The Relationship between Exogenous Hormonal use and Cerebral Vein Thrombosis in Women: A Comprehensive Review

INTRODUCTION

Cerebral venous thrombosis (CVT) constitutes only 0.5–1% of all types of cerebrovascular disease (CVA).^[1] According to the World Health Organization, a stroke caused by CVT is defined as an infarction or hemorrhage in the brain, spinal cord, or retina due to thrombosis of a cerebral vein, most commonly of intracranial veins and dural sinuses.^[2] Risk factors for CVT include pregnancy or puerperium, oral contraceptives (OCP), and, less commonly, hormone replacement therapy (HRT) usage.^[3] This association was first published in 1970, but little attention was given to OCP at that time, as there were only a few studies and observations available.^[4] This review highlights the link between OCP use and CVT. As OCPs are widely used, understanding their potential to increase CVT risk is vital, given that CVT, though serious, is often treatable with good outcomes.

A systematic search was conducted across PubMed, Cochrane Library, EMBASE, Scopus, Web of Science, and cumulative index to nursing and allied health literature (CINAHL) databases to identify relevant studies published after 2005. The search strategy employed two key themes: Exposure terms (hormonal contraceptives, OCPs, birth control pills, ethinyl estradiol, desogestrel, levonorgestrel, and HRT) and outcome terms (cerebral vein thrombosis, intracranial venous thrombosis, sinus thrombosis, dural venous thrombosis, and venous infarct), linked using the and operator. Medical subject headings (MeSH) terms and controlled vocabulary synonyms were utilized without study design restrictions to maximize sensitivity. Two independent reviewers screened articles using a two-stage process, beginning with title and abstract review, followed by full-text assessment of potentially relevant studies. Inclusion criteria required studies of women aged 15–60 years with documented hormonal contraceptive exposure and objectively confirmed cerebral venous sinus thrombosis (CVST) through imaging. Studies were excluded if they included pregnant or postpartum women, trauma patients, those with malignancy, autoimmune diseases, or prolonged immobilization. Given CVST’s rarity, both cohort and case–control designs were considered acceptable. Quality was evaluated using five criteria: Appropriate case–control definitions, adequate matching for confounders, recent exposure documentation, control for thrombophilia and other risk factors, and use of statistical adjustments before reporting associations.

EPIDEMIOLOGY AND PATHOPHYSIOLOGY

CVT incidence exceeds 10 per million, with women affected 3 times more than men.^[1] During pregnancy and puerperium, CVT accounts for 6–64% of strokes and occurs at a rate of 11.6/100,000 deliveries in developed nations.^[5] Meta-analyses report a markedly increased CVT risk with OCP use (odds radio [OR] 5.59, 95% confidence interval [CI] 3.95–7.91).^[6] In the ISCVT study, 46% of women with CVT used OCPs, while other analyses show risk increases up to 7.59-fold.^[7] An Asian series identified OCP use in 12% of CVT cases.^[8] OCP use among CVT patients rose from 6.9% before 1990 to 46.7% after 2000 in a review of 112 studies.^[9]

The pathophysiology of CVT is complex due to anatomical differences and a lack of animal models. The leading theory, Virchow’s Triad, involves blood stasis, endothelial injury, and a hypercoagulable state disrupting the balance between clot formation and breakdown.^[10] The coagulation cascade, key to thrombosis, has three phases: Initiation, amplification, and propagation [Figure 1]. Injury to the vessel wall activates Factor VII and tissue factor, converting prothrombin to thrombin. Thrombin then activates platelets and Factor VIII, which, with Factor IX, activates Factor X, forming the prothrombinase complex that promotes fibrin clot formation. Natural anticoagulants regulate this to prevent excessive clotting.^[11] OCPs, especially those with estrogen, significantly affect coagulation [Figure 2]. Second- and third-generation OCPs increase procoagulant factors (VII, X, fibrinogen, and prothrombin fragments) and Factor VIII, while reducing Factor V and natural anticoagulants such as antithrombin and protein S.^[11] Estrogen also alters gene transcription by binding to nuclear receptors, promoting a hypercoagulable state. It raises tissue plasminogen activator and plasminogen but inhibits fibrinolysis by increasing plasminogen activator inhibitor and thrombin-activatable fibrinolysis inhibitor, increasing venous thrombosis risk.^[11] Genetic factors compound risk; for example, OCP users with prothrombin G20210A or Factor V Leiden mutations show markedly increased CVT risk (OR up to 30). A Dutch study confirmed this heightened risk in women using OCP with thrombophilia (OR 13 to 30).^[12]

Close

Figure 1:

Schematic overview of the coagulation cascade. Coagulation factors are indicated by Roman numerals, with the suffix “a” indicating the activated form. (PT: Prothrombin time)

Export to PPT

Close

Figure 2:

Coagulation profile changes seen with second and third generations of oral contraceptive pills (OCP) and combined vaginal contraception.

