Indian Journal of Radiology Indian Journal of Radiology  

   Login   | Users online: 340

Home Bookmark this page Print this page Email this page Small font sizeDefault font size Increase font size     


NEURORADIOLOGY Table of Contents   
Year : 2004  |  Volume : 14  |  Issue : 2  |  Page : 129-137
Cerebral venous thrombosis-spectrum of CT findings

Dept. of Radiodiagnosis, K.G. Hospital and Postgraduate Medical Institute, 5, Govt. Arts College Road, Coimbatore-641018, Tamil Nadu, India

Click here for correspondence address and email

Keywords: Cerebral Venous Thrombosis, CT

How to cite this article:
Karthikeyan D, Vijay S, Kumar T, Kanth L. Cerebral venous thrombosis-spectrum of CT findings. Indian J Radiol Imaging 2004;14:129-37

How to cite this URL:
Karthikeyan D, Vijay S, Kumar T, Kanth L. Cerebral venous thrombosis-spectrum of CT findings. Indian J Radiol Imaging [serial online] 2004 [cited 2014 Apr 16];14:129-37. Available from:

   Introduction Top

Cerebral venous thrombosis (CVT) is an uncommon condition, which over the past 5 to 10 years has been diagnosed more frequently due to greater awareness and the availability of better non-invasive diagnostic techniques [1]. CVT is an elusive diagnosis because of its nonspecific presentation and its numerous predisposing causes. Imaging plays a key role in the diagnosis [2]. The imaging findings are easily overlooked if not specifically sought. The radiologist may be the first physician to suggest the diagnosis, which is a radiological (and not merely a clinical) one. The radiologist often plays an important role in establishing the diagnosis of CVT [3]. Though angiography is still considered to be the 'gold standard', MRI and MR venography are currently the preferred techniques for making the diagnosis of CVT. Before the advent of MRI, conventional CT was the best noninvasive method of diagnosing CVT [1]. In the present day, CT is the initial modality of choice in the course of investigating most neurological conditions. CT is likely to remain so because of its widespread availability, comparatively shorter scan times and lower cost. We present a spectrum of the CT features of CVT.

   Clinical features Top

The first description of CVT appeared in the French literature in 1825 as a postmortem report by Ribes. Though many series have been published since then, the true incidence of CVT is still unknown. Recent publication of large clinical series suggests that the true incidence of CVT is higher than previously thought [4].

The symptoms and signs associated with CVT are relatively nonspecific [5]. They are often vague and can include symptoms due to increased intracranial pressure [3]. These include headache, papilloedema, vomiting, seizures, and focal neurological deficits. CVT may be difficult to diagnose clinically because of its various and nonspecific manifestations and the multiple associated conditions and etiologies [5]. The thrombosis can also extend in a retrograde manner into cerebral veins and cause venous infarction; in this instance, focal neurologic signs can be seen. Symptoms are easily mistaken for those of tension headache, migraine headache, or other disease processes, such as pseudotumor cerebri or an intracranial mass [3]. The clinical spectrum is wide and recognition remains a challenge for the clinician.

Headache is the presenting symptom in 70-90% of cases. Focal deficits such as hemiparesis and hemisensory disturbance, seizures, impairment of level of consciousness and papilloedema occur in one-third to three-quarters of cases. The onset may be acute, subacute or insidious, most patients presenting with symptoms that have evolved over days or weeks. There are several typical clinical constellations: 18-38% of cases present with a syndrome resembling benign intracranial hypertension with headache, papilloedema and visual disturbances; up to 75% of cases are characterised by a focal neurological deficit and headache; a third group of between 30% and 50% may present with seizures often followed by a Todd's paresis. Rare but classical clinical pictures are that of superior sagittal sinus thrombosis (4%) with bilateral or alternating deficits and/or seizures and cavernous sinus thrombosis (3%) with chemosis, proptosis and painful ophthalmoplegia. An even less frequent presentation is a rapidly progressive illness with deepening coma, headache, nausea and pyramidal signs, due to extensive involvement of the deep cerebral veins. Thunderclap headache with neck stiffness mimicking subarachnoid haemorrhage has been described [5].

Acute focal presentations of CVT may clinically resemble stroke due to arterial infarction. Chronic presentations of CVT may raise concern for possible intracranial neoplasm. The clinical differential diagnosis for CVT also includes the various causes of acute and chronic intracranial hemorrhage (including subdural haematoma, hypertensive hemorrhage, vascular malformations, ruptured aneurysms, and other less common causes) [1].

