Version May 2024
INTRODUCTION — Cerebral vein and dural sinus thrombosis (CVT) is less common than most other types of stroke but can be more challenging to diagnose. Due to the widespread use of magnetic resonance imaging (MRI) and rising clinical awareness, CVT is recognized with increasing frequency. In addition, it is now known to have a more varied clinical spectrum than previously realized. Because of its myriad causes and presentations, CVT is a disease that may be encountered not only by neurologists and neurosurgeons but also by emergency clinicians, internists, oncologists, hematologists, obstetricians, pediatricians, and family practitioners.
This topic will review the epidemiology, pathogenesis, clinical features, and diagnosis of CVT. Treatment and prognosis are discussed separately. (See "Cerebral venous thrombosis: Treatment and prognosis".)
CVT in newborns is also reviewed elsewhere. (See "Stroke in the newborn: Classification, manifestations, and diagnosis", section on 'Cerebral sinovenous thrombosis'.)
EPIDEMIOLOGY — The available data suggest that CVT is uncommon [1]. The annual incidence ranges from 1.16 to 2.02 per 100,000 [2-4] and is more common in females than males, with a female-to-male ratio of 3:1 [5,6]. The imbalance may be due to the increased risk of CVT associated with pregnancy and puerperium and with oral contraceptives [7]. (See 'Acquired risk factors' below.)
In adults, CVT affects patients who are younger on average than those with arterial types of stroke. In the International Study on Cerebral Vein and Dural Sinus Thrombosis (ISCVT), the median age of patients with CVT was 37 years [5], and only 8 percent of the patients were older than 65 [8]. Compared with males, females were significantly younger (median age 34 years, versus 42 years for men) [6].
There is a low risk of recurrent CVT and venous thromboembolism after a first CVT, as reviewed separately. (See "Cerebral venous thrombosis: Treatment and prognosis", section on 'Recurrence'.)
PATHOGENESIS — The pathogenesis of CVT remains incompletely understood because of the high variability in the anatomy of the venous system and the paucity of experiments in animal models of CVT. However, there are at least two different mechanisms that may contribute to the clinical features of CVT (figure 1) [9]:
●Thrombosis of cerebral veins or dural sinus obstructs blood drainage from brain tissue, leading to cerebral parenchymal lesions (eg, stroke) or dysfunction and to increased venous and capillary pressure with disruption of the blood-brain barrier.
●Occlusion of dural sinus resulting in decreased cerebrospinal fluid (CSF) absorption and elevated intracranial pressure.
Obstruction of the venous structures (figure 2) results in increased venous pressure, decreased capillary perfusion pressure, and increased cerebral blood volume. Dilatation of cerebral veins and recruitment of collateral pathways play an important role in the early phases of CVT and may initially compensate for changes in pressure.
The increase in venous and capillary pressure leads to blood-brain barrier disruption, causing vasogenic edema, with leakage of blood plasma into the interstitial space. As intravenous pressure continues to increase, localized cerebral edema and venous hemorrhage may occur due to venous or capillary rupture. The increased intravenous pressure may lead to an increase in intravascular pressure and a lowering of cerebral perfusion pressure, resulting in decreased cerebral blood flow (CBF) and failure of energy metabolism. In turn, this allows intracellular entry of water from failure of the Na+/K+ ATPase pump and consequent cytotoxic edema [10]. Venous infarction and hemorrhage may be confluent (ie, venous hemorrhagic infarction).
Advances in understanding the pathophysiology of venous occlusion have been aided by the use of MRI, mainly diffusion-weighted MRI and perfusion-weighted MRI [11-14]. These techniques have demonstrated the coexistence of both cytotoxic and vasogenic edema in patients with CVT [11,13-15].
The other effect of venous thrombosis is impairment of CSF absorption. Normally, CSF absorption occurs in the arachnoid granulations and glymphatic system, which drain CSF into the venous system. Thrombosis of the dural sinuses leads to increased venous pressure, impaired CSF absorption, and consequently elevated intracranial pressure. Elevated intracranial pressure is more frequent if superior sagittal sinus thrombosis is present, but it may also occur with thrombosis of the jugular vein or the lateral sinus. Note that the lateral sinus consists of two segments; the proximal segment is termed the transverse sinus, and the distal segment is termed the sigmoid sinus (figure 2).
RISK FACTORS AND ASSOCIATED CONDITIONS — Many conditions are associated with CVT. The major risk factors for CVT in adults can be grouped as transient or permanent (table 1). The most frequent risk factors for CVT are [1,5]:
●Prothrombotic conditions, either genetic or acquired
●Obesity
●Oral contraceptives
●Pregnancy and the puerperium
●Malignancy
●Infection
●Head injury and mechanical precipitants
In more than 85 percent of adult patients, at least one risk factor for CVT can be identified, most often an inherited or acquired prothrombotic condition [5]. In the Canadian pediatric ischemic stroke registry, a risk factor was identified in 98 percent of the children [16]. A prothrombotic state was found in 41 percent. In infants older than four weeks of age and in children, head and neck disorders, mostly infections and chronic systemic diseases (eg, connective tissue disease, hematologic disorder, and cancer) were common. The most common risk factors in those ≥65 years old are genetic or acquired thrombophilia, malignancy, and hematologic disorders such as polycythemia [8,17].
