ORIGINAL ARTICLE

Year : 2022  |  Volume : 39  |  Issue : 1  |  Page : 28-34

Evaluation of the relationship between epileptic seizures and type of parenchymal lesion in patients with cerebral venous thrombosis

Yasemin Dinc, Aylin Bican Demir, Mustafa Bakar, İbrahim Bora
Department of Neurology, Uludağ University Faculty of Medicine, Bursa, Turkey

Correspondence Address:
Yasemin Dinc
Uludag Üniversitesi Nöroloji, Polikliniği, Nilüfer
Turkey

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/nsn.nsn_139_21

Introduction: Epileptic seizures occur in approximately 35%–40% of patients with cerebral venous thrombosis (CVT). The relationship between parenchymal lesions and epileptic seizures in CVT has been investigated, but the most associated types of parenchymal lesions have not been determined. This study, therefore, aimed to identify high-risk groups. Methods: A total of 159 patients were diagnosed as having CVT between 2015 and 2021 at our tertiary center. The risk factors for epileptic seizures after CVT were determined. Results: A total of 159 patients who were diagnosed with having CVT, 109 (68.5%) females and 50 (31.5%) males, were included in this study. The mean ages of the women and men were 41.20 ± 14.15 years and 43.60 ± 16.30 years, respectively. We found that superior sagittal sinus involvement (P = 0.019), sigmoid sinus involvement (P = 0.010), cortical vein involvement (P < 0.001), parenchymal lesion (P < 0.001), and the postpartum period (P = 0.003) increased the risk of epileptic seizures. When the significant variables associated with epileptic seizures in the patients were analyzed using binary logistic regression, the most significant variable was found to be the presence of parenchymal lesions. Conclusion: We found that the most significant variable for epileptic seizures after CVT was parenchymal lesions. Juxtacortical hemorrhages and nonhemorrhagic venous infarcts were the most common causes of epileptic seizures. CVT is a heterogeneous group of diseases caused by multiple aetiologies and may show ethnic and racial differences. For this reason, more precise information can be obtained with multi-center prospective studies in our population.

Keywords: Cerebral venous thrombosis, epileptic seizures, parenchymal lesions

How to cite this article: Dinc Y, Demir AB, Bakar M, Bora &. Evaluation of the relationship between epileptic seizures and type of parenchymal lesion in patients with cerebral venous thrombosis. Neurol Sci Neurophysiol 2022;39:28-34

How to cite this URL: Dinc Y, Demir AB, Bakar M, Bora &. Evaluation of the relationship between epileptic seizures and type of parenchymal lesion in patients with cerebral venous thrombosis. Neurol Sci Neurophysiol [serial online] 2022 [cited 2023 May 29];39:28-34. Available from: http://www.nsnjournal.org/text.asp?2022/39/1/28/342363

Introduction

Cerebral venous thrombosis (CVT) is a rare form of venous thromboembolism.^[1] There are three to four new cases per million adults each year.^[2] CVT may present with different clinical symptoms, such as isolated intracranial hypertension, focal neurologic deficit, epileptic seizures, and encephalopathy. Epileptic seizures are a frequent clinical presentation of CVT^[3],[4],[5] that occur in approximately 35%–40% of patients with CVT^[6],[7],[8] either early (in the first 14 days) or late (after the 14^th day)^[5],[6] in the disease process. Focal neurologic deficits, supratentorial parenchymal lesions, cortical vein thrombosis, and superior sagittal sinus thrombosis are the main risk factors.^[3],[5],[7] Epileptic seizures are a poor prognostic factor in patients with CVT.^[6] Parenchymal lesions, such as nonhaemorrhagic venous infarcts, juxtacortical haemorrhage, subarachnoid haemorrhage, and parenchymal haematomas, may occur in CVT.^[9] The relationship between parenchymal lesions and epileptic seizures has been investigated, but the most associated parenchymal lesions have not been determined. This study, therefore, aimed to study which parenchymal lesions were related to the occurrence of epileptic seizures. Knowing factors that can lead to seizures is important for taking the necessary precautions. In this study, we determined the 1-year seizure rate of patients with CVT diagnoses who were followed in our clinic and identified high-risk groups.

