A 16-month-old boy presented as generally unwell, hypersomnolent and lethargic for 2-3 weeks and a short history of gait disturbance and photophobia with a possible antecedent mild non-specific viral infection. Standard MR brain imaging was performed. The patient was transferred to a regional centre for further assessment and treatment.
The patient initially underwent a standard paediatric MRI brain protocol (sagittal T1-w, axial spin echo T2-w, axial and coronal non-fat saturated fluid-attenuated inversion recovery (FLAIR), axial diffusion weighted imaging (DWI) and axial gradient echo T2-w (T2*) sequences) in the local hospital followed by a repeat study with IV gadolinium the following day. The initial study showed diffuse high T2-w signal within both thalami (Figure 1A), most prominent on the FLAIR sequences (Figure 1B).
The obtained T1 & T2 spin echo sequences demonstrated loss of normal flow voids within the straight sinus (Figure 2A) and the left transverse (Figure 2B) sinuses. There was corresponding FLAIR hyperintensity (Figure 2C) and low apparent diffusion coefficient (ADC) values on diffusion weighted sequences (Figure 2E, F). On review of the original MR study, many of these findings were also retrospectively apparent. Additionally prominent blooming artefact was seen with the same venous sinuses and adjacent draining veins on the T2* images (Figure 3A, B). Gadolinium contrast-enhanced images demonstrated clear and extensive filling defects involving the straight (Figure 2D, white bordered black arrowheads) and left transverse and sigmoid sinuses, extending into the vein of Galen (Figure 2D, white arrow) and both internal cerebral veins.
At the time of transfer to the regional centre the primary differential diagnosis was acute disseminated encephalomyelitis (ADEM) due to the thalamic abnormality. This is why the repeat imaging was tailored to assess for demyelination. However, although not optimised for venous sinus images, the repeat study clearly demonstrated the presence of extensive cerebral venous sinus thrombosis (CVST).
CVST is a rare but important entity seen throughout childhood. Its incidence varies with age with single-series studies suggesting a reported incidence around 2.6 per 100,000 in neonates  with a lower incidence of 0.4-0.7 per 100,000 in older children [1,2,3].
Although there are commonalities between adults and children in terms of the aetiology of CVST, there are several critical differences with further differences observed between older and younger children. Age-independent risk factors are vast but important examples include dehydration, anaemia, infection (local or systemic), clotting and autoimmune disorders, cardiac disease and malignancy [1,2]. In older children, head and neck infection becomes an important and more common risk factor .
As exemplified by our case, the presentation of paediatric CVST can be non-specific and may overlap with the presentation other predisposing factors such as severe dehydration or infection . Common presenting features include headaches, vomiting, altered mental status, seizures and focal neurological deficits [1,2,3]. The possibility of CVST may therefore be just one of many potential differential diagnoses in a child presenting with such non-specific symptoms. Such differential diagnoses include arterial stroke, vasculitides, intracranial infection or, as in this case, inflammatory disorders .
CT venography and MR venography (usually based on phase-contrast imaging techniques) present two easily accessible and highly sensitive methods for diagnosing CVST  but due to the non-specific nature of the presentation, and variable degrees of experience in interpretating such studies, these specific examinations may not always be performed.
However, as shown by our case, thrombus can be seen on many standard MR sequences [6,7]. Loss of T1-w and T2-w flow voids within the affected sinuses is an unsurprising consequence of luminal thrombus . The signal characteristics of the thrombus itself can be used to estimate its age appearing isointense to brain on T1-w and hypointense on T2-w images in the acute phase (due to the prevalence of deoxyhaemoglobin), T1-w and T2-w hyperintense in the more subacute phase (due to presence of methaemoglobin) [6,8]. In this case the thrombus demonstrated T1 hyperintensity and T2 isointensity suggesting that it was in an early subacute time course. Cerebral venous thrombus also demonstrates high signal on FLAIR sequences and, in a large minority of cases, reduced mean apparent diffusion coefficient (ADC) values [ 6,8]. Blooming artefacts are also commonly seen in gradient recall echo sequences due to blood breakdown products in the thrombus . This is a particularly useful sign with a high PPV for the detection of acute thrombus which may be more difficult to perceive based on other signal characteristics . Our case provides a good example of the utility of the gradient echo T2* sequence as the blooming artefact within the right transverse sinus and straight sinus is extremely prominent. Finally, thrombus can often be seen as filling defects on standard delayed post-contrast T1-w MR images even if not timed to the venographic phase . However, an important potential pitfall is that thrombus may show intense contrast enhancement similar to that of the patent venous sinus thereby not being visible . In our case, the presence of clear venous sinus filling defects on the post-contrast T1 sequences obviated the need for dedicated venographic sequences and therefore allowed for earlier confident diagnosis than may otherwise have occurred.