In the early stages there may be cortical vein thrombosis without sinus thrombosis, the latter developing only later due to progression of the thrombotic process. There is no well-defined clinical syndrome to suggest this, although the rapid onset of focal deficit and/or seizures is thought to be typical of this situation. There is strong overlap between all these outlined groups and patients may progress from one to the other in the course of their illness [5].

CVT can occur on a spontaneous basis or be secondary to a hypercoagulable state, dehydration, infection, or dural sinus compression [3]. Multiple pathophysiologic mechanisms and predisposing factors exist, including: a hypercoagulable state, extrinsic compression or local invasion of a vein by tumor or an adjacent infectious process (e.g., mastoiditis), a low-flow state within the venous sinus, dehydration or pregnancy and the postpartum state [2] [Table 1] [6]. As many as 25% of patients present with no predisposing risk factor, however, in some patients, an etiologic factor is discovered subsequently [2],[7]. Evidence suggests that oral contraceptive use, pregnancy, and the peripartum state increase the risk of CVT [3]. Hypercoagulable states associated with puerperium as well as infectious diseases are believed to be the major causes in the third world, but these are less significant in western countries [4]. Elderly or debilitated patients (e.g., those with an underlying illness) are more likely to have spontaneous CVT. Neonates and infants suffering from dehydration may develop CVT [2].

A mortality range of 10-80% has been reported although the higher rate is based on older data [2]. Recent studies estimate a morbidity range of 6-20% [2],[7], including residual focal neurologic deficits and blindness secondary to optic nerve atrophy. The prognosis for return of function is believed to be somewhat better than for arterial stroke [2].

Because of the generally good prognosis and variable clinical signs, many cases remain clinically undetected [5].

Venoocclusive disease of the brain most commonly affects the superior sagittal sinus, followed by the transverse sinus, sigmoid and straight sinuses [7]. The main sinuses affected by CVT are the superior sagittal sinus (72%) and the lateral sinuses (70%). In about one-third of cases more than one sinus is affected. In a further 30-40% both sinuses and cerebral or cerebellar veins are involved [5]. CVT often presents with haemorrhagic infarction [Figure - 1] in areas atypical for arterial vascular distribution [2]. Haemorrhagic infarction occurs in approximately 10-50% of cases, principally affecting the cortex and adjacent white matter. This is thought to be primarily due to elevated venous and capillary pressure caused by the persistence of thrombosis [5].

   Diagnosis Top

CT, MRI, conventional angiography, and, more recently, CT venography have been used to detect CVT [7]. The main indication for assessment of venous structures is the clinical setting of suspected venous thrombo-occlusive disease [8]. Investigations should focus on establishing the diagnosis and searching for underlying causes. MRI combined with MR venography (MRV) has largely replaced invasive cerebral angiography and conventional CT. CT will, however, often remain the first imaging modality to be used simply due to availability and also to exclude other conditions such as intracerebral haemorrhage or abscess. CT is entirely normal in 10-20% of cases with proven CVT [5]. MRV in conjunction with conventional MRI can accurately diagnose CVT and is reliable as the sole examination for this condition [2],[5]. MRV is currently considered to be the noninvasive test of choice for evaluation of the dural sinus. However, flow-related and susceptibility artifacts can impair the evaluation of the venous structures. The more invasive arterial DSA is still the standard of reference [8]. Recently, CT venography has been shown to have sensitivity equal or superior to MRV in visualizing thrombosed sinuses or cerebral veins [5],[7]. This technique is not used routinely at present.

