Cerebral Sinus Thrombosis

Abstract & Commentary

Source: Renowden S. Cerebral venous sinus thrombosis. Eur Radiol. 2004;14:215-226.

Cerebral sinus thrombosis most commonly affects the superior sagittal sinus (up to 80%) or the transverse and sigmoid sinuses (70%), with concomitant involvement of the cortical veins in up to 40% of cases. Women are more commonly affected, reflecting their unique risk factors, including pregnancy, the puerperium, and oral contraceptive medication. Other risk factors include surgery, head and neck infections, head injury, malignancy, dehydration, and connective tissue diseases.

Clinical presentation may be acute (30%), subacute (40%), or chronic (30%) and includes headache (up to 90%), increased intracranial pressure with dizziness, nausea, visual obscurations and papilledema, cranial neuropathies, altered sensorium, seizures, and, with deep venous involvement, stroke.

Diagnosis usually requires the increased sensitivity of magnetic resonance imaging (MRI), although in its absence, computerized tomography (CT) venography may enhance the accuracy of cranial CT. On cranial CT, pathognomonic signs include the cord sign (a hyperdense thrombosed vein on noncontrast CT) in up to 25% and the empty delta sign (enhancing superior sagittal sinus wall with intraluminal thrombus) in up to 46%. Brain edema, generalized (50%) or asymmetrical (3%), white matter edema (11%), hypodensity due to infarction (13%), intraparenchymal hemorrhage (32%), subdural hematoma (8%), and gyral (17%), tentorial, or falcine enhancement (19%) may also be observed on cranial CT.

MRI, with MR venography, is the examination of choice in suspected cerebral sinus thrombosis and should include T1 sagittal, T2 axial, and dual echo coronal sequences. Diffusion-weighted images have yet to prove their value, but CT venography is equivalent or superior to MRI/MRV for diagnosis. Cerebral angiography is generally not needed for diagnosis but may be performed during thrombolysis.

Treatment is not standardized, but heparin rather than warfarin is the treatment of choice, even in the presence of hemorrhagic infarction. Low-molecular-weight heparin has yet to prove its value, and systemic intravenous thrombolysis is not recommended. Either mechanical thrombolysis or local delivery of thrombolytics (urokinase or tPA, tissue plasminogen activator) into the occluded sinus may be attempted if deterioration occurs despite heparin, but data comparing them to heparin are lacking. Catheter-mediated thrombectomy is in clinical trials for carotid artery occlusion and may offer another mode of therapy. In all instances, supportive care mandates addressing any underlying condition and includes, as warranted, hydration, anticonvulsants, antibiotics, lowering intracranial pressure, or surgery.

Mortality is as high as 30%, with an additional 15-25% experiencing permanent neurologic deficit. Causes of death include cerebral edema, status epilepticus, sepsis, and pulmonary emboli. Recurrences occur in 12%, but obstetric patients do well and future pregnancy need not be avoided.


Prognosis appears better than previously appreciated. Among 624 adults with cerebral-vein and dural sinus thrombosis prospectively studied in a multinational, observational study from 1998 to 2001 and followed up for a median of 16 months, 57.1% (n = 356) were normal, with only minor symptoms in an additional 22% (n = 137). Mild or moderate impairment was seen in 7.5% (n = 47) and 2.9% (n = 18), respectively, whereas severe impairment or death was seen in 2.2% (n = 14) and 8.3% (n = 52), respectively. Age older than 37, male gender, coma, abnormal mental status, CT evidence of hemorrhage, deep-cerebral-vein thrombosis, CNS infection, and cancer were multivariate predictors of death or dependence.1 — Michael Rubin

Dr. Rubin, Professor of Clinical Neurology, New York Presbyterian Hospital-Cornell Campus is Assistant Editor of Neurology Alert.


1. Ferro JM, et al. Stroke. 2004;35:664-670.