Description of the condition

Intracranial venous thrombosis (ICVT) is the obstruction of one or more of the cerebral veins or venous sinuses, which causes a rise of venous pressure in its drainage territory (that is, venous hypertension) leading to venous engorgement and brain oedema (Arquizan 2002). This can lead to reduced blood flow in the region. The affected cerebral cortex and underlying white matter may become congested, swollen and haemorrhagic, leading to venous infarction (Arquizan 2002).

ICVT was first described by Ribes in 1825. At that time the condition was largely regarded as fatal, with the majority of diagnoses being made at post‐mortem (Benamer 2000). More recently, the diagnosis of ICVT has become easier to confirm with new and less invasive neuroimaging techniques such as magnetic resonance imaging (MRI) (Connor 2002). This has resulted in early treatment and thus much better prognosis than was previously thought possible. In the International Study on Cerebral Vein and Dural Sinus Thrombosis, the 30‐day case fatality rate was only 3.4%, and the median time from onset to death was 13 days (Canhão 2005a). Another prospective study found that at six months after ICVT 16% of participants had died, 7% were dependent for daily activities, and 77% were independent (Stolz 2005). Treatments of ICVT with anticoagulation, thrombolysis, and corticosteroids have been subjects of a couple of Cochrane reviews (Ciccone 2004; Coutinho 2011).

The exact frequency of ICVT in the general population is unclear (Arquizan 2002). Nevertheless, it is relatively rare and accounts for less than 1% of all strokes (Masuhr 2004). ICVT can occur in all age groups, with the highest incidence being amongst neonates and adults in their third decade. ICVT is more common amongst females (male/female ratio of 1.5 to 5). Due to the wide spectrum of clinical symptoms and gradual onset of symptoms, the diagnosis is often missed or delayed. Important causes and predisposing factors include intracranial or systemic infections, coagulation disorders (for example protein C or S deficiency), vascular trauma (for example neurosurgery, head injury), space‐occupying lesions (for example intracranial and extracranial malignancies), hormonal factors (for example the oral contraceptive pill, pregnancy, puerperium), vasculitis (for example lupus erythematosus, Behçet's disease), metabolic disorders (for example homocystinuria, hyperhomocysteinaemia), and others (for example following a lumbar puncture). However, in up to a third of people there may be no identifiable cause (Allroggen 2000; Canhão 2005b; Heller 2003; Masuhr 2004).

There are several key pathophysiological differences between arterial and venous thrombosis in the brain. Firstly, ICVT is widely regarded as a continuing process in which the balance of pro‐thrombotic and thrombolytic processes is disturbed, leading to progression of the venous thrombus with time. This slow growth of the thrombus, together with good collateralisation of venous vessels, could explain why some people present with more gradual onset of symptoms, often over days or weeks. In one retrospective multicentre study of 48 participants 44% presented acutely, 35% presented subacutely, and 21% presented in a chronic state (Terazzi 2005). Secondly, haemorrhagic transformation occurs in a significant proportion of people, probably as a result of raised venous and capillary pressure at and around the site of occlusion (Allroggen 2000).

The clinical presentation of ICVT depends on the extent and site of venous occlusion, the progression of the thrombotic process, and the existence of venous collaterals (Masuhr 2004; Van den Bergh 2005). Occlusion of a large venous sinus may lead to more generalised neurological sequelae, such as intracranial hypertension, epileptic seizures, and altered consciousness, with a poorer outcome. Involvement of isolated cortical veins may present with more focal neurological symptoms such as motor or sensory deficits and focal seizures (Van den Bergh 2005). In some mild cases headache might be the only presenting feature (Cumurciuc 2005).

Description of the intervention

Compared with people with strokes caused by arterial occlusion, those with ICVT are much more likely to experience seizures at the initial presentation and during follow‐up (Buccino 2003; De Bruijn 2001; Ferro 2003; Masuhr 2004). In one series of 59 participants with ICVT, 47% experienced seizures on admission (De Bruijn 2001). In another series of 142 participants with ICVT, 34% experienced seizures within the first two weeks of the event, and 10% experienced seizures after the first two weeks (Ferro 2003). In yet another series of 77 participants, 36% experienced seizures in the acute period (undefined) but only 5% developed recurrent seizures (Preter 1996). The frequency of seizures may be even higher amongst children with ICVT. In a series of 58 children with ICVT, 58% experienced seizures on admission, and those presenting with seizures were more likely to have a bad outcome (DeVeber 2001). In Ferro 2003, early seizures were found to be more frequent in people with motor and sensory deficits and in those with computed tomography (CT) or MRI evidence of focal brain oedema, venous infarction, or intracerebral haemorrhage on admission. Furthermore, late seizures were more frequent in people with early seizures and CT or MRI evidence of haemorrhage. However, the relationship between post‐ICVT seizures and outcome remains unclear; some studies have identified seizures as a poor prognostic indicator (Benamer 2000; DeVeber 2001; Stolz 2005) whilst others have not found such a relationship (Ferro 2003).

How the intervention might work

The prophylactic use of antiepileptic drugs (AEDs) for the management of post‐ICVT seizures is controversial. In Ferro 2003, of the 31 participants who experienced early seizures after ICVT, 18 (58%) were prescribed AEDs, and of the 60 participants who did not experience early seizures 12 (20%) were prescribed AEDs. There is currently no consensus on the optimal management of post‐ICVT seizures. Whilst some experts recommend prophylactic treatment with AEDs for all people with ICVT because of the high risk of seizures and their potential adverse physiological effects (Einhaupl 1994), others have more recently recommended reserving prophylactic treatment only for those who have already had a seizure, that is, for secondary prevention only (Buccino 2003; Masuhr 2004). On the other hand, Ferro 2003 was specific in recommending that prophylactic treatment should be used in those with proven risk factors for seizures (for example those who have already experienced seizures or who have CT or MRI evidence of haemorrhage).

Why it is important to do this review

This is an update of a review in the Cochrane Database of Systematic Reviews first published in 2006 (Kwan 2006), which found no evidence to support or refute the use of AEDs for the primary or secondary prevention of seizures related to ICVT.

Overall, it remains unclear which people with ICVT should receive prophylactic AEDs and which drug should be used, at what dosage, and for how long (Masuhr 2004). Any potential benefits of using AEDs should also be balanced with their potential side effects. This systematic review aimed to review updated evidence to assess the potential benefits and risks of AEDs for the prevention of seizures related to ICVT.