Methods

A total of 159 patients who were diagnosed as having CVT between 2015 and 2021 at the Department of Neurology, Faculty of Medicine, XXX University, were retrospectively studied. Approval for the study was obtained from the clinical research ethics committee of the faculty (Date: July 30, 2021, Number: 2021-9/4). Patient consent was not required because this was a retrospective study. In our center, all medical records are stored in a medical operating system. All patients were examined by neurologists in the emergency room. Each patient's symptoms, neurologic examinations, diseases, and drugs used were recorded in an epicrisis. All patients with suspected CVT underwent noncontrast cranial computed tomography (CT), cranial magnetic resonance imaging, and 3T contrast-enhanced cranial magnetic resonance venography. Patients with CVT were admitted to the neurology ward for further examination and treatment and were evaluated for rheumatologic disease and hypercoagulopathy. The postdischarge follow-up of the patients was accomplished in the stroke outpatient clinic of Uludag University Medical School. All patients were followed up regularly for 1 year. The clinical outcomes of the patients were evaluated at 3 months. modified rankin scale scores of 0, 1, and 2 were considered good clinical outcomes, and scores of 3, 4, 5, and 6 were considered poor clinical outcomes. Parenchymal lesions were divided into nonhaemorrhagic venous infarcts and intracranial haemorrhages. As shown in [Figure 1], intracranial haemorrhages were classified into three categories: Small juxtacortical haemorrhages, large parenchymal haematomas, and subarachnoid haemorrhages. Juxtacortical haemorrhages were defined as haemorrhages that were <20 mm in concave diameter and located in the white matter adjacent to the cortex.^[10] Whether the patients experienced seizures was determined at the end of the 1^st year.

Figure 1: (a) Juxtacortical hemorrhage of 2–3 mm in the left frontoparietal cortex on noncontrast cranial computed tomography. (b) Scattered hematoma cavity approximately 5 cm in size in the right parietal region in flair sequence in cranial magnetic resonance imaging. (c) Hyperdense signal change in left frontoparietal cortical sulcuses compatible with subarachnoid hemorrhage on noncontrast cranial computed tomography. (d) In the temporal cortex on the left, an area compatible with venous infarct is observed in the temporal cortex with hyperintense T2A hyperintense and high B values Click here to view

Early (late) seizures after CVT were defined as seizures occurring within (after) the first 2 weeks.^[5],[6] The seizure classification of the patients was made according to the 2017 International League Against Epilepsy classification.^[11]

Frequent seizures were defined as one to three seizures per month, as reported by the patients during follow-up. Those who had fewer than one seizure per month were identified as having infrequent seizures.

Drug-resistant epilepsy is defined as seizures that persist in a patient despite the use of two appropriately selected anti-epileptic drugs (AEDs).^[12] The clinical, radiologic, and demographic features of the patients who did and did not have epileptic seizures after CVT were compared. The risk factors for epileptic seizures after CVT were then determined.

Statistical analysis

The radiologic, demographic, and clinical data of the patients with and without epileptic seizures after CVT were compared. Statistical analysis was performed using the IBM SPSS Statistics 25.0 package (IBM Corp., Armonk, New York, USA). The Shapiro-Wilk test, a histogram, and a Q-Q plot were used to determine the normality of data distribution. The means and standard deviations or medians (25%–75% quartiles) were then used for the analysis of the continuous variables. Frequencies and percentages are given for the categorical variables. Levene's test was used to determine the variance homogeneity. A two-sided independent sample t-test or a two-sided Mann–Whitney U test was used to analyze the differences in the continuous variables between the groups. A two-sided Fisher's exact test, Pearson test, Chi-square test, and continuity correction test for 2 × 2 or r × c tables were used to analyse the differences in the categorical variables between the groups. Binary logistic regression analysis was used to determine the significant risk factors for epileptic seizures after CVT. P < 0.05 were accepted as statistically significant.

Results

A total of 159 patients who were diagnosed with having CVT, 109 (68.5%) females and 50 (31.5%) males, were included in the study. The mean ages of the women and the men were 41.20 ± 14.15 years and 43.60 ± 16.30 years, respectively. An insignificant statistical relationship was found between sex and mean age (P = 0.32). Forty-six (28.9%) patients had epileptic seizures. Forty (25.1%) patients had early seizures (in the first 14 days), and 27 (16.9%) patients had late seizures (after 14 days). Twenty-one patients both early and late seizures.