Before the advent of MRI, conventional CT was the best noninvasive method of diagnosing CVT [5]. The diagnosis may be made or suggested by CT before and after intravenous contrast medium injection. With careful interpretation of the findings, which may be subtle and a high degree of clinical suspicion, CT also may lead to the diagnosis [2]. On CT, infarctions in a non-arterial distribution in the white matter and/or cortical white matter junction, often associated with hemorrhage [Figure - 2], should suggest the possible diagnosis of CVT. Bilateral cerebral involvement can occur, including the superior cerebral white matter of the convexities from superior sagittal sinus thrombosis [Figure - 3], or the basal ganglia and thalami from internal cerebral vein thrombosis in which the internal cerebral veins appear hyperdense on NECT [2]. In most instances, the NECT examination in patients with suspected CVT serves mainly to depict secondary changes in the brain parenchyma, such as venous infarcts or edema. It also excludes other abnormalities in the initial workup [8]. On non-contrast-enhanced CT (NECT), the classic finding is the delta sign [Figure - 4], which is observed as a dense triangle (from hyperdense thrombus) within the superior sagittal sinus [2]. While the presence of haemorrhagic cortical venous infarcts [Figure - 2] or [Figure - 3], the cord sign [Figure - 5], the dense vein sign [Figure - 6],[Figure - 7], or subcortical hemorrhage [Figure - 8] may be helpful at NECT, intravenous contrast material is necessary for definitive CT diagnosis of CVT [1]. On contrast-enhanced CT (CECT), the empty delta sign, or reverse delta sign, or reverse triangle sign, [Figure - 9] can be observed in the superior sagittal sinus reflecting the opacification of collateral veins in the dural leaves surrounding the comparatively less dense thrombosed sinus [2],[5]. The empty delta sign can be diagnostic for sinus thrombosis, but it has been reported in only about 28%-72% of cases with conventional CT [1]. False-positive causes of the empty delta sign include subdural haematoma, subdural empyema and arachnoid granulations [7]. A high splitting sagittal sinus can also mimic the appearance of the empty delta sign [8]. In superior sagittal sinus thrombosis, outward bowing of the superior sagittal sinus wall is an abnormal finding [7]. False-negative empty delta signs result from partial volume averaging effects, a small thrombus, or recanalised thrombus [7]. Indirect CT signs include focal cerebral cortical ischemia with gyral enhancement, small ventricles compressed by cerebral edema, and intense tentorial enhancement [Figure - 10]. Occasionally, the transcerebral medullary cortical veins and pial collaterals [Figure - 11] can be observed [2].

Conventional thin-section CECT improves visualization of dural sinus disease; however, the number of sections obtained during the peak of the contrast agent bolus is limited by scanning speed, and so high-resolution (non-helical) sections cannot encompass the entire brain [1]. Helical CT technology has produced a new method for evaluating CVT with CT venography [1]. The term CT venography was used by Casey et al [9], who described the technique as a rapid method to depict the intracranial venous circulation with consistently high quality [8]. Subsecond spiral CT revolution times and multi-detector-row CT (MDCT) scanners have allowed whole-brain CT venograms to be obtained in as little as 45 seconds. Unlike older conventional CT scanners, helical and MDCT scanners have sufficient speed for obtaining high-resolution images of the entire brain and all the dural sinuses during peak venous enhancement. The new technique of CT venography provides several advantages. At most institutions, NECT is performed during the initial workup of the patient presenting with acute disease. This can be immediately followed by CT venography in the appropriate patients undergoing the initial workup, at the same sitting, thus decreasing time to diagnosis and implementation of therapy. A rapid and reliable diagnosis of venous thrombo-occlusive disease is important, because early therapy is favorable in patients with CVT. Since the scan duration is very short, the image quality is hardly impaired by patient motion, and patient monitoring is easier in critically ill patients as compared with MRV. In situations in which MRI has absolute or relative contraindications (patients on non-MR-compatible life-support systems, claustrophobic patients, or patients with ferromagnetic implants or cardiac pacemakers), CT venography can still be performed [1],[9]. MRV, however, has the advantage of not requiring ionizing radiation or intravascular contrast material. Thus it is expected to remain the modality of choice in the evaluation of the pregnant patient with suspected CVT. Wetzl et al [8] claim that the effective radiation dose for CT venography was lower than that incurred with routine contiguous NECT. In preliminary reports, CT venography has been shown to be a useful method for evaluating the venous structures [8].

CT venography is an excellent alternative to MRI for diagnosis of CVT. CT venography is similar to CT angiography in terms of the helical data acquisition but differs by using a delay after contrast material administration, thus allowing optimal visualization of the venous system [3]. Ninety-hundred ml of non-ionic contrast is injected intravenously at 3 ml/sec through a pressure injection with a 40-sec pre scan delay [8],[9],[10]. Bone detail is subtracted from the resultant images with a thresholding technique; thereafter, data can be analyzed from source images or reconstructed with a variety of methods, such as maximum intensity projection [9]. The appearance of dural sinus thrombosis at CT venography is similar to that at MRV (i.e., a filling defect within the thrombosed dural sinus, alone or in association with a venous infarct). CT venograms are easier to interpret and have fewer artifacts than MR venograms [9].

Pathologic findings of a venous sinus can be displayed either with reformatted images [Figure - 12] or MIP reconstructions in multiple projections and can be directly related to parenchymal abnormalities [8]. CT venography has a high sensitivity for depicting the intracerebral venous circulation. Wetzl et al [8], claim that time consuming MIP images provided no added information when compared with the simple MPR images. All large sinus are clearly depicted on MPR images. Casey et al [1] claim that detection and delineation of intraluminal venous filling defects on the 3D CT venograms is best seen with integral display [Figure - 13] and that integral display provided a clearer depiction of intraluminal thrombus margins than MIP displays. They also claim that CT venography with integral display may allow detection of the uncommon but underdiagnosed condition of isolated cortical vein thrombosis, for which there is presently no ideal noninvasive test [1]. The integral algorithm displays the average intensity of a five-voxel-deep layer along the proximal surface, thus essentially providing the viewer with a five-voxel-thick curved reformation over a predetermined 3-D surface [11].

Alberico et al [12], while evaluating the feasibility of CT cerebrovascular imaging in pediatric patients using a hand-injected bolus of contrast material and a helical scanning technique, found that integral display algorithms showed venous structures better. In our experience of more than a hundred patients with CVT in the past 3 years (unpublished data, presented in part at the 56th IRIA 2003 at Jaipur) we have been able to obtain CT venographic images of adequate diagnostic quality with a hand-injected bolus of 60 ml of iodinated contrast medium and a helical scanning technique with a pre-scan delay of about 20-30 seconds. We routinely use multiplanar reformations and integral display algorithms for depicting the dural sinuses and cortical veins. In addition to the excellent depiction of dural sinus thrombosis, integral display also shows any nearby subcortical haemorrhagic infarcts [Figure - 14].

The improved detail of CT venograms, especially in the 3D reconstructions, should decrease the frequency of false-negative studies. The use of 3D CT venography models should help decrease the possibility that normal anatomic variation such as high splitting of the superior sagittal sinus could be mistaken for an empty delta sign [1]. CT venography is completed during the first minute of contrast material injection, there is insufficient time for strong nonvascular enhancement. It is unlikely that chronic thrombus, which is believed to convert to vascularized connective tissue, would have time to strongly enhance with iodine during CT venography.

   Conclusion Top

The clinical manifestations of CVT are often vague and are easily mistaken for those caused by other neurological disease processes. A variety of different imaging techniques are available for the noninvasive evaluation of the cerebral venous vasculature [8]. NECT is the most common initial imaging study for many of the clinical diagnostic considerations that CVT mimics. Unfortunately, NECT may occasionally show only subtle findings or may even appear normal. Cerebral hemorrhage or focal edema due to venous congestion or infarction are often findings at CT that lead to further imaging evaluations. Such findings, especially in young patients and when not readily explained by history, should raise concern for possible CVT. Subcortical hemorrhage, while nonspecific, has been reported as a common finding of CVT at CT and MRI [1]. The characteristic CT appearances and signs strongly suggest CVT but CT is seldom conclusively diagnostic [2]. Because of the subtlety of the findings, the prospective diagnosis of venous thrombosis may not be made unless a high index of suspicion is maintained during interpretation of the CT study [2]. CT venography is an alternative technique that enables excellent vascular delineation. In addition to being the study of choice when MRV is not possible, it may be the only test required when clarifying a question of thrombus raised on NECT [13].