Acquired risk factors — The most common acquired risk factors are pregnancy and the puerperium, the use of oral contraceptives, malignancy, and obesity [18,19]. (See "Cerebrovascular disorders complicating pregnancy" and "Combined estrogen-progestin contraception: Side effects and health concerns" and "Risk and prevention of venous thromboembolism in adults with cancer".)
In the prospective International Study on Cerebral Vein and Dural Sinus Thrombosis (ISCVT) cohort of 624 adults with CVT, females comprised 75 percent [6]. Furthermore, a sex-specific risk factor (ie, oral contraceptives, pregnancy, puerperium, and hormone replacement therapy) was identified in 65 percent of females. Estrogen receptor modulators, such as tamoxifen, and hormone replacement therapy have been associated with CVT in case reports [20-23]. In an earlier report of the ISCVT cohort, a prothrombotic condition was found in 34 percent of all patients, and a genetic prothrombotic condition was found in 22 percent of all patients [5].
●Oral contraceptives – The most frequent risk factor for CVT in younger female patients is the use of oral contraceptives [24,25]. Furthermore, the risk for CVT in females using oral contraceptives is increased in the presence of a prothrombotic defect and obesity [25,26]. (See "Combined estrogen-progestin contraception: Side effects and health concerns", section on 'Venous thromboembolism'.)
●Chemotherapeutic agents – Several medications used to treat cancer (eg, L-asparaginase, all-trans retinoic acid, and cisplatin) may increase the risk of venous thromboembolism including CVT through procoagulant effects. (See "Cancer-associated hypercoagulable state: Causes and mechanisms", section on 'Therapy-related factors'.)
Obesity is a risk factor for CVT and other forms of venous thromboembolism. In an observational study of 186 patients with CVT and matched controls, obesity was associated with an elevated risk of CVT for females (adjusted odds ratio [aOR] 3.5, 95% CI 2.0-6.1) but not males (aOR 1.2, 95% CI 0.3-5.3) [26]. The risk was highest for females with obesity using oral contraceptive medications (aOR 29.3, 95% CI 13.5-63.6). (See "Overview of the causes of venous thrombosis", section on 'Obesity'.)
Genetic thrombophilia — The risk for CVT is influenced by the individual's genetic background [27]. In the presence of some prothrombotic conditions, patients are at an increased risk of developing a CVT when exposed to a precipitant such as head trauma, lumbar puncture, jugular catheter placement, pregnancy, surgery, infection, and drugs. These prothrombotic conditions include the following:
●Antithrombin deficiency [28,29]
●Protein C deficiency or protein S deficiency [16,30,31]
●Factor V Leiden pathologic variant [24,32,33]
●G20210 A prothrombin gene pathologic variant [24,33-35]
●Hyperhomocysteinemia [36]
In a meta-analysis of case-control studies, with over 200 neonatal and pediatric cases of sinovenous thrombosis (ie, CVT) and 1200 control subjects, the prevalence of factor V Leiden (FVL) variant among cases and controls was 12.8 and 3.6 percent, respectively, and carriers of the FVL variant were significantly more likely to develop CVT (odds ratio [OR] 3.1, 95% CI 1.8-5.5) [37]. Similarly, the prevalence of the prothrombin gene variant among cases and controls was 5.2 and 2.5 percent, respectively, and carriers were significantly more likely to develop CVT (OR 3.1, 95% CI 1.4-6.8).
The association of CVT with hyperhomocysteinemia due to genetic variants in methylene tetrahydrofolate reductase (MTHFR) is controversial [27,38,39].
A 2010 meta-analysis of case-control studies found that the frequency of the MTHFR 677C>T polymorphism in adults was similar for 382 patients with CVT compared with 1217 controls (15.7 versus 14.6 percent; OR 1.12, 95% CI 0.8-1.58), suggesting that the MTHFR 677C>T polymorphism is not a risk factor for CVT [40]. In contrast, a 2011 meta-analysis, after controlling for heterogeneity among studies, found that the MTHFR 677C>T polymorphism was associated with CVT (OR 2.30, 95% CI 1.20-4.42) [27].
There is no association of CVT with PAI-1 or protein Z polymorphisms.
COVID-19 infection and COVID-19 vaccine-associated thrombosis — Several cases of CVT have been observed in the setting of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, the majority of which occurred in patient without other predisposing risk factors [41]. In a review of 34,331 patients hospitalized with SARS-CoV-2 infection, the frequency of CVT was 0.08 percent (95% CI 0.01-0.5) [41]. The in-hospital mortality was 40 percent. (See "COVID-19: Neurologic complications and management of neurologic conditions", section on 'Cerebrovascular disease'.)