Export to PPT

CLINICAL FEATURES

CVT manifests in acute, subacute, chronic, or acute-on-chronic forms, presenting through five primary syndromes tied to the thrombosis location and extent. Raised intracranial pressure is seen in acute to chronic forms of CVT. The rest all present in acute to subacute forms.

1. Raised intracranial pressure: Headache, vomiting, blurred vision, and sixth cranial nerve palsy are often associated with superficial sinus thrombosis.

2. Headache with focal neurological deficits and seizures: Cortical vein thrombosis can lead to significant complications such as edema and venous infarction. This can result in increased intracranial pressure, potentially causing brain herniation.

3. Multiple cranial nerve lesions: This mainly affects cranial nerves III-XI, causing symptoms such as vision changes, facial weakness, or difficulties with swallowing, depending on the affected nerves.

4. Subacute encephalopathy: This is a rare presentation, seen in deep venous thrombosis, with altered sensorium and rapid progression, and is often fatal.

5. Cavernous sinus syndrome: Painful proptosis, chemosis, and ocular palsy can occur due to infections originating in the face or paranasal sinuses. These conditions often affect cranial nerves III to VI, leading to various symptoms, including visual disturbances and limited eye movement.

There are no unique signs or symptoms described in the literature for OCP-related CVT. Headache is the most common, being present in 90% of cases.^[13] The study done by Ferro et al. reported that CVT commonly presented with intracranial hypertension in 92%, headache in 89.1%, focal deficits in 46.4%, motor deficits in 35.5%, vomiting in 32.6%, seizures in 22.5%, mental status disturbance in 31.2%, and papilledema in 6.5%. Isolated intracranial hypertension occurred in 41.3%, and aphasia in 18.1% of patients.^[13] The most common differentials to be considered are migraine, ischemic stroke, intracranial cerebral hemorrhage, brain neoplasm, meningoencephalitis, hypertensive emergency, idiopathic intracranial hypertension, and preeclampsia/eclampsia (in postpartum females).^[14-16]

INVESTIGATIONS

Investigations for CVT require a combination of imaging and laboratory tests due to its variable clinical presentation. Magnetic resonance imaging (MRI) combined with magnetic resonance venography (MRV) is considered the most sensitive method for diagnosing CVT.^[17] However, computerized tomography with venography remains widely used due to its availability, lower cost, and rapid results.^[17]

1. Apart from routine blood investigations, a focused prothrombotic panel should be done, which includes protein C, protein S, antithrombin, Vitamin B12, homocysteine levels, and testing for methylenetetrahydrofolate reductase gene mutations. According to the American Stroke Association, D-dimer testing can help exclude CVT in low-probability cases, but a normal D-dimer does not rule out the diagnosis.^[13] Thrombophilia testing – covering protein C, protein S, antithrombin, Factor VIII, lupus anticoagulant, and genetic mutations such as Factor V Leiden and prothrombin G20210A – is ideally performed 8–12 weeks after stopping anticoagulation to guide prognosis and treatment. Some genetic and antibody assays (e.g., antiphospholipid antibodies) should be performed at diagnosis as they are unaffected by anticoagulation.^[18]

2. Other investigations depend on clinical suspicion and include pregnancy tests, iron studies, tests for COVID-19 infection, antiphospholipid and anticardiolipin antibodies, antinuclear antibodies, and anti-neutrophil cytoplasmic antibodies for autoimmune conditions. Biomarkers for cardiac stress should also be assessed to evaluate the presence of an associated cardiac etiology.^[19,20]