   References Top

1.Casey SO, Ozsvath RR, Alberico RA, Rubinstein D. CT Venography of Dural Sinus Thrombosis. RSNA eJournal (online). 1998 March. [9 screens]. Available from: URL:  Back to cited text no. 1    
2.Patel MR. Brain, Venous Sinus Thrombosis. Article online. June 2003. Available from: URL:  Back to cited text no. 2    
3.Provenzale JM. Nontraumatic Neurologic Emergencies: Imaging Findings and Diagnostic Pitfalls. Radiographics. 1999; 19:1323-1331.  Back to cited text no. 3    
4.Daif A, Awada A, Al-Rajeh S, Abduljabbar M, Al Tahan AR, Obeid T, Malibary T. Cerebral Venous Thrombosis in Adults. A Study of 40 Cases From Saudi Arabia. Stroke. 1995;26:1193-1195.  Back to cited text no. 4    
5.Allroggen H, Abbott RJ. Cerebral venous sinus thrombosis. Postgrad Med J 2000;76:12-15.   Back to cited text no. 5  [PUBMED]  [FULLTEXT]
6.Karthikeyan D. Cerebral Venous Thrombosis. In: Karthikeyan D. Computed Aided Tomographic Case Histories - Brain. New Delhi: Jaypee Brothers. 2003; 96-99.   Back to cited text no. 6    
7.Greiner FG, Takhtani D. Neuroradiology Case of the Day. Radiographics 1999; 19:1098-1101.  Back to cited text no. 7  [PUBMED]  [FULLTEXT]
8.Wetzel SG, Kirsch E, Stock KW, Kolbe M, Kaim A, Radue EW. Cerebral Veins: Comparative Study of CT Venography with Intraarterial Digital Subtraction Angiography. Am J Neuroradiol 20:249-255.   Back to cited text no. 8    
9.Casey SO, Alberico RA, Patel M, Jimenez JM, Ozsvath RR, Maguire WM, Taylor ML. Cerebral CT venography. Radiology 1996;198:163-170.  Back to cited text no. 9  [PUBMED]  
10.Shetty PG, Jhaveri KS. Neurovascular Applications of CT Angiography. Ind J Radiol Imag 2000;10:211-220.  Back to cited text no. 10    
11.Casey SO, Rubinstein D, Lillehei KO, Alberico RA, Ozsvath RR, Cajade-Law AG, Weprin BE, Michel E, Truwit CL. Integral and Shell-MIP Display Algorithms in MR and CT Three-dimensional Models of the Brain Surface. Am J Neuroradiol 1998;19:1513-1521.   Back to cited text no. 11  [PUBMED]  [FULLTEXT]
12.Alberico RA, Barnes P, Robertson RL, Burrows PE. Helical CT Angiography: Dynamic Cerebrovascular Imaging in Children. Am J Neuroradiol 1999;20:328-334.  Back to cited text no. 12  [PUBMED]  [FULLTEXT]
13.Gotwald TF, Beauchamp NJ.CTA and MRA-Evaluation of the intracranial vasculature. Appl Radiol 2000 Aug:13-18.  Back to cited text no. 13    

Correspondence Address:
S Vijay
Dept. of Radiodiagnosis, K.G. Hospital and Postgraduate Medical Institute, 5, Govt. Arts College Road, Coimbatore-641018, Tamil Nadu
Login to access the Email id

Get Permissions


[Figure - 1], [Figure - 2], [Figure - 3], [Figure - 4], [Figure - 5], [Figure - 6], [Figure - 7], [Figure - 8], [Figure - 9], [Figure - 10], [Figure - 11], [Figure - 12], [Figure - 13], [Figure - 14]

This article has been cited by
1 Cerebral venous sinus thrombosis: Comparison of multidetector computed tomography venogram (MDCTV) and magnetic resonance venography (MRV) of various fi eld strengths
Ho, J.S. and Rahmat, K. and Ramli, N. and Fadzli, F. and Chong, H.T. and Tan, C.T.
Neurology Asia. 2012; 17(4): 281-291
2 Overlooked early CT signs of cerebral venous thrombosis with lethal outcome
Kozic, D., Zarkov, M., Semnic, R.R., Ostojic, J., Radovanovic, B., Ivanovski, D.
Acta Neurologica Belgica. 2010; 110(4): 345-348
3 Mastoiditis, subdural empyema and cortical venous thrombosis
Agrawal, A.
Journal of the Pakistan Medical Association. 2008; 58(3): 151-152
4 Computed Tomographic Angiography in Stroke Imaging: Fundamental Principles, Pathologic Findings, and Common Pitfalls
Gupta, R., Jones, S.E., Mooyaart, E.A.Q., Pomerantz, S.R.
Seminars in Ultrasound, CT and MRI. 2006; 27(3): 221-242


    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Email Alert *
    Add to My List *
* Registration required (free)  

    Clinical features
    Article Figures

 Article Access Statistics
    PDF Downloaded0    
    Comments [Add]    
    Cited by others 4    

Recommend this journal