Multiple reports describe thromboembolism including CVT associated with thrombocytopenia among patients immunized with the adenovirus-vector AstraZeneca (ChAdOx1 nCov-19) COVID-19 and Janssen/Johnson & Johnson (Ad26.COV2.S) COVID-19 vaccines [42-48]. In cohort studies, those with vaccine-induced thrombotic thrombocytopenia (VITT) have been typically younger and less likely to have other thromboembolic risk factors than those with CVT not attributed to VITT [47,49,50]. However, patients with VITT have a higher burden of both intracranial and extracranial venous thromboses [49]. In addition, the clinical presentation is more severe and mortality is higher for those with CVT associated with VITT than other causes of CVT, ranging from 22 to 47 percent in studies, compared with 3 to 5 percent among those with other causes of CVT [46,47,49-51].
Cases of CVT associated with VITT have typically occurred in patients who are between 5 and 30 days post-vaccination [42-44,52,53]. Some cases occurring after the second dose of the AstraZeneca COVID-19 vaccine have also been reported [54]. The evaluation and management of patients with VITT is discussed in greater detail separately. (See "COVID-19: Vaccine-induced immune thrombotic thrombocytopenia (VITT)" and "COVID-19: Vaccines", section on 'Thrombosis with thrombocytopenia'.)
Other conditions — Although infectious causes of CVT were frequently reported in the past, they are responsible for only 6 to 12 percent of cases in modern-era studies of adults with CVT [5,18]. Local infections (eg, involving the ears, sinuses, mouth, face, or neck) are typically responsible, although systemic infection is sometimes the only cause.
Head injury and mechanical precipitants are less common causes of CVT [55-57].
Inflammatory diseases are also risk factors for CVT, including systemic lupus erythematosus, Behçet disease, granulomatosis with polyangiitis, thromboangiitis obliterans, inflammatory bowel disease, and sarcoidosis.
As with venous thrombosis in other parts of the body, multiple risk factors may be found in about half of adult patients with CVT [5]. In light of this, a thorough search for additional causes should be carried out even when a specific risk factor is identified in a given patient. (See "Overview of the causes of venous thrombosis".)
Cryptogenic CVT — No underlying etiology or risk factor for CVT is found in a minority of children (≤10 percent) and adults (13 percent) with CVT [8,16,58]. In older adult CVT patients, the proportion of cases without identified risk factors is higher (37 percent) than it is in adults under age 65 (10 percent) [8].
CLINICAL ASPECTS — Cerebral vein and dural sinus thrombosis has a highly variable clinical presentation [59,60]. The onset can be acute, subacute, or chronic. CVT most often presents with new headache or as a syndrome of isolated intracranial hypertension. Additional manifestations include focal neurologic deficits, seizures, and/or encephalopathy.
Symptoms and signs — Symptoms and signs of CVT can be grouped into three major syndromes:
●Isolated intracranial hypertension syndrome (headache with or without vomiting, papilledema, and visual problems) [61]
●Focal syndrome (focal deficits, seizures, or both)
●Encephalopathy (multifocal signs, mental status changes, stupor, or coma) [59,62]
Less common presentations include cavernous sinus syndrome, subarachnoid hemorrhage, and multiple cranial nerve palsies. A case of CVT mimicking a transient ischemic attack has also been reported [63].
The clinical symptoms and signs in CVT depend upon several factors, including patient age and sex, the site and number of occluded sinuses and veins, the presence of parenchymal brain lesions, and the interval from CVT onset to presentation. In children, signs of diffuse brain injury, coma, and seizures are the main clinical manifestations, especially in neonates [16]. In older children, the manifestations of CVT resemble those in adults, with headache and hemiparesis [64]. Females are more likely than males to have a headache on presentation and less likely to have a chronic onset of symptoms [6]. Older adults may also have a distinctive presentation; depressed consciousness and mental status changes are more common, while headaches and isolated intracranial hypertension are less frequent than in younger patients [8].
Cerebral edema, venous infarction, and hemorrhagic venous infarction are associated with a more severe syndrome; patients are more likely to be comatose or to have motor deficits, aphasia, and seizures and less likely to present with isolated headache.
Headache — Headache is the most frequent symptom of CVT. In the International Study on Cerebral Vein and Dural Sinus Thrombosis (ISCVT) cohort, headache was present in 89 percent of patients [5]. Headaches associated with CVT are more frequent in females and young patients than in males or older adults [65]. Headache is usually the first symptom of CVT and can be the only symptom [66] or can precede other symptoms and signs by days or weeks [67].
The features of CVT-related headache are quite variable. Head pain may be localized or diffuse [67]. Headache caused by intracranial hypertension from CVT is typically characterized by severe head pain that worsens with Valsalva maneuvers and with recumbency.
The site of the headache has no relationship with the localization of the occluded sinus or the parenchymal lesions [68,69]. Headache onset with CVT is usually gradual, increasing over several days [7]. However, some patients with CVT have sudden explosive onset of severe head pain (ie, thunderclap headache) that mimics subarachnoid hemorrhage [70,71]. (See "Overview of thunderclap headache" and "Aneurysmal subarachnoid hemorrhage: Clinical manifestations and diagnosis".)