All patients with epileptic seizures started undergoing AED therapy after the first seizure. Four patients experienced focal seizures with awareness, 28 patients had focal seizures with impaired awareness, and 14 patients had bilateral tonic-clonic seizures. During electroencephalography evaluations, slow-wave activity (delta and theta) was observed in 16 patients, epileptic activity (sharp and spike-wave) in 12 patients, and epileptic and slow-wave activity in 14 patients. The patients with epileptic seizures after CVT were followed up for an average of 38.6 ± 4.62 (range, 12–72) months to evaluate the frequency of seizures. Epileptic seizures recurred in 36 (22.6%) patients. Fifteen (9.4%) patients had frequent epileptic seizures, and 29 (18.2%) patients had rare seizures. Thirty-four (21.3%) patients were taking a single AED, whereas 12 (7.5%) were taking multiple AEDs. Ten (6.2%) patients were found to have drug-resistant epilepsy. The AEDs of the patients with monotherapy were levetiracetam (n = 17), carbamazepine (n = 9), oxcarbazepine (n = 7), and zonisamide (n = 1).

The treatments of the patients receiving multiple AED therapies were levetiracetam and carbamazepine (n = 6), levetiracetam and oxcarbazepine (n = 3), and levetiracetam and lacosamide (n = 3).

The epileptic seizures and the clinical, radiologic, and demographic characteristics of the patients with CVT were analyzed. Statistically significant results were obtained from superior sagittal sinus involvement (P = 0.019), sigmoid sinus thrombosis (P = 0.010), cortical vein thrombosis (P < 0.001), the presence of a parenchymal lesion (P < 0.001), being in the postpartum period (P = 0.003), poor clinical outcome (P < 0.001), death of CVT (P = 0.016), and the development of intracranial herniation (P = 0.001).

No statistically significant results were obtained from age, sex, the presence of malignancy, the presence of coagulopathy, the use of oral contraceptives, pregnancy, transverse sinus thrombosis, inferior sagittal sinus thrombosis, jugular vein thrombosis, and sinus rectus thrombosis (P > 0.05) [Table 1].

Table 1: Evaluation of epileptic seizure after cerebral venous thrombosis and their clinical, radiological, and demographic properties Click here to view

The significant variables associated with epileptic seizures in the patients were analyzed using binary logistic regression. The most significant variable was found to be the presence of parenchymal lesions (P < 0.001, odds ratio = 11.036) and cortical vein thrombosis (P = 0.020, odds ratio = 3.499) [Table 2].

Table 2: Logistic regression analysis of epileptic seizure after cerebral venous thrombosis based on clinical variables Click here to view

Parenchymal lesions due to CVT were grouped into four categories. In our study, there were 24 (15%) patients with nonhemorrhagic venous infarctions and 35 (22.1%) patients with haemorrhagic venous infarctions. Hemorrhagic venous infarcts were classified into three categories; juxtacortical hemorrhage was present in 23 (14.5%) patients, subarachnoid hemorrhage in four (2.5%) patients, and parenchymal haematoma in 12 (7.5%) patients [Table 3].{Table 2}

When the relationship between parenchymal lesion type and epileptic seizures after CVT was analyzed, a statistically significant relationship was found with juxtacortical hemorrhage (P < 0.001) and nonhemorrhagic venous infarcts (P = 0.047). By contrast, insignificant statistical results were obtained from subarachnoid hemorrhage and parenchymal haematoma (P > 0.05) [Table 3].

The relationship between parenchymal lesion type and early seizures was likewise examined. A statistically significant relationship was identified with juxtacortical hemorrhage (P < 0.001), subarachnoid hemorrhage (P = 0.020), and nonhemorrhagic venous infarcts (P = 0.043) [Table 4].

Table 4: Evaluation of the relationship between the type of parenchymal lesions and early seizure after cerebral venous thrombosis Click here to view

When the relationship between parenchymal lesion type and late seizures was examined, a statistically significant relationship was found with juxtacortical hemorrhage (P = 0.001) and parenchymal haematoma (P = 0.002). Nonsignificant results were found with subarachnoid hemorrhage and nonhaemorrhagic venous infarction (P > 0.05) [Table 5].