Other indirect MR findings that may suggest underlying CVST include ischaemic infarcts (which classically do not conform to an arterial territory), parenchymal haemorrhage, signs of raised intracranial pressure and prominent collateral and emissary veins [6,7,8]. Parenchymal T2 hyperintensity can also be seen on MR reflecting venous ischaemia or parenchymal oedema . The location of the parenchymal abnormality often reflects the location of the CVST [9,10], although symmetry is not guaranteed in thrombosis involving midline sinuses . Such signal abnormality may be an important secondary clue to an underlying CVST when non-dedicated studies have been performed. Of particular interest, the bilateral thalamic signal abnormality seen in our case is considered a hallmark feature of deep CVST and is reported to be present in as many as 86% of cases . Therefore, in retrospect, this could have been an early clue in the patient’s initial study. Although deep CVST is comparatively less common than superficial CVST , it is associated with worse clinical outcomes .
Our patient followed a standard management pathway with anticoagulation which is considered the mainstay of treatment [2,4]. Other options including venous sinus thrombectomy are typically reserved for severe cases due to the lack of an evidence base [4,12]. Identification and treatment of the patient’s iron deficiency anaemia was also an important tenet of management as this is a well-recognised risk factor for CVST. Although there was likely some element of dehydration which also played a role, no other predisposing factor could be found in this patient. In particular, there was no evidence of an underlying thrombophilia. Some studies suggest that 32-56% of patients may have underlying prothrombotic pathologies such as thrombophilia [1,2]. CVST-related mortality is not insubstantial, reported at approximately 10%  with variable rates of long-term neurological deficits reported in children [2,13]. Poor neurological outcomes are more likely to occur in those with parenchymal haemorrhage or infarct .
Take home messages:
Written informed patient consent for publication has been obtained.
 Heller, C., Heinecke, A., Junker, R., Knöfler, R., Kosch, A., Kurnik, K., Schobess, R., von Eckardstein, A., Sträter, R., Zieger, B. and Nowak-Göttl, U., 2003. Cerebral Venous Thrombosis in Children. Circulation, 108(11), pp.1362-1367.
 Dlamini, N., Billinghurst, L. and Kirkham, F., 2010. Cerebral Venous Sinus (Sinovenous) Thrombosis in Children. Neurosurgery Clinics of North America, 21(3), pp.511-527.
 deVeber, G., Andrew, M., Adams, C., Bjornson, B., Booth, F., Buckley, D., Camfield, C., David, M., Humphreys, P., Langevin, P., MacDonald, E., Meaney, B., Shevell, M., Sinclair, D., Yager, J. and Gillett, J., 2001. Cerebral Sinovenous Thrombosis in Children. New England Journal of Medicine, 345(6), pp.417-423.
 Ichord, R., 2017. Cerebral Sinovenous Thrombosis. Frontiers in Pediatrics, 5.
 Capecchi, M., Abbattista, M. and Martinelli, I., 2018. Cerebral venous sinus thrombosis. Journal of Thrombosis and Haemostasis, 16(10), pp.1918-1931.
 6 Wasay, M. and Azeemuddin, M., 2005. Neuroimaging of Cerebral Venous Thrombosis. Journal of Neuroimaging, 15(2), pp.118-128.
 Bonneville, F., 2014. Imaging of cerebral venous thrombosis. Diagnostic and Interventional Imaging, 95(12), pp.1145-1150.
 Leach, J., Fortuna, R., Jones, B. and Gaskill-Shipley, M., 2006. Imaging of Cerebral Venous Thrombosis: Current Techniques, Spectrum of Findings, and Diagnostic Pitfalls. RadioGraphics, 26(suppl_1), pp.S19-S41.
 Teksam, M., Moharir, M., deVeber, G. and Shroff, M., 2008. Frequency and Topographic Distribution of Brain Lesions in Pediatric Cerebral Venous Thrombosis. American Journal of Neuroradiology, 29(10), pp.1961-1965.
 Tsai, F., Wang, A., Matovich, V., Lavin, M., Berberian, B., Simonson, T. and Yuh, W., 1995. MR staging of acute dural sinus thrombosis: correlation with venous pressure measurements and implications for treatment and prognosis. AJNR Am J Neuroradiol., 16(5), pp.1021-1029.
 Bergui, M., Bradac, G. and Daniele, D., 1999. Brain lesions due to cerebral venous thrombosis do not correlate with sinus involvement. Neuroradiology, 41(6), pp.419-424.
 Yeo, L., Lye, P., Yee, K., Cunli, Y., Ming, T., Ho, A., Sharma, V., Chan, B., Tan, B. and Gopinathan, A., 2020. Deep Cerebral Venous Thrombosis Treatment. Clinical Neuroradiology, 30(4), pp.661-670.
 deVeber, G., MacGregor, D., Curtis, R. and Mayank, S., 2000. Neurologic Outcome in Survivors of Childhood Arterial Ischemic Stroke and Sinovenous Thrombosis. Journal of Child Neurology, 15(5), pp.316-324.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.