High-resolution susceptibility-weighted imaging of clots in cerebral venous thrombosis

Detection of cerebral cortical vein thrombosis with high-resolution susceptibility weighted imaging — A comparison with MR venography and standard MR sequences

“…The majority of these studies, mainly evaluated the susceptibility contrast of COVT with T2* GRE from a small cohort of cases [ 3 – 7 , 9 , 11 , 16 ]. Recently, Boukerche et al were able to demonstrate the success of SWI to distinguish cortical vein clots from deoxyhemoglobin of patent veins, although the sequence is designed deliberately to make the patent cortical veins appear hypointense from the physiological deoxyhemoglobin [ 14 ]. In our study, we used a relatively large cohort of COVT cases to compare SWI against other MR sequences and the results revealed the best sensitivity of 93%, NPV of 96% and accuracy of 97% for SWI to identify cortical vein clots (Table 2 ).…”

“…Each MR sequence including T1 weighted turbo spin echo (TSE), T2 weighted TSE, FLAIR, diffusion weighted imaging (DWI), enhanced T1 TSE with fat saturation (T1C TSE), enhanced 3D T1 MPRAGE (3D T1c MPRAGE), SWI and MR venography (2D Time of flight (TOF) or 3D phase contrast (PC)) was evaluated independently on PACS on separate occasions over a month. Prominent hypointense signals with engorgement of a cortical vein on SWI was considered a sign of COVT [ 14 ]. Signs of cortical vein clot for each of the other MR sequences were considered as hyperintense signals or bulky isointense signals on T1 TSE, iso or hyperintense signals in the lumen of a cortical vein on T2 TSE, iso or hyperintense signals in cortical veins on FLAIR, prominent cord like susceptibility or hyperintense signals in the location of cortical veins on DWI, filling defect in the lumen of a cortical vein on 3D T1c MPRAGE or T1C TSE and hypointense signals in a cortical vein in the raw images of MR venography (MRV).…”