Headache due to CVT may also resemble migraine with aura [72-74]. (See "Pathophysiology, clinical manifestations, and diagnosis of migraine in adults" and "Pathophysiology, clinical features, and diagnosis of migraine in children".)
CVT must be included as a possible cause of persisting headache following lumbar puncture, because lumbar puncture can rarely precipitate a CVT. (See "Post dural puncture headache".)
Isolated intracranial hypertension syndrome — Isolated intracranial hypertension syndrome (ie, headache associated with papilledema or visual problems) accounts for a significant proportion of CVT cases [61]. Visual obscurations may occur, coinciding with bouts of increased headache intensity.
Isolated intracranial hypertension is more frequent in patients with a chronic presentation than in those who present acutely [75]. In addition, patients with chronic course or delayed clinical presentation may show papilledema on fundoscopy, a finding that is less frequent in acute cases. (See "Overview and differential diagnosis of papilledema".)
Seizures — Focal or generalized seizures, including status epilepticus, are more frequent in CVT than in other cerebrovascular disorders. In the ISCVT cohort of 624 patients, seizures at presentation occurred in 39 percent, and seizures after the diagnosis of CVT occurred in 7 percent [76]. In a retrospective cohort of 131 children (including 33 neonates) with CVT, seizures at presentation occurred in 13 of 33 neonates (42 percent) and 19 of 98 children (19 percent) [77]. Variables associated with seizures include supratentorial parenchymal brain lesions, sagittal sinus and cortical vein thrombosis, and motor deficits [76].
Encephalopathy — Severe cases of CVT can cause disturbances of consciousness and cognitive dysfunction, such as delirium, apathy, a dysexecutive syndrome, multifocal deficits, or seizures.
Focal syndrome — Weakness with monoparesis or hemiparesis, sometimes bilateral, is the most frequent focal deficit associated with CVT. In the ISCVT cohort, motor weakness was present in 37 percent of patients [5]. Aphasia, in particular of the fluent type, may follow sinus thrombosis, especially when the left lateral sinus is affected. Sensory deficits and visual field defects are less common.
Syndromes associated with isolated sinus or vein thrombosis — Isolated thrombosis of the different sinuses and veins produces diverse clinical pictures.
●In cavernous sinus thrombosis, ocular signs dominate the clinical picture with orbital pain, chemosis, proptosis, and oculomotor palsies [78-81].
●Isolated cortical vein occlusion produces motor/sensory deficits and seizures [82-84].
●With sagittal sinus occlusion, motor deficits, bilateral deficits, and seizures are frequent, while presentation as an isolated intracranial hypertension syndrome is infrequent.
●Patients with isolated lateral sinus thrombosis frequently present with isolated headache or isolated intracranial hypertension [85]. Less often, they may also present with focal deficits or seizures. Aphasia often follows if the left transverse sinus is occluded.
●Jugular vein or lateral sinus thrombosis may present as isolated pulsating tinnitus [86,87].
●Multiple cranial nerve palsies may occur in thrombosis of the lateral sinus, jugular, or posterior fossa veins [88].
●When the deep cerebral venous system (ie, the straight sinus and its branches) is occluded, the signs and symptoms of CVT are generally severe, with coma or other alterations in mental status and motor deficits, often bilateral [89-91]. However, more limited thrombosis of the deep venous system can produce relatively mild symptoms [92].
Neuroimaging — The neuroimaging features of CVT can include focal areas of edema or venous infarction, hemorrhagic venous infarction, diffuse brain edema, or (rarely) isolated subarachnoid hemorrhage [1]. In patients with CVT, the proportion who present with intracerebral hemorrhage is 30 to 40 percent [93,94]. Small nontraumatic juxtacortical hemorrhages (image 1), which are located just below the cortex in the white matter and have a diameter of <2 cm, account for up to one-fourth of intracerebral hemorrhages in patients with CVT and are associated with superior sagittal sinus occlusion [95].
In a minority of cases, computed tomography (CT) may demonstrate direct signs of CVT, which include the dense triangle sign, the empty delta sign, and the cord sign, described below (see 'CT' below). Brain MRI in combination with magnetic resonance venography is the most informative technique for demonstrating the presence of dural thrombus, cortical vein thrombosis, and extent of brain injury. The imaging findings of CVT are discussed in greater detail below. (See 'Urgent imaging' below.)
DIAGNOSIS — In patients with clinically suspected CVT (eg, presenting with new headache, isolated intracranial hypertension syndrome, focal neurologic deficits, seizures, and/or encephalopathy), urgent neuroimaging is necessary as the first step in the diagnostic evaluation. Aside from neuroimaging, there is no simple confirmatory laboratory test that can confidently rule out CVT in the acute phase of the disease.