3. Imaging techniques vary in sensitivity and specificity. Plain CT has moderate sensitivity (~79%) and specificity (~81%) for CVT diagnosis, improved by contrast enhancement.^[21] CT venography offers a high sensitivity of 95% for cerebral vein evaluation, outperforming some MRV techniques in detecting small cortical vein thromboses.^[22] MRI provides a sensitivity of approximately 82% and specificity of 92%, with MRV showing even higher sensitivity (91.3%).^[22] Contrast-enhanced MRV, using gadolinium, offers superior delineation of cerebral venous structures with sensitivity and specificity reported at 83% and 100%, respectively.^[22] Digital subtraction angiography remains the gold standard, especially valuable for complex cases, and allows for therapeutic interventions such as thrombolysis and thrombectomy.^[22] The common signs seen in neuroimaging of CVT cases are mentioned in Table 1. Typical and characteristic CT features of cerebral venous thrombosis are shown in Figure 3. Furthermore, in cases of CVT with concurrent ischemic stroke, a work-up should ideally include detailed echocardiographic imaging as part of a broader evaluation for cardioembolic sources, complemented by clinical assessment and possibly extended cardiac rhythm monitoring.^[23,24]

Close

Figure 3:

Plain and contrast computed tomography (CT) scan of the brain showing (respective white arrows) – (a) transverse sinus thrombosis; (b) straight sinus thrombosis; (c) bilateral parietal venous hemorrhagic infarcts with perilesional edema and diffuse cerebral edema; (d) empty delta sign; (e) cord sign and hyperdense sagittal sinus thrombosis; and (f) CT venogram (axial) showing extension of the cerebral venous thrombosis down to the jugular vein.

Export to PPT

TREATMENT

Acute CVT management focuses on sinus recanalization and combines anticoagulation, symptomatic treatment, and risk factor correction.

1. Initial CVT treatment starts with heparin, followed by Vitamin K antagonists or direct-acting oral anticoagulants (DOACs). Anticoagulation continues for 3–12 months in transient cases and indefinitely in high-risk or recurrent thrombosis. A recent systematic review summarizing three randomized trials (RE-SPECT, ACTION-CVT, and SECRET) and 16 observational studies comparing DOACs with Vitamin K antagonists (VKAs) found similar rates of recurrent venous thromboembolism, major hemorrhage, and complete recanalization (42.9% vs. 42.3%; relative risk, 0.98; 95% CI, 0.87–1.11).^[25] DOACs are contraindicated in pregnant and breastfeeding women (American heart association [AHA], 2024).^[13]

2. According to American society of anesthesiologists (ASA) guidelines, patients who experience seizures, with or without parenchymal brain lesions, may benefit from early initiation of antiepileptic drug (AED) therapy to prevent seizure recurrence. AEDs can be discontinued at follow-up visits once the patient stabilizes. However, those without seizures do not derive benefit from prophylactic AED use.^[13]

3. Management of isolated intracranial hypertension involves both pharmacological and supportive approaches. Pharmacological treatment includes mannitol, glycerol, and acetazolamide. Supportive measures consist of elevating the head of the bed to 30 degrees and implementing controlled hyperventilation to maintain PaCO₂ between 30- and 35-mm Hg.^[26] Mannitol reduces vascular congestion, improves cerebral perfusion, and may decrease intracranial pressure, resulting in headache relief; however, current guidelines do not recommend its routine use.^[13] A meta-analysis indicated that glycerol could be a more effective and better-tolerated alternative to mannitol for reducing cerebral edema and elevated intracranial pressure, especially in high-risk patients with renal failure.^[27]

4. According to AHA guidelines, ventriculoperitoneal shunting is advised for secondary hydrocephalus, while lumbo-peritoneal shunting is considered for refractory intracranial hypertension associated with progressive visual loss.^[2] Nevertheless, a few case reports have described CVT development following shunting procedures (Iimori et al.).^[28] For patients with optic nerve compression, an optic nerve sheath fenestration (ONSF) procedure may be performed. ONSF relieves pressure within the subarachnoid space in cases of elevated intracranial pressure and may help prevent progressive visual damage related to cerebral venous disease. Despite its potential benefits, there is no consensus regarding the optimal timing of this surgery.^[29]