Table 5: Evaluation of the relationship between the type of parenchymal lesions and late seizure after cerebral venous thrombosis Click here to view

Discussion

In this study, a statistically significant relationship was found between epileptic seizures and superior sagittal sinus involvement, sigmoid sinus involvement, cortical vein involvement, the presence of parenchymal lesions, being in the postpartum period, formation of intracranial herniation, death of CVT, and poor clinical outcomes after CVT. According to binary logistic regression analysis, parenchymal lesions were the most significant risk factor for epileptic seizures after CVT.

The incidence of epileptic seizures in the 1^st year after the diagnosis of CVT was 29%. A wide range of rates (12%–44.3%) has been reported in different studies.^{[3],[13],[14],[15],[16]} CVT is a heterogeneous group of diseases caused by multiple etiologies, and the broad range of rates of epileptic seizures after CVT may be due to factors such as racial differences and variations in study methodologies. A more extreme rate of epileptic seizures after CVT has been reported in postpartum patients, which is consistent with our study.^[13],[14]

Epileptic seizures occur more frequently in patients with CVT than in patients with haemorrhagic and ischaemic strokes. Epileptic seizures can develop at different stages (early or late) in the course of the disease; however, the pathophysiology of these different onsets varies.^[5],[6] Vasogenic edema, changes in extracellular ions, loss of the blood-brain barrier, excessive release of excitatory neurotransmitters, changes in the way cells produce energy and glutamate, and free radicals cause early seizures.^[15],[16] Late seizures, by contrast, result from the development of gliosis and meningocerebral scarring.^[17] In addition, changes in membrane properties, deafferentation, selective neuronal loss, and collateral sprouting can result in hyperexcitability at levels high enough to cause epileptic seizures.^[18]

According to previous studies, the predictors of epileptic seizures in patients with CVT are a low Glasgow Coma Scale score, the presence of superior sagittal sinus involvement accompanied by cortical vein thrombosis, and the presence of hemorrhagic infarctions. Our study is thus similar to the literature.^{[19],[20],[21]}

In our study, the variables that had the greatest significant statistical relationship with epileptic seizures after CVT were the presence of parenchymal lesions and cortical vein thrombosis. The close relationship between parenchymal lesions and epileptic seizures after CVT has been shown in many studies.^{[4],[19],[21]} Superior sagittal sinus thrombosis and cortical vein thrombosis probably increase the risk of epileptic seizures by causing parenchymal lesions, and in cortical vein thrombosis, close contact of cortical veins to the cerebral cortex can cause local alterations of the blood-brain barrier and trigger seizures. Various parenchymal lesions with different pathophysiologies can be seen in patients with CVT.^[9]

The brain contains two venous systems: The superficial and deep cerebral veins. The superficial venous system consists of intracortical and subcortical veins. Connections between the superficial and deep systems may occur through anastomotic medullary and transcerebral veins. In patients with CVT, when rupture occurs in the subcortical veins due to pressure increase, a juxtacortical hemorrhage may occur; when rupture occurs in the intracortical veins due to a pressure increase, subarachnoid hemorrhage may occur.^{[10],[22],[23]} Infarctions that do not fit into the major arterial vascular region, such as multiple isolated lesions and the involvement of a subcortical region where the cortex is preserved and exceeds one arterial distribution, may raise suspicion of venous infarctions due to CVT.

The typical location affects venous structures, and the pathophysiologies of these lesions are different. Superior sagittal sinus thrombosis often leads to impaired venous drainage and thus parenchymal changes in the parasagittal region. Labbé vein thrombosis can lead to temporal lobe infarction. Bilateral or unilateral infarction of the thalamus, basal ganglia, and inner capsule is typically seen in internal cerebral vein thrombosis.^[8],[24]

Juxtacortical hemorrhages are small, concave hemorrhages located in the white matter adjacent to the cortex; as described recently, they tend to be more supratentorial. Previous studies indicated that juxtacortical hemorrhages were characteristic of CVT.^[10] Juxtacortical haemorrhages are associated with superior sagittal sinus involvement and cortical vein involvement, which are provocateurs of epileptic seizures after CVT.

