| Part 1 Blood suppression inversion time settings of non-contrast-enhanced magnetic resonance portography Objective:To investigate the effect of the blood suppression inversion time(BSP TI)settings on the imaging quality of portal vein with non-contrast-enhanced magnetic resonance portography(NCE-MRP)using multiple spatial inversion pulses(SLEEK).Materials and Methods:Twenty-eight patients suspected of portal hypertension were included in the study.After overnight fast all the patients were examined with SLEEK-MRP series performed with different blood suppression inversion time(BSP TI)settings of 600 ms,900ms,1200 ms and 1500 ms.The original images were transmitted to an imaging workstation and processed with multiple plane reconstruction(MPR),volume rendering(VR)and maximum intensity projection(MIP)to finely delineate small branches of the portal vein.Assessments of imaging quality were evaluated by 2 physicians and portal venous branching grades were rated.Regions of interest(ROIs)were placed in main portal vein(MPV),left portal vein(LPV),right portal vein(RPV)and adjacent liver parenchyma.Signal intensity was recorded and the signal contrast(Cv/l)between MPV,LPV,RPV and liver parenchyma was calculated.The consistency of the imaging quality evaluated by 2physicians was assessed.The consistency of the imaging quality of different BSP TI setting groups was assessed.Branching grading scores and Cv/l of MPV,LPV,RPV of different BSP TI setting groups were compared.Results: The consistency between the two physicians evaluated imaging quality scores was good,and the Kappa coefficient was 0.839.There was no significantdifference in the imaging quality scores between the BSP TI settings groups(p=0.481).Branching grading scores of LPV and RPV increased significantly in each group,while the difference between BSP TI=1200ms group and BSP TI=1500ms group decreased.In group BSP TI=600ms,900 ms,1200ms,1500 ms,branching grades of LPV were 2.0±0.7,2.6±1.0,3.3±1.1,3.5±1.1 separately;branching grades of RPV were 2.6±0.8,3.5±0.9,4.0±0.9,4.2±0.8 separately;the sum grades of LPV and RPV were 4.6±1.4,6.0±2.0,7.4±2.3,7.7±2.2 separately.Cv/l of MPV,LPV and RPV decreased in each group,while the difference between BSP TI=1200ms group and BSP TI=1500ms group was not statistically significant.In group BSP TI=600ms,900 ms,1200ms,1500 ms,Cv/l of MPV were 3.9±1.2,2.8±0.6,2.4±0.5,2.2±0.4separately,Cv/l of LPV were 3.5±1.3,2.7±0.7,2.4±0.5,2.2±0.5 separately,Cv/l of RPV were 3.7±1.4,2.8±0.7,2.4±0.5,2.3±0.5 separately.Conclusion:Increasing of BSP TI,an important parameter of SLEEK-MRP,helps to display small branches of portal vein,while reducing the contrast between vessels and background tissue.At a BSP TI setting of 1200 ms,portal vein can be visualized with small branches clearly.Part 2 Feasibility of Non-Contrast-Enhanced Magnetic Resonance Angiography for Evaluating Portal Venous HemodynamicsObjective : The purpose of the current study is to investigate the feasibility of non-contrast-enhanced magnetic resonance angiography(NCE-MRA)using spatial labeling with multiple inversion pulses(SLEEK)in evaluating portal venous hemodynamics.Materials and Methods:Twenty-five patients suspected of portal hypertension were included in the study.After overnight fast all the patients were examined with SLEEK-MRA series performed with different blood suppression inversion time(BSP TI)settings of 600 ms,900ms,1200 ms and 1500 ms.Subsequently they underwent color doppler ultrasonography(CDUS)within one day,measuring velocity of the main portal vein(MPV),left portal vein(LPV)and right portal vein(RPV).Original images of SLEEK-MRA were transmitted to an imaging workstation and processed with multiple plane reconstruction(MPR),volume rendering(VR)and maximum intensity projection(MIP)to finely delineate small branches of the portal vein.Portal venous branching grades of SLEEK-MRA series