5. The rate of partial or complete recanalization in CVT varies between 47% and 100% with anticoagulation therapy alone.^[30] Interventional options include catheter-directed thrombolysis and mechanical thrombectomy, with surgical decompression reserved for cases with large venous infarcts or hemorrhages causing severe intracranial pressure. In one review involving 13 patients with severe CVT who underwent decompressive craniectomy, 11 (84.6%) achieved favorable outcomes.^[31] Endovascular therapy (EVT) has also been investigated. The multicenter, randomized TO-ACT trial (Thrombolysis or anticoagulation for CVT) demonstrated that patients with severe CVT did not experience clinical improvement with EVT compared with standard anticoagulation.^[32] At present, EVT is employed as a rescue treatment for individuals who deteriorate clinically or present with contraindications to conventional therapy (AHA, 2024).^[13]

OUTCOME AND PROGNOSIS

The prognosis of CVT is generally favorable, with up to 80% of patients achieving complete recovery. However, a significant minority, approximately 13%, may experience poor outcomes, including death or severe disability.^[33]

Data from large cohort studies such as the ISCVT indicate that women tend to have better outcomes than men, with complete recovery rates of 81% versus 71%, respectively.^[34] There remains insufficient evidence regarding the risk of CVT recurrence related to OCP use, likely due to the variety of OCP formulations available. Secondary analyses from the ACTION-CVT study, which included 947 patients, identified that Black race, history of venous thromboembolism, and the presence of antiphospholipid antibodies correlate with a higher risk of CVT recurrence.^[35] For patients experiencing recurrent CVT, VTE post-CVT, or those with severe thrombophilia – including homozygous prothrombin G20210A mutation, homozygous Factor V Leiden, deficiencies of protein C, protein S, antithrombin, combined thrombophilia, or antiphospholipid syndrome – indefinite anticoagulation may be considered, targeting an INR of 2.0 to 3.0 as recommended by AHA guidelines.^[13,36] Cardiovascular rehabilitation and physiotherapy, notably incorporating high-intensity interval training, have demonstrated positive effects on functional recovery in stroke patients, suggesting potential benefits in CVT management protocols by enhancing patients’ quality of life and clinical outcomes.^[37]

CVT AND HORMONAL CONTRACEPTION

Hormonal contraception encompasses birth control methods that influence the endocrine system, including birth control pills, contraceptive skin patches, vaginal rings, and hormone-releasing intrauterine devices. The mechanism of action is illustrated in Figure 4, which involves estrogen and progesterone interacting with the hypothalamic-pituitary axis. Estrogen prevents implantation by altering the endometrium, while progesterone thickens cervical mucus to hinder sperm. OCPs are contraindicated in patients with venous thromboembolism, certain cancers, liver disease, uncontrolled hypertension, or thrombophilia.^[38] Postpartum women who are not breastfeeding and <21 days post-delivery should avoid OCPs due to an elevated risk. OCPs are classified into combined oral contraceptive pills (COCPs), which contain both estrogen and progesterone; progestogen-only pills (POPs); emergency contraceptive pills (ECPs), used after intercourse; and non-hormonal preparations.

Close

Figure 4:

Schematic illustration showing the main actions of combined oral contraceptives in the female reproductive system. SHBG: Sex hormone binding globulin.

Export to PPT

COCPs, containing synthetic estrogen and progesterone, have evolved through generations, reducing estrogen doses and mitigating risk. Early high-dose formulations (>50 µg estrogen) increased VTE risk four to ten-fold, while modern low-dose formulations (15–35 µg) reduce this risk to three- to six-fold.^[39] The sex hormone-binding globulin, which increases with estrogen and certain progestins (e.g., cyproterone acetate and gestodene), correlates with thrombogenicity. Activated protein C resistance, more common with third- and fourth-generation pills, is another contributor to VT risk.^[39] Table 2 depicts the risk of CVT with different generations of COCP formulations. Various case series and case studies concerning CVT and OCP intake in the past 20 years are summarized in Table 3.^[40-48] OCP users are at the highest risk during the 1^st year, but risk persists until OCP is used; however, not persist beyond discontinuation. Compared with nonusers, the relative risk of VT in OCP users is around 7.0% during the 1^st year, 3.6 % during 1–5 years, and 3.1% for >5 years of usage.^[49] The risk estimates for prothrombotic risk factors along with OCP use for developing any VT, and the recommendations suggested for decreasing that risk is mentioned in Table 4.^[50,51] OCPs containing levonorgestrel were associated with a nearly fourfold increased risk of venous thrombosis (odds ratio 3.6, range 2.9–4.6) compared to non-users.^[50] The risk for gestodene users was elevated by 5.6 times (range 3.7–8.4), for desogestrel by 7.3 times (range 5.3–10.0), for cyproterone acetate by 6.8 times (range 4.7–10.0), and for drospirenone by 6.3 times (range 2.9–13.7).^[51] Progestin-only pills, commonly used postpartum, have not shown a significant increase in thrombosis risk but may alter coagulation markers.^[52] POPs increase total and free protein S levels, decrease plasma protein C concentrations, and enhance plasma sensitivity to activated protein C. ECPs are employed after intercourse to prevent pregnancy and include hormonal regimens (such as the Yuzpe regimen, levonorgestrel, and ulipristal acetate) and non-hormonal copper intrauterine device (IUDs). The Yuzpe regimen involves two doses of 100 µg ethinyl estradiol with either 0.5–1 mg levonorgestrel or 1 mg norgestrel taken 12 h apart.^[53] Levonorgestrel can be administered as a single 1.5 mg dose or as two 0.75 mg doses taken 12 h apart (Split dose).^[54] While rare case reports document CVT following ECP use (Hogra et al.), systematic safety reviews have not identified increased thrombosis incidence.^[55] Overall, hormonal contraception impacts CVT risk variably, warranting individualized risk assessment before prescription, especially in patients with thrombophilia predispositions or other risk factors.