Similar to a recently published study, our results showed that juxtacortical hemorrhage was more common than subarachnoid hemorrhage in patients with CVT.^[9] A probable reason for this is that the veins in the cortex are more resistant to rupture because the cortex itself has a more dense structure, which may provide resistance against the increased pressure inside the veins. In contrast, the white matter may have a looser structure, making veins in the white matter the weakest (most susceptible to rupture) link in the superficial venous system.^[10]

Previous studies did not consider juxtacortical hemorrhages a risk factor for epileptic seizures after CVT. The fact that juxtacortical hemorrhages cause seizures may be due to the deterioration of the blood-brain barrier, permanent damage to the parenchyma, and its proximity to the cortex.

Our study is the first to examine the relationship between parenchymal lesion type and epileptic seizures after CVT. A statistically significant relationship was found between seizures in the 1^st year after CVT and juxtacortical hemorrhage and nonhemorrhagic venous infarcts. No significant relationship was identified between subarachnoid hemorrhages and parenchymal hematomas. Early and late seizures were analyzed separately, and juxtacortical hemorrhage and nonhemorrhagic venous infarcts were found to be correlated with early and late seizures. By contrast, subarachnoid hemorrhage triggered only early seizures, and parenchymal hematomas triggered only late seizures.

Poor clinical outcomes are less common in patients with CVT than in patients with arterial thrombosis. Most patients with CVT have good clinical outcomes because of early diagnosis and improvements in treatment. The mortality rate has been decreasing over time.^[25] Some previous studies emphasized that patients with epileptic seizures had poor prognostic features,^{[15],[26],[27],[28],[29],[30]} but other studies found no association with poor clinical outcomes.^[20]

Epileptic seizures are deemed an independent risk factor for death and poor clinical outcomes in an intracranial sinus venous thrombosis study.^[7] In our study, intracranial herniation, which is also closely related to parenchymal lesions, death of CVT, and poor clinical outcome, was found to be associated with epileptic seizures after CVT.

Limitations of the study

The greatest limitation of our study is that it uses a single center and a limited sample size. Moreover, this study is retrospective; the patients were evaluated only according to their medical records in our tertiary center.

Conclusion

Our study found that parenchymal lesions were the most associated variable with epileptic seizures after CVT, and juxtacortical hemorrhages and nonhemorrhagic venous infarcts are the most common causes of epileptic seizures. Patients with seizures have poor clinical outcomes and increased mortality due to parenchymal lesions, and higher rates of intracranial herniation. CVT is a heterogeneous group of diseases caused by multiple aetiologies and may show ethnic and racial differences. For this reason, more precise information can be obtained with multi-center prospective studies in our population.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.



 