Diagnostic approach — The diagnosis of CVT should be suspected in patients who present with one or more of the following:
●New-onset headache
●Headache with features that differ from the usual pattern (eg, progression or change in attack frequency, severity, or clinical features) in patients with a previous primary headache
●Symptoms or signs of intracranial hypertension
●Encephalopathy
●Focal neurologic symptoms and signs, especially those not fitting a specific vascular distribution or those involving multiple vascular territories
●Seizures
In addition, the diagnosis of CVT should be suspected in patients who have atypical neuroimaging features on routine CT or MRI at presentation, such as cerebral infarction that crosses typical arterial boundaries, hemorrhagic infarction, or lobar intracerebral hemorrhage of otherwise unclear origin [1]. In any of these scenarios, suspicion for CVT should be particularly high for patients with known risk factors, including prothrombotic conditions, oral contraceptive use, pregnancy and the puerperium, malignancy, infection, and head injury, even if the initial neuroimaging study (most often a CT) is normal.
Urgent imaging — For patients with any presentation raising concern for CVT, we recommend urgent neuroimaging with brain MRI and magnetic resonance (MR) venography or with cranial CT with CT venography if MRI is not an option [1]. The clear demonstration of absence of flow and intraluminal venous thrombus by CT or MRI is the most important finding for confirming the diagnosis. However, these findings are not always evident, and the diagnosis may rest on imaging features demonstrated by MR venography or CT venography showing only absence of flow in a venous sinus or cortical vein.
A number of normal anatomic variants may mimic sinus thrombosis, including sinus atresia, sinus hypoplasia, asymmetric sinus drainage, and normal sinus filling defects associated with arachnoid granulations or intrasinus septa [1]. For example, a study of 100 subjects (without CVT) with normal brain MRI found artifactual transverse sinus flow gaps on MR venography (in nondominant or codominant but not in dominant transverse sinuses) in 31 percent [96]. Another report of 100 subjects without venous pathology found asymmetric lateral sinuses in 49 percent and partial or total absence of one lateral sinus in 20 percent [97].
CT — Head CT is normal in up to 30 percent of CVT cases, and most of the findings are nonspecific [59]. However, CT is often the first investigation to be performed in clinical practice, and it is useful to rule out other acute or subacute cerebral disorders.
In about one-third of cases, CT demonstrates direct signs of CVT, including the following (image 2 and image 3) [59,98-100]:
●The cord sign is a curvilinear or linear hyperdensity from a thrombosed cortical vein typically found over the cerebral cortex.
●The dense triangle sign is a triangular or round hyperdensity reflecting a thrombosed sinus on a cross-section view.
●The empty delta sign (also called the empty triangle or negative delta sign) is a triangular pattern of contrast enhancement surrounding a central region without contrast enhancement found in the posterior part of the superior sagittal sinus on head CT performed with contrast.
Indirect signs of CVT on head CT are more frequent. These can include intense contrast enhancement of falx and tentorium, dilated transcerebral veins, small ventricles, and parenchyma abnormalities. In addition, associated brain lesions may be depicted in 60 to 80 percent of patients with CVT. These may be hemorrhagic or nonhemorrhagic:
●Hemorrhagic lesions include intracerebral hemorrhage, hemorrhagic infarcts, or rarely (<1 percent) subarachnoid hemorrhage usually limited to the convexity [101-103].
●Nonhemorrhagic lesions include focal areas of hypodensity caused by vasogenic edema or venous infarction, usually not respecting the arterial boundaries, as well as diffuse brain edema. With serial imaging, some lesions may disappear ("vanishing infarcts"), and new lesions may appear.
CT venography — Head CT is often normal in patients with CVT, and MRI techniques for confirming the diagnosis are not readily available in some hospitals and geographic locations. In this situation, we suggest CT venography as a useful alternative to MR venography or digital subtraction angiography for the diagnosis of CVT, in agreement with guidelines from the United States and Europe [1,104]. The overall accuracy of head CT combined with CT venography is 90 to 100 percent, depending on the occlusion site [105]. Compared with digital subtraction intra-arterial angiography, the combination of head CT and CT venography has a sensitivity and specificity of 95 and 91 percent [106].
CT venography gives a good visualization of the major dural sinuses [107,108], is readily available, and is quicker than MRI. It can be used for patients who have contraindications to MRI [1]. When combined with head CT, it may demonstrate filling defects, sinus wall enhancement, and increased collateral venous drainage (image 4) [106,109,110]. CT venography is often particularly helpful in subacute or chronic CVT because it can demonstrate heterogeneous density in thrombosed venous sinuses. However, its use may be limited because of low resolution of the deep venous system and cortical veins, the risk of contrast reactions, and radiation exposure [105,111,112].
MRI — MRI using gradient echo T2* susceptibility-weighted sequences in combination with MR venography is the most sensitive imaging method for demonstrating the thrombus and the occluded dural sinus or vein (image 4 and image 5) [15,111,113-116]. The characteristics of the MRI signal depend on the age of the thrombus [117,118]:
●In the first five days, the thrombosed sinuses appear isointense on T1-weighted images and hypointense on T2-weighted images.
