to extrapolate to patients with intracranial hemorrhage associated with hormonal contraception. Patients with intracranial hemorrhage associated with hormonal contraception tend to be younger and healthier and, thus, have a better prognosis (53).

Cerebral venous thrombosis associated with hormonal contraception. In a single-center study of 24 patients with cerebral venous thrombosis between 1992 and 2002, oral contraceptive use was the presumed etiology in four patients (02). All patients were treated with heparin followed by oral anticoagulation. Of the four patients with oral contraceptive use, one reported a new severe headache, one reported cognitive dysfunction, and two were normal at hospital discharge.

In a prospectively maintained database of patients with cerebral venous thrombosis (all causes) between 1998 and 2001, mortality at the end of the median follow-up period of 16 months was 8.3%, and excellent outcome, defined as an mRS of 1 or less, was seen in 79.1% of patients and increased to 86.6% with an mRS of 2 or less (12). In this cohort, hormone replacement therapy was noted in 4.3% of the cohort and oral contraceptives in 54.3% of the female patients over 50 years of age.

Reversible cerebral vasoconstriction syndrome associated with hormonal contraception. Due to the low reported incidence of reversible cerebral vasoconstriction syndrome, data on the prognosis of reversible cerebral vasoconstriction syndrome associated with hormonal contraception are lacking. In one study analyzing the reversible cerebral vasoconstriction syndrome registries of two academic hospitals, approximately 97% of patients were functionally independent based on Barthel Index scores at a median follow-up time of 78 months (21). The time course for improvement or resolution of differing symptoms associated with reversible cerebral vasoconstriction syndrome is variable.

Biological basis

The precise effects of hormonal contraception on various components of the coagulation cascade are unknown. Much research has found an effect of hormonal contraception on thrombogenic agents (eg, fibrinogen, prothrombin, and coagulation factors) and endogenous antithrombotic factors (eg, protein C, protein S, and antithrombin).

The effects on the coagulation cascade are more pronounced in third-generation hormonal contraceptives as compared to second-generation, which use differing formulations of progesterone (29; 46; 51; 35). These effects include an increase in fibrinogen, prothrombin, and factors VII, VIII, and X (29). There is also a decrease in protein S and antithrombin, both of which regulate the coagulation cascade via inhibition (46; 22; 24), and in the tissue factor pathway inhibitor (17; 09; 51; 35). Of note, hormonal replacement therapy has also been shown to affect similar components of the coagulation cascade in a dose-dependent manner (39).

Protein C, an antithrombotic factor in the coagulation cascade, is a potentially relevant site of action of hormonal contraception. In a mouse model and in clinical studies, both free tissue factor pathway inhibitor and protein S decrease in response to hormonal therapy (19; 51; 47). There has, however, been one case-control study that did not find an association between tissue factor pathway inhibitor and protein S levels with stroke (38).

A linear relationship between the levels of sex hormone binding globulin (SHBG) during the use of oral contraceptive and the risk of venous thrombosis has been suggested (31). However, a study of postmenopausal women in the Woman’s Health Initiative (WHI) found an inverse relationship between the level of SHBG and ischemic stroke (27).

Although not well understood, estrogen appears to have a stronger effect on coagulation factors and hypercoagulability than progesterone.

Etiology and pathogenesis

The precise mechanism by which oral contraception promotes coagulability remains unknown.

In normal hemostasis, injury to the vascular endothelium results in a cascade of interactions between coagulation factors and components of the exposed vascular endothelium. This cascade results in the aggregation of platelets locally around the region of vascular injury. Fibrin fibrils stabilize this aggregation with a fibrin mesh, which forms a thrombus, which, in turn, can prevent blood from flowing distally (54). In some cases, the thrombus will embolize, moving distally. The embolus may subsequently get lodged within a smaller portion of the circulatory system, block flow distally, and lead to ischemia of the affected tissue.

A change in the plasma levels of hemostatic agents, including coagulation factors, inhibitors, and fibrinolytic factors, has been associated with increases in stroke risk (43; 42; 28). The effects on hemostasis following an increase in thrombogenic agents and a decrease in antithrombotic agents result in a hypercoagulable state, which, in turn, leads to an increased risk of thrombosis in the large and small vessels throughout the body.

In the setting of a hypercoagulable state, cerebral ischemic injury can result from the formation of in situ thrombus in the cerebral arteries, leading to ischemic stroke, or from emboli from arterial thrombus elsewhere. Venous emboli may also form in the cerebral veins, leading to cerebral venous thrombosis, or in the extremity or pelvic veins allowing for the potential for paradoxical emboli to the brain if an intracardiac shunt is present.

Any amount of ischemia results in cytotoxicity within brain parenchyma and causes focal neurologic deficits based on the region in which brain parenchyma sustained injury. Clinical symptoms are based on both the size and location of ischemia.

Epidemiology

Risk of ischemic stroke

Combined estrogen and progesterone formulations. In 1996, the World Health Organization Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception published the results of its case-control study, which examined the risk of ischemic stroke in association with current use of combined hormonal birth control in 697 cases and in 1962 age-matched controls (52). The risk of ischemic stroke in contraceptive users compared to nonusers was higher in Europe (aOR 2.99; 95% CI: 1.65–5.40), developing countries (aOR 2.93; 95% CI: 2.15–4.00), Asia (aOR 3.55; 95% CI: 2.13–5.92), and Latin America (aOR 2.76; 95% CI: 1.86–4.11).

Many but not all cohort and case-control studies have found an association between combined hormonal contraception and ischemic stroke. A 2015 Cochrane review identified 24 studies that assessed the risk of ischemic stroke in users of combined oral contraception versus nonusers (37). Combined oral contraceptive users showed an increased risk of stroke (RR 1.7; 95% CI: 1.5–1.9) compared to nonusers. However, significant heterogeneity was found among studies.

One study in this Cochrane review warrants mention for its large study sample. This cohort study of Danish women aged 15 to 48 years investigated the risk of thrombotic stroke in women according to contraception type, estrogen dose, progestin type, and route of administration (26). The study was massive, with a cohort size of 1,626,158 women with 14,251,063 person-years of observation. Thrombotic stroke was observed in 3311 (0.20%) women. Oral contraceptives containing 20 µg, 30 to 40 µg, and 50 µg of ethinyl estradiol were associated with a relative risk of thrombotic stroke of 1.60 (95% CI: 1.37–1.87), 1.75 (95% CI: 1.61–1.92), and 1.97 (95% CI: 1.45–2.66) respectively, though the linear trend was not significant (p=0.24).

A 2016 systematic review identified two studies that reported the risk of arterial events among combined hormonal patch users compared to combined oral contraceptive users (49). Neither study found a difference in the risk of ischemic stroke between the patch and oral contraceptive users.

In the 2002 Risk of Arterial Thrombosis in Relation to Oral Contraceptives (RATIO) study, the risk of stroke was higher in women on oral contraception who were obese (OR 4.6; 95% CI: 2.4–8.9) compared to women who were not (OR 2.2; 95% CI: 1.5–3.0) (23).

In the 1996 WHO study, the odds ratio of ischemic stroke for smokers compared to nonsmokers was 7.20 (95% CI: 3.23–16.1) among contraceptive users in Europe (52). For smokers versus nonsmokers in developing countries, the odds ratio was 4.83 (95% CI: 2.76–8.43). Similarly, for women on oral contraceptives in the RATIO study, the risk of stroke was higher in women who smoked (OR 4.4; 95% CI: 2.7–7.3) compared to women who did not smoke (OR 2.8%; 95% CI:1.5–5.0).

The risk of stroke appears to increase with higher estrogen doses. In a 2019 meta-analysis, an increased risk of ischemic stroke was associated with every 10-µg increase in estrogen dose (OR 1.20; 95% CI: 1.17–1.22) (25). The investigators also found that duration of exposure may also increase stroke risk (OR for every 5-year increment in duration of contraception 1.24; 95% CI: 1.03–1.49).

A case-control analysis sought to clarify how estrogen dose may impact stroke risk (03). Of 203,853 combination hormonal contraception users in the study (127 with ischemic stroke), stroke risk was higher with 30 or more micrograms of ethynyl estradiol compared to less than 30 micrograms (OR 1.52, 95% CI: 1.02–2.26).

Progesterone-only formulations. A 2018 systematic review found six studies (five case-control and one cohort) that examined the risk of stroke in women taking progesterone-only contraception (15). On meta-analysis, the use of progesterone-only contraception was not associated with an increased risk of stroke (adjusted pooled RR 1.02; 95% CI: 0.72–1.33). Similarly, in the aforementioned large Danish cohort study, progestin-only contraceptives, including the levonorgestrel-releasing IUD, were not associated with thrombotic stroke, regardless of route of administration (26).

Risk of intracranial hemorrhage

A 2018 meta-analysis sought to estimate the risk of hemorrhagic stroke associated with the current use of oral contraceptives (55). Five cohort studies and 10 case-control studies including 4271 patients with hemorrhagic stroke were included in the analysis. An increased risk of hemorrhagic stroke (subarachnoid hemorrhage and intracerebral hemorrhage) was found for contraceptive users (summary OR 1.39; 95% CI: 1.05–1.83). When each type of hemorrhage was examined individually, contraceptive users were at an increased risk of subarachnoid hemorrhage (OR 1.60; 95% CI: 1.21–2.12) but not intracerebral hemorrhage (summary OR 0.92; 95% CI: 0.33–2.54). However, it should be noted that ten studies evaluated subarachnoid hemorrhage, whereas only four evaluated intracerebral hemorrhage.

Risk of cerebral venous thrombosis

Because it is well-established that hormonal contraception is associated with venous thrombosis, it is not surprising that hormonal contraception is associated with cerebral venous thrombosis. A 2015 meta-analysis identified nine case-controlled studies published between 1996 and 2005 that investigated the association between oral contraception and cerebral venous thrombosis (01). Across the studies, the pooled odds ratio for developing cerebral venous thrombosis for contraceptive users compared to nonusers was 7.59 (95% CI: 3.82–15.09). However, heterogeneity among the nine included studies was high (I2 = 72.1%). Current guidelines recommend screening for contraceptive use in the initial clinical assessment of women with cerebral venous thrombosis (05).

Risk of reversible cerebral vasoconstriction syndrome

Case reports have reported an association between estrogen withdrawal and reversible cerebral vasoconstriction syndrome (30; 56). In one report, a woman developed reversible cerebral vasoconstriction syndrome 1 month after starting a combination oral contraceptive pill (44). Because reversible cerebral vasoconstriction syndrome can present with ischemic stroke, subarachnoid hemorrhage, and intracerebral hemorrhage, some of the patients included in the aforementioned studies may have had this diagnosis.

Prevention

Because the absolute risk of stroke in the setting of hormonal contraception is low, primary prevention is not indicated. Universal screening for thrombophilia in women who initiate oral contraceptives is not recommended (05). Secondary stroke prevention is discussed under Management.

Differential diagnosis

Confusing conditions

The differential diagnosis for stroke generally includes migraine and seizure. Hypoglycemia, and rarely hyperglycemia, can mimic strokes by causing focal neurologic symptoms that resolve with resolution of the severe glucose imbalance.

Associated or underlying disorders

Other risk factors for ischemic stroke include age, high blood pressure, hyperlipidemia, male sex, active malignancy, and obstructive sleep apnea. There are genetic disorders that can place people at higher risk for stroke (eg, cerebral autosomal dominant arteriopathy with subcortical infarctions and leukoencephalopathy [CADASIL], sickle cell disease) and genetic disorders that are associated with hypercoagulable states, like Factor V Leiden.

Diagnostic workup

Beyond imaging techniques to confirm the diagnosis of ischemic stroke, intracranial hemorrhage, cerebral venous thrombosis, or reversible cerebral vasoconstriction syndrome, diagnostic work-up to suggest or establish the etiology as related to hormonal contraception relies on a thorough evaluation of other potential cardiac and vascular etiologies. This work-up will typically entail vessel imaging with a CT angiogram, MR angiogram, or carotid dopplers; cardiac imaging with a transthoracic echocardiogram; and cardiac rhythm monitoring and may include hypercoagulable serologic studies to evaluate for an underlying inherited or acquired thrombophilia apart from the use of hormonal contraception.

Ischemic stroke. A non-contrast head CT is required to diagnose ischemic stroke, as intracranial hemorrhage must be ruled out. If a head CT is negative for hemorrhage, a patient’s history and examination may be sufficient to confer a diagnosis of ischemic stroke. If the diagnosis is uncertain, or if clarifying the location and the extent of the infarct is important, an MRI brain may be pursued.

Intracranial hemorrhage. A non-contrast head CT is a sensitive, specific, and quick test for intracranial hemorrhage.

Cerebral venous thrombosis. Cerebral venous thrombosis can sometimes be suggested based on characteristic findings on a non-contrast head CT. Often a dedicated CT venogram or an MRI brain or MR venogram is required to confirm the diagnosis.

Reversible cerebral vasoconstriction syndrome. MRI or CT scanning is used to determine whether reversible cerebral vasoconstriction syndrome has caused an associated intracranial hemorrhage (intraparenchymal hemorrhage or subarachnoid hemorrhage) or ischemic stroke. In addition, vessel imaging with a CT angiogram, MR angiogram, or conventional angiogram should be performed to evaluate for focal narrowing of the intracranial vessels. Vessel changes can be dynamic in reversible cerebral vasoconstriction syndrome, so repeat vessel imaging may be needed to secure the diagnosis.

Management

Although high-quality evidence is lacking, most providers recommend discontinuation of hormonal contraception in women who have had an ischemic stroke. In most cases, antiplatelet agents are used in secondary stroke prevention in patients who suffer from stroke related to hormonal contraception.

Current guidelines suggest blood pressure measurement prior to initiating hormonal contraception. This recommendation is based on results from the WHO Study and the Transnational Study (48). In the WHO study, women on oral contraception who had their blood pressure checked prior to initiating contraception had lower odds of ischemic stroke than did women on oral contraception who did not have their blood pressure checked (OR 3.90; 95% CI: 1.83–8.33 versus OR 2.26; 95% CI: 1.12–4.55) (52). Similarly, in a population of 220 women with ischemic stroke and 775 controls in the Transnational Study, women who had their blood pressure checked had lower odds of thromboembolic stroke compared to women who had not (18). The authors suspected that blood pressure checks served as a surrogate for better healthcare or health behavior. In two Danish studies, stroke incidence was lower than in non-Danish studies, and this was hypothesized to be related to the fact that the majority of Danish women have their blood pressure checked before obtaining prescriptions for contraceptives (33; 26).

Outcomes

The natural history of stroke may be more favorable for patients with contraceptive-associated stroke as these patients tend to be healthier and younger than typical stroke patients. There is no evidence to suggest a difference in treatment effect for patients with contraceptive-associated stroke compared to those with stroke from other causes.

Special considerations

Pregnancy

Although patients who are currently pregnant do not receive exogenous hormonal contraceptives, there continues to be a risk of stroke associated with pregnancy, especially in the third trimester and the postpartum period (20). In patients with a history of cerebral venous thrombosis regardless of the association with hormonal contraception, prophylaxis with low molecular weight heparin is suggested during the pregnancy and immediate postpartum period (40; 05; 11). Postpartum angiopathy, a form of reversible cerebral vasoconstriction syndrome associated with pregnancy and a postpartum state, has been described; there is no evidence for a high risk of recurrence with future pregnancies, although data are limited (36).

Migraine

Migraine headache is one of the most common neurologic diseases and is known to disproportionately affect women. Migraine with aura independently increases the risk of ischemic stroke (10; 50). An association between hormonal contraception and stroke in women with headache was first identified in the 1960s, raising concern about the safety of oral contraception in women with migraine (13). Over time, many case-control and cohort studies have sought to quantify this risk.

In 2000, the International Headache Society Task Force on Combined Oral Contraceptives & Hormone Replacement Therapy noted no contraindication to the use of oral contraception in women with migraine who did not have migraine aura or additional risk factors, such as age over 35 years, ischemic heart disease, hypertension, smoking, or diabetes mellitus (04). A potentially increased risk of ischemic stroke in women with migraine with aura who are using oral contraception was noted.

A 2017 study using a nationwide healthcare claims database investigated the association among combined hormonal contraceptive use, migraine with aura versus without, and ischemic stroke (06). Among females aged 15 to 49 between 2006 and 2012, 1884 women with a first-ever ischemic stroke were identified and age-matched to four controls (n=7536). Among women on combined hormonal contraceptives, the risk of ischemic stroke was almost two times higher in women with migraine without aura compared to women without migraine (OR 1.77; 95% CI: 1.09–2.88). For women with migraine with aura, the odds were greater than six (OR 6.08; 95% CI: 3.07–12.05). Per the current guidelines, smoking cessation should be strongly recommended for women with migraine headaches with aura (05).

A 2018 systematic review and meta-analysis evaluated 15 studies of women with migraine on hormonal contraception that reported stroke risk (41). The risk of stroke in migraine patients who used any dose of estrogen was higher than nonusers, with odds ranging from 1.08 to 16.9. The dose of estrogen was reported in seven of the studies. The authors found limited evidence regarding the effect of estrogen dosing on stroke risk.

In 2020, a systematic review of observational studies that included 12,256 women with migraine on low-dose combination hormonal contraception suggested increased odds of ischemic stroke in contraception users than in nonusers (32). Due to significant heterogeneity, a meta-analysis was not performed.

A case-control analysis sought to clarify how migraine history may impact stroke risk (03). Of 203,853 combination hormonal contraception users in the study (127 with ischemic stroke), stroke risk was twice as high in women with a history of migraine (OR 2.00; 95% CI: 1.27–3.17).

COVID-19

In 2023, a systematic review of three nonrandomized studies of interventions that included 314,704 patients and two case series of 13 patients with COVID-19 compared outcomes of those on hormonal contraception versus those not on hormonal contraception (34). Hormonal contraception use was not associated with an increase in the severity of COVID-19 disease course or with an increased risk of thromboembolism. Due to significant heterogeneity among the included studies, a meta-analysis was not performed.