References

1.	Bousser MG, Crassard I. CVT, pregnancy and oral contraceptives. Thromb Res 2012;130:19-22.
2.	Stam J. Thrombosis of the cerebral veins and sinuses. N Engl J Med 2005;352:1791-8.
3.	Ferro JM, Correia M, Rosas MJ, Pinto AN, Neves G; Cerebral Venous Thrombosis Portuguese Collaborative Study Group. Seizures in cerebral vein and dural sinus thrombosis. Cerebrovasc Dis 2003;15:78-83.
4.	Ferro JM, Canhão P, Bousser MG, Stam J, Barinagarrementeria F; ISCVT Investigators. Early seizures in cerebral vein and dural sinus thrombosis: Risk factors and role of antiepileptics. Stroke 2008;39:1152-8.
5.	Masuhr F, Busch M, Amberger N, Ortwein H, Weih M, Neumann K, et al. Risk and predictors of early epileptic seizures in acute cerebral venous and sinus thrombosis. Eur J Neurol 2006;13:852-6.
6.	Koubeissi MZ, Alshekhlee A, editors. Seizures in Cerebrovascular Disorders: A Clinical Guide. New York: Springer; 2015. p. 95-101.
7.	Ferro JM, Canhão P, Bousser MG, Stam J, Barinagarrementeria F. Prognosis of cerebral vein and dural sinus thrombosis: Results of the international study on cerebral venous sinus thrombosis (ISCVT). Stroke 2004;35:664-70.
8.	Ameri A, Bousser MG. Cerebral venous thrombosis. Neurol Clin 1992;10:87-111.
9.	Afifi K, Bellanger G, Buyck PJ, Zuurbier SM, Esperon CG, Barboza MA, et al. Features of intracranial hemorrhage in cerebral venous thrombosis. J Neurol 2020;267:3292-8.
10.	Coutinho JM, Vandenberg R, Zuurbier SM, VanBavel E, Troost D, Majoie CB, et al. Small juxtacortical haemorrhages in cerebral venous thrombosis. Ann Neurol 2014;75:908-16.
11.	Fisher RS. The new classification of seizures by the international league against epilepsy 2017. Curr Neurol Neurosci Rep 2017;17:48.
12.	International League Against Epilepsy Commission on Classification and Terminology of the International League Against Epilepsy. Proposal for revised classification of epilepsies and epileptic syndromes. Epilepsia 1989;30:389-99.
13.	Bajko Z, Motataianu A, Stoian A, Barcutean L, Andone S, Maier S, et al. Postpartum cerebral venous thrombosis – A single-center experience. Brain Sci 2021;11:327.
14.	Ferro JM, Correia M, Pontes C, Baptista MV, Pita F; Cerebral Venous Thrombosis Portuguese Collaborative Study Group (Venoport). Cerebral vein and dural sinus thrombosis in Portugal: 1980-1998. Cerebrovasc Dis 2001;11:177-82.
15.	Luhmann HJ. Ischemia and lesion induced imbalances in cortical function. Prog Neurobiol 1996;48:131-66.
16.	Heiss WD, Huber M, Fink GR, Herholz K, Pietrzyk U, Wagner R, et al. Progressive derangement of periinfarct viable tissue in ischemic stroke. J Cereb Blood Flow Metab 1992;12:193-203.
17.	Jennett B. Posttraumatic epilepsy. Adv Neurol 1979;22:137-47.
18.	Stroemer RP, Kent TA, Hulsebosch CE. Neocortical neural sprouting, synaptogenesis, and behavioral recovery after neocortical infarction in rats. Stroke 1995;26:2135-44.
19.	Kalita J, Chandra S, Misra UK. Significance of seizure in cerebral venous sinus thrombosis. Seizure 2012;21:639-42.
20.	Mehvari Habibabadi J, Saadatnia M, Tabrizi N. Seizure in cerebral venous and sinus thrombosis. Epilepsia Open 2018;3:316-22.
21.	Uluduz D, Midi I, Duman T, Yayla V, Karahan AY, Afsar N, et al. Epileptic seizures in cerebral venous sinus thrombosis: Subgroup analysis of VENOST study. Seizure 2020;78:113-7.
22.	Okudera T, Huang YP, Fukusumi A, Nakamura Y, Hatazawa J, Uemura K. Micro-angiographical studies of the medullary venous system of the cerebral hemisphere. Neuropathology 1999;19:93-111.
23.	Kurokawa Y, Hashi K, Okuyama T, Uede T. Regional ischemia in cerebral venous hypertension due to embolic occlusion of the superior sagittal sinus in the rat. Surg Neurol 1990;34:390-5.
24.	Schaller B, Graf R. Cerebral venous infarction: The pathophysiological concept. Cerebrovasc Dis 2004;18:179-88.
25.	Luo Y, Tian X, Wang X. Diagnosis and treatment of cerebral venous thrombosis. A review. Front Aging Neurosci 2018;10:1-15.
26.	Shindo A, Wada H, Ishikawa H, Ito A, Asahi M, Ii Y, et al. Clinical features and underlying causes of cerebral venous thrombosis in Japanese patients. Int J Hematol 2014;99:437-40.
27.	Busch MA, Hoffmann O, Einhäupl KM, Masuhr F. Outcome of heparin-treated patients with acute cerebral venous sinus thrombosis: Influence of the temporal pattern of intracerebral haemorrhage. Eur J Neurol 2016;23:1387-92.
28.	Jalili M, Ghourchian S, Shahidi GA, Rohani M, Rezvani M, Zamani B. A study of factors associated with cerebral venous thrombosis. Neurol Sci 2013;34:321-6.
29.	Dinc Y, Ázpar R, Hakyemez B, Bakar M. The relationship between early neurological deterioration, poor clinical outcome, and venous collateral score in cerebral venous sinus thrombosis. Neurol Sci Neurophysiol 2021;38:158-65.     [Full text]
30.	Canhao P, Ferro JM, Lindgren AG, Bousser MG, Fernando B; ISCVT Investigors. Causes and predictors of death in cerebral venous thrombosis. Stroke 2005;36:1720-5.

Figures

  [Figure 1]
 
 

Tables

  [Table 1], [Table 2], [Table 3], [Table 4], [Table 5]