●Beyond five days, venous thrombus becomes more apparent because signal is increased on both T1- and T2-weighted images.
●After the first month, thrombosed sinuses exhibit a variable pattern of signal, which may appear isointense.
On T2*-weighted gradient echo and T2* susceptibility-weighted MRI sequences, the acute thrombus can be directly visualized as an area of hypointensity in the engorged sinus or cortical vein (image 5) [84,113,119,120]. In addition, a chronically thrombosed sinus may also demonstrate low signal on these sequences. Limited data from a series of 28 patients with CVT suggest that the presence of hyperintensities in the veins or sinuses on diffusion-weighted MRI sequences predicts a low recanalization rate [121].
Parenchymal brain lesions secondary to venous occlusion, including brain swelling, vasogenic edema, or venous infarction, are hypointense or isointense on T1-weighted MRI and hyperintense on T2-weighted MRI (image 6). Venous congestion may show reversible reduced diffusivity on diffusion-weighted MRI sequences.
Agreement among observers for the diagnosis of CVT with MRI varies with the location of sinus or vein thrombosis. It is good or very good for most of the occluded sinus and veins, moderate to very good for the left lateral sinus and straight sinus, and poor to good for the cortical veins [122]. The diagnosis of isolated cortical vein thrombosis can be difficult, but the use of T2* susceptibility-weighted MRI may enable a diagnosis of isolated cortical vein thrombosis by demonstrating thrombus as an area of hypointensity [84,113,119].
MR venography — MR venography, usually performed using the time-of-flight (TOF) technique, is useful for demonstrating absence of flow in cerebral venous sinuses, though interpretation can be confounded by normal anatomic variants such as sinus hypoplasia and asymmetric flow [1]. Other MR techniques may be useful to distinguish these variants from venous thrombosis. Contrast-enhanced MR venography can provide better visualization of cerebral venous channels, and gradient echo or susceptibility-weighted sequences will show normal signal in a hypoplastic sinus and abnormally low signal in the presence of thrombus. A chronically thrombosed hypoplastic sinus will show absence of flow on two-dimensional TOF MR venography and enhancement on contrast-enhanced MRI and MR venography.
Conventional angiography — Cerebral digital subtraction angiography is typically reserved for cases when the clinical suspicion for CVT is high but CT venography or MR venography are inconclusive [1]. Angiography may be helpful for making this diagnosis by showing the sudden termination of a cortical vein surrounded by dilated and tortuous collateral "corkscrew veins" or by the filling of a cortical vein that was not apparent on an earlier angiographic study during the acute phase of CVT. Other typical signs of CVT on angiography are nonvisualization of all or part of a venous sinus, delayed venous emptying with pathologically increased collaterals, and reversal of venous flow.
As with MR and CT venography, conventional cerebral angiography may be limited by potential pitfalls. Anatomic variations, such as variability of number and location of cortical veins, hypoplasia of the anterior part of the superior sagittal sinus, duplication of the superior sagittal sinus, and hypoplasia or aplasia of the transverse sinuses, may make the diagnosis of CVT by all types of angiography difficult [59]. While the interobserver agreement for a diagnosis of CVT is not perfect, the combination of conventional contrast angiography plus brain MRI has a higher interobserver agreement than angiography alone (94 versus 62 percent) [123].
Laboratory tests — Aside from neuroimaging, there is no simple confirmatory laboratory test that can confidently rule out CVT in the acute phase of the disease. We suggest routine blood studies consisting of a complete blood count, chemistry panel, prothrombin time, and activated partial thromboplastin time for patients with suspected CVT, in agreement with guidelines from the American Heart Association/American Stroke Association [1]. The findings from these tests may suggest the presence of conditions that contribute to the development of CVT such as an underlying hypercoagulable state, infection, or inflammatory process. The guidelines recommend screening for these and other potential prothrombotic conditions that may predispose to CVT, including use of contraceptives, at the initial clinical presentation.
The utility of D-dimer testing and lumbar puncture is reviewed in the following sections.
D-dimer — An elevated plasma D-dimer level supports the diagnosis of CVT, but a normal D-dimer does not exclude the diagnosis in patients with suggestive symptoms and predisposing factors.
The potential utility of D-dimer for the diagnosis of CVT is illustrated by the following observations:
●A 2012 meta-analysis included 14 studies that evaluated D-dimer in 1134 patients for the diagnosis of suspected or confirmed CVT [124]. In seven studies that evaluated patients with suspected CVT, D-dimer was elevated in 145 of 155 patients in whom CVT was confirmed and was normal in 692 of 771 patients in whom CVT was ruled out, yielding a sensitivity and specificity of 94 and 90 percent, respectively. D-dimer performed less well in seven studies that enrolled subjects with already-confirmed CVT; the sensitivity and specificity were 89 and 83 percent, respectively. The sensitivity of D-dimer for CVT was also lower in patients with isolated headache as the presenting symptom (82 percent), in those with subacute or chronic clinical presentations of CVT (83 percent), and in those with a single affected venous sinus (84 percent).
