ESWAN In The Detection Of Diffuse Axonal Injury Related Hemorrhagic Lesions And The Investigation Of Imaging Parameters

Purpose:To compare the effectiveness of enhanced T2 star weighted angiography(ESWAN) sequence with that of a conventional T2*-weighted gradient-recalled-echo(GRET2*WI) MR imaging technique for detection of hemorrhagic shearing lesions associated with diffuse axonal injury(DAI) and to determine whether hemorrhagic lesions correlated with clinical variables such as initial Glasgow Coma Scale(GCS) score,duration of coma and reaction of pupil to light.Methods:Twenty-eight DAI patients were included in our study.The global and regional values of lesion number determined with both examinations were compared by using Wilcoxon Signed Ranks Test(two-tailed) using SPSS Statistic for Windows(Release 13.0).Differences were considered to be significant at P≤0.05. Global and regional number,volume,and distribution of hemorrhagic lesions were compared with dischotomized clinical variables by nonparametric Mann-Whitney U test (two-tailed) and independent samples t test. Differences were considered to be significant at P≤0.05.Spearman's rank correlation analysis also was performed to compare the dischotomized clinical variables with the global and regional number and volume of lesions.Correlations were considered significant at P≤0.05.Results:Hemorrhagic lesions were much more visible on ESWAN MR images than on conventional GRET2*WI MR images.ESWAN MR imaging depicted a significant higher global (P=0) and regional(P=0.001 and P=0.007) number of lesions than did GRE MR imaging.Patients with lower GCS score(≤8,n=16) or prolonged coma(>4 days,n=15) abnormal reaction of pupil to light(n=15) had a greater global number(P≤0.01) and apparent volume(P≤0.01) of hemorrhagic lesions.Significant differences(P≤0.05) also were observed when comparing regional injury (such as frontal white matter, parietotemporaloccipital white matter, corpus collasum, thalamus and brain stem)to clinical variables(with an exception that there was no statistical difference in lesion volume of frontal white matter between the two groups by coma duration).Significant correlations were seen between the dischotomized clinical variables and global number(P≤0.01) and global volume(P≤0.01) of lesions.Regional analysis showed significant correlations(P≤0.05) between the dischotomized clinical variables and regional lesion extent in frontal white matter, parietotemporaloccipital white matter, corpus collasum, thalamus and brain stem(with two exceptions that there were no statistical correlations between the lesion volume of frontal white matter and coma duration as well as reaction of pupil to light).Conclusion:ESWAN is much more sensitive than conventional GRET2*WI in detecting hemorrhagic diffuse axonal injury,more accurate and objective assessment of injury can be obtained after injury. Purpose:The goal of this study was to identify and describe the signal changes on magnetic resonance imaging(MRI),including diffusion-weighted imaging(DWI),T2-weighted fluid-attenuated inversion recovery (T2/FLAIR) and enhanced gradient echo T2 star weighted angiography(ESWAN) in diffuse axonal injury(DAI) in the brain.,and to estimate the diagnostic value of every sequence in order to design more reasonable MRI program with respect to DAI patients.Methods:Ten DAI patients(five male,five female) within two weeks after trauma were analyzed with MRI(T2/FLAIR, ESWAN,and DWI).DWI in three orthogonal directions with calculation of apparent diffusion coefficient(ADC) maps was also performed.Lesions were identified and compared on all sequences.Results:Four hundred seventy-two lesions were counted by the combined use of all sequences.DWI identified 307 lesions,in which 19 lesions were not visible on T2/FLAIR or ESWAN sequences; FLAIR identified 331 lesions,in which 15 lesions were not visible on DWI or ESWAN sequences;ESWAN identified 408 hemorrhagic lesions,in which 103 lesions were not visible on DWI or T2/FLAIR sequences;The majority of DWI-positive non-hemorrhagic shearing lesions showed decreased diffusion(80.8%).Conclusion:All of the three sequences(DWI,T2/FLAIR and ESWAN) are valuable in DAI because each of them identified additional shearing injuries not visible on either of the two other sequences.ESWAN is extremely sensitive for detecting hemorrhagic shearing lesions,and its application is not restricted by the time after trauma;DWI is valuable in detecting non-hemorrhagic shearing lesions especially in acute and subacute phase after trauma.Furthermore,DWI/ADC maps differentiate between lesions with decreased or increased diffusion.On account of the short acquisition time,DWI is more convenient and appropriate for unstable and sever DAI patients in acute phase.T2/FLAIR is more worthful than DWI in indentifying non-hemorrhagic shearing lesions after acute phase.With respect to DAI patients in subacute or chronic phase after trauma,the combined use of ESWAN and T2/FLAIR is recommended. Objective:Use of magnetic susceptibility effect as an image contrast is a new technique that has proven useful for evaluation iron containing tissues, cerebral microvasculature, and pathologic processes resulting in susceptibility changes (hemorrhage, metastasis, traumatic brain injury, metabolic disorders). ESWAN(enhanced gradient echo T2 star weighted angiography) is a relatively new susceptibility enhanced imaging sequence which acquires multiple images at different echo times corresponding to different T2* weighting and employs flow compensated gradients to minimize spatial misregistration due to moving tissues.Cerebral veins show a profoundly hypointense signal on these long TE images while the intracranial arteries are hyperintense on the short TE images.The purpose of these study was to assess the potential of ESWAN for the depiction of both cerebral veins and arteries,and compare this technique to conventional time-of-flight angiography.Methods:13 healthy volunteers were includeded in the study.MRI was performed on a 1.5Tesla MR scanner(HD propeller,General Electrics,USA) using the following parameters (1) a 3D multi-echo enhanced gradient echo T2 Star Weighted Angiography (ESWAN) with 11 echos (TR= 77ms,TE of the first echo=10ms,the echo sounding=5.1-5.2ms, slice thickness= 2 mm,416 x 356 matrix, FOV= 240 mm x 192 mm, Flip angle=30°,bandwidth=+/-62.5 kHz, scan time 7:31 min), (2) an arterial 3D time of flight MR angiography (TOF-MRA;TR= 25 ms, TE= 3 ms, flip angle= 20°, bandwidth=+/-20.83 kHz, slice thickness= 1.2 mm matrix= 320 x192, FOV= 200 mm x 176 mm, scan time= 8:45min) (3) a venous 2D time of flight (TOF-MRV; TR= 24 ms, TE= 4.9 ms, flip angle= 60°, bandwidth=+/-15.63 kHz, matrix= 256 x 160, FOV-200 mm x 200 mm, scan time= 8:18 min). With regard to the ESWAN sequence,a new set of magnitude images corresponding to the first three echos were reconstructed by utilizing a simple sum-of-square weighted averaging, another new set of magnitude images corresponding to the last seven echos were reconstructed in the same way. The magnitude images corresponding to short echos were reformatted to depict cerebral arteries by use of maxium intensity projection(MIP) while the magnitude images corresponding to long echos were reformatted to visualize veins by use of minium intensity projection(mIP). Regarding the TOF-MRA and the TOF-MRV, MIP reformations were performed.ESWAN-MIP were evaluated according to TOF-MRA,in collaboration to score the success with which the two different sequences depicted the segments of intracerebral arteries as continuous (score-2), noncontinuous(score=1) and invisible(score=0).According to each segment.comparison of scores was performed by use of paired wilcoxon test. Eswan-mIP were compared to TOF-MRV,the number of revealed deep cerebral veins were counted on the slice of foramen level of the two sequences separately,we compared the two groups of counted number of veins by use of paired wilcoxon test.Results:The results in visualization of Willis circle constituent segments and the segments of posterial cerebral artery(PCA) as well as the A1,A2,A3 of anterial cerebral artery(ACA),M1,M2 of middle cerebral artery(MCA) showed the excellent agreements(P>0.05) between ESWAN-MIP and TOF-MRA. Concerning the depiction of the distal segments of ACA and MCA,ESWAN-MIP was inferior to TOF-MRA(P<0.01). With regard to the visualization of the deep cerebral veins,ES WAN-mIP were considerably superior to the TOF-MRV(P<0.05).Conclusion:The simultaneous acquisition of MRA and MRV can be performed by use of ESWAN sequence with appropriate parameters.The imaging effect of ESWAN-MIP requires improvements. Objective:The purpose of our study was to choose reasonable parameters for the detection of micro hemorrhagic lesions of diffuse axonal injury(DAI) through investigating the impact of parameters of ESWAN,such as TE and TR,on the magnetic susceptibility and causal analysis.Methods:10 DAI patients were included in the study. MRI was performed on a 1.5Tesla MR scanner(HD propeller,General Electrics,USA) using the following parameters(1)short TE ESWAN sequence:a 3D multi-echo enhanced gradient echo T2 Star Weighted Angiography (ESWAN) with 11 echos (TR= 77ms,TE of the first echo=10ms,the echo sounding=5.1-5.2ms, slice thickness= 2 mm, slice number=48,416 x 356 matrix, FOV= 240 mm x 192 mm, Flip angle=30°,bandwidth=+/-62.5 kHz, scan time=5min59s(2) long TE ESWAN sequence:a 3D multi-echo enhanced gradient echo T2 Star Weighted Angiography (ESWAN) with 11 echos (TR= 104ms,TE of the first echo=48ms.the echo sounding=5.1-5.2ms, slice thickness= 2 mm, slice number=48.416 x 356 matrix. FOV= 240 mm x 192 mm, Flip angle=30°,bandwidth=+/-62.5 kHz, scan time=8min17s.With regard to the short ESWAN sequence,a new set of magnitude images corresponding to the last six echos were reconstructed by utilizing a simple sum-of-square weighted averaging, and were reformatted by use of minium intensity projection(mIP). With regard to the long ESWAN sequence,a new set of magnitude images corresponding to all echos were reconstructed by utilizing a simple sum-of-square weighted averaging, and were reformatted by use of minium intensity projection(mIP).On either kind of mIP images micro hemorrhagic lesions were counted,the volume of lesions were measured and the number of revealed deep cerebral veins were counted on the slice of foramen level separately. We compared the two groups of data by use of paired t test..Results:The distribution and amount of hemorrhagic lesions on either of the two ESWAN sequences were identical,while the volume of lesions measured on long TE ESWAN sequence were larger than short TE ESWAN sequence(P=0.014),the amount of detected deep cerebral veins on long TE ESWAN sequence were more than short TE ESWAN sequence (P<0.0001). Conclusion:ESWAN is composed of 11 echos,it is a favorable virtue that guarantees sufficient magnetic susceptibility to detect micro hemorrhagic lesions of DAI,even if the TE of the first several echos is so short that arteries are bright from time-of-flight inflow bright. Along with the prolongation of TE,the amount of detected cerebral veins increase,but the blooming effect increase either and the magnify ratio of hemorrhagic size on ESWAN increase as well.We think that short TE ESWAN sequence is more appropriate for cerebral trauma examination.