CVT WITH NON-ORAL HORMONAL PREPARATIONS

A 2006 study reported no significant increase in cerebral CVT risk associated with the contraceptive patch.^[56] However, a few published case reports have linked CVT to other hormonal devices such as the ethinyl estradiol vaginal ring.^[57] The BMJ follow-up study (Lidegaard et al.) also indicated that women using non-oral hormonal contraceptives – such as transdermal, subcutaneous, intrauterine, or vaginal preparations – had a modestly higher risk of venous thrombosis compared to non-users.^[58] Although the relative risk was lower than that associated with OCPs, certain patient factors, including age >35 years, smoking, obesity, or hypertension, appeared to amplify this risk. Furthermore, the meta-analysis by Tepper et al. showed that available evidence remains insufficient to confirm the safety of non-oral hormonal methods regarding CVT specifically.^[59] Hence, while these preparations are often preferred for convenience or reduced systemic absorption, caution is warranted in women with prothrombotic predispositions or cardiovascular comorbidities. Future prospective analyses are needed to clarify their long-term cerebrovascular safety.

CVT WITH HRT

HRT – including estrogen or combined estrogen-progestin regimens is known to increase venous thrombosis risk, particularly during the 1^st year of therapy. Epidemiological studies demonstrate a two- to five-fold higher risk of VT in HRT users compared to non-users, which rises considerably in those with underlying thrombophilia or advancing age. Combined HRT has shown a greater risk than estrogen-only therapy (OR ≈ 1.6), and oral HRT carries a higher thrombotic burden than transdermal formulations (OR ≈ 4.0).^[60] The ESTHER study group further highlighted that oral – but not transdermal – estrogen significantly increases venous thromboembolism risk.^[61] Similarly, according to the ISCVT cohort, about 4.3 % of CVT cases occurred in women on HRT.^[62] Taken together, these findings suggest that oral estrogen preparations are associated with a prothrombotic shift in coagulation markers and clinical events, whereas transdermal routes appear comparatively safer. HRT should therefore be individualized, using the lowest effective estrogen dose and preferring transdermal forms, especially in postmenopausal or genetically predisposed women.

CONCLUSION

CVT is an uncommon but preventable cerebrovascular condition strongly associated with hormonal factors, especially OCPs use. Estrogen-containing COCs increase CVT risk by approximately 5–7 times compared with non-users, particularly in women with thrombophilia, obesity, or smoking history. Estrogen induces a prothrombotic state by elevating procoagulant factors and reducing natural anticoagulants, while newer third- and fourth-generation progestins such as drospirenone or desogestrel further heighten thrombotic potential. Diagnosis relies on a high index of suspicion followed by MRI or MR venography, as presentations are often nonspecific, with headache being the most frequent symptom. HRT and some non-OCPs also carry a thrombosis risk, but transdermal estrogen-based HRT has proved much safer.Low-dose, second-generation levonorgestrel-based or progestin-only contraceptives appear to be the safest options for women at risk of CVT, emphasizing the need for individualized risk assessment before hormonal therapy initiation.