●In a subsequent study of 233 patients with suspected CVT and symptom onset of less than seven days, D-dimer demonstrated a sensitivity and specificity of 94 and 98 percent, respectively, for predicting CVT [125].
Thus, D-dimer measurement may have some value as a diagnostic screening tool for the assessment of patients with possible CVT. However, a normal D-dimer value cannot exclude CVT, especially in patients with isolated headache or with thrombosis of a single sinus. Individual assays used to measure D-dimer vary, but it is reasonable to use the same threshold levels as used in diagnostic protocols for deep venous thrombosis (eg, D-dimer >500 ng/mL of fibrinogen equivalent units). (See "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity", section on 'D-dimer'.)
Lumbar puncture — Lumbar puncture may be useful to exclude meningitis in patients with CVT who present with isolated intracranial hypertension, a syndrome that may account for up to 25 percent of all patients with CVT [5]. In addition, lumbar puncture is valuable in such patients to measure and decrease cerebrospinal fluid (CSF) pressure when vision is threatened. However, in the absence of suspicion for meningitis, CSF analysis is usually not helpful diagnostically for patients with focal neurologic findings and neuroimaging confirmation of CVT [1].
The CSF abnormalities in CVT are nonspecific and may include a lymphocytic pleocytosis, elevated red blood cell count, and elevated protein; these abnormalities are present in 30 to 50 percent of patients with CVT [60,61].
Performing a lumbar puncture is not harmful in patients with CVT, as suggested by the findings of a study that analyzed 624 patients with CVT and identified 224 who had lumbar puncture [126]. The groups with and without lumbar puncture did not differ on any of the outcome measures, which were neurologic worsening within 30 days of CVT onset, acute death, complete recovery at six months, or death or dependency at six months. Nevertheless, lumbar puncture is contraindicated in patients with large brain lesions because they have an increased risk of herniation.
Evaluation for thrombophilic state — Searching for a thrombophilic state, either genetic or acquired, should be done for patients with CVT who have a high pretest probability of severe thrombophilia, a category that includes those with a personal and/or family history of venous thrombosis, CVT at a young age, and CVT in the absence of a transient or permanent risk factor (table 1) [104]. When appropriate, screening should include:
●Antithrombin
●Protein C
●Protein S
●Factor V Leiden
●Prothrombin G20210A pathologic variant
●Lupus anticoagulant, anticardiolipin, and anti-beta2 glycoprotein-I antibodies
●Homocysteine
Acute thrombosis can transiently reduce levels of antithrombin, protein C, and protein S, so the utility of testing for these disorders in the acute phase of CVT is limited. In practice, it is preferable to test for protein C, protein S, and antithrombin at least two weeks after oral anticoagulation has been discontinued, since warfarin therapy reduces measurements of protein C and protein S and may raise plasma antithrombin concentrations into the normal range in patients with hereditary antithrombin deficiency. It is possible to test for protein C and protein S levels while receiving heparin therapy, which does not alter plasma protein C or protein S concentrations. However, testing for antithrombin should be performed when off heparin, which can lower antithrombin levels. (See "Antithrombin deficiency" and "Protein C deficiency" and "Protein S deficiency".)
No underlying etiology or risk factor for CVT is found in approximately 13 percent of adult patients. However, it is important to continue searching for a cause even after the acute phase of CVT, as some patients may have a condition such as the antiphospholipid syndrome, polycythemia, thrombocythemia, malignancy, or inflammatory bowel disease that is discovered weeks or months after the acute phase. (See 'Risk factors and associated conditions' above.)
If abnormal results are found in assays for lupus anticoagulant, anticardiolipin, or anti-beta2 glycoprotein-I antibodies, testing should be repeated at least 12 weeks later, as the diagnosis of antiphospholipid syndrome requires two positive determinations of these biomarkers. (See "Diagnosis of antiphospholipid syndrome", section on 'Antiphospholipid antibody testing'.)
An evaluation for paroxysmal nocturnal hemoglobinuria should be pursued if the complete blood count shows unexplained hemolytic anemia, iron deficiency, or pancytopenia. (See "Clinical manifestations and diagnosis of paroxysmal nocturnal hemoglobinuria", section on 'Diagnosis and classification'.)
In patients older than 40 years without identified etiology, we suggest searching for an occult malignancy. In patients with sepsis or with fever and no obvious cause of infection, we recommend performing a lumbar puncture.
DIFFERENTIAL DIAGNOSIS — The clinical presentation of CVT can be nonspecific (eg, headache, seizure, encephalopathy) and the cause may not be apparent on initial routine neuroimaging studies.
●For patients presenting with symptoms and signs of isolated intracranial hypertension syndrome (headache with or without vomiting, papilledema, and visual problems), the main considerations in the differential are idiopathic intracranial hypertension (pseudotumor cerebri) and meningitis. Other conditions associated with elevated intracranial pressure (eg, intracranial mass lesions from tumor or abscess) are usually apparent on neuroimaging with CT or MRI. If neuroimaging reveals no structural intracranial lesion responsible for intracranial hypertension, a lumbar puncture is indicated with measurement of opening pressure and cerebrospinal fluid for analysis. (See "Idiopathic intracranial hypertension (pseudotumor cerebri): Clinical features and diagnosis".)
●For patients presenting with a focal neurologic syndrome (eg, focal deficits, seizures, or both) the differential is broad and includes other vascular etiologies (eg, intracerebral hemorrhage from a variety of other causes, subdural hemorrhage, ischemic stroke), infection (eg, meningitis, abscess), and tumor.
●For patients presenting with encephalopathy (eg, multifocal signs, mental status changes, stupor, or coma), the differential includes infection (eg, bacterial and viral meningoencephalitis), inflammation (eg, paraneoplastic and autoimmune encephalitis), demyelination (eg, acute disseminated encephalomyelitis, neuromyelitis optica spectrum disorders), and toxic and metabolic disturbances.
●For patients presenting with thunderclap headache, which is rare in CVT, the differential (table 2) includes subarachnoid hemorrhage, other types of intracranial hemorrhage, reversible cerebral vasoconstriction syndromes (RCVS), cervical artery dissection, viral and bacterial meningitis, acute complicated sinusitis, spontaneous intracranial hypotension, ischemic stroke, acute hypertensive crisis, third ventricular colloid cyst, and pituitary apoplexy. If initial neuroimaging is nondiagnostic, patients with thunderclap headache should have lumbar puncture with measurement of opening pressure and cerebrospinal fluid analysis to exclude subarachnoid hemorrhage and meningitis (algorithm 1). If lumbar puncture is also nondiagnostic, imaging of the cerebral circulation is necessary, preferably with magnetic resonance angiography/venography. (See "Aneurysmal subarachnoid hemorrhage: Clinical manifestations and diagnosis" and "Reversible cerebral vasoconstriction syndrome" and "Overview of thunderclap headache".)
●For patients during pregnancy or puerperium, preeclampsia or eclampsia are diagnostic considerations with any of the presentations listed above or otherwise when presenting with ischemic stroke or intracerebral hemorrhage. (See "Cerebrovascular disorders complicating pregnancy".)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Stroke in adults" and "Society guideline links: Stroke in children".)
SUMMARY AND RECOMMENDATIONS
●Epidemiology and risk factors – Cerebral venous thrombosis (CVT) is uncommon, with an estimated incidence of <2.5 per 100,000 annually. Among young adults, CVT is more common in females than males. The mean age of onset is 39 years old. (See 'Epidemiology' above.)
The major risk factors for CVT in adults (table 1) are prothrombotic (hypercoagulable) conditions, oral contraceptives, pregnancy and the puerperium, malignancy, obesity, infection, head injury, and mechanical precipitants. (See 'Risk factors and associated conditions' above.)
●Clinical features – The clinical presentation of CVT is highly variable. The onset can be acute, subacute, or chronic. Headache (of gradual, acute, or thunderclap onset) is the most frequent symptom, occurring in almost 90 percent of patients, and may occur as part of an isolated intracranial hypertension syndrome, with or without vomiting, papilledema, and visual problems. In other cases, headache may be accompanied by focal neurologic deficits, focal or generalized seizures, and encephalopathy with altered mental status or coma. (See 'Clinical aspects' above.)
●Neuroimaging features – Parenchymal brain lesions, including brain swelling, edema, venous infarction, or hemorrhagic venous infarction, may occur secondary to venous occlusion. (See 'Neuroimaging' above.)
•Head CT scan is normal in up to 30 percent of CVT cases, and most of the findings with CVT are nonspecific. However, in about one-third of patients, CT demonstrates direct signs of CVT, which include the cord sign, the dense triangle sign, and the empty delta sign (image 2 and image 3).
CT venography may demonstrate filling defects, sinus wall enhancement, and increased collateral venous drainage and is an alternative to magnetic resonance (MR) venography (image 4 and image 2). (See 'CT' above and 'CT venography' above.)
•Brain MRI along with MR venography is the most informative technique for demonstrating the presence of dural thrombus, cortical vein thrombosis, and the extent of brain injury (image 6 and image 5). (See 'MRI' above and 'MR venography' above.)
●Diagnosis – The combination of an abnormal signal in a venous sinus on brain MRI and the corresponding absence of flow on MR venography confirms the diagnosis of CVT. However, these findings are not always evident, and the diagnosis may rest on imaging features showing only absence of flow in a venous sinus or cortical vein. Other than neuroimaging, there is no simple confirmatory laboratory test that can confidently rule out CVT in the acute phase of the disease. (See 'Diagnosis' above.)
●Evaluation for underlying thrombophilia – Screening for thrombophilia should be done for patients with CVT who have a high pretest probability for severe thrombophilia, a category that includes those with a personal and/or family history of venous thrombosis, CVT at a young age, and CVT in the absence of a transient or permanent risk factor (table 1). (See 'Evaluation for thrombophilic state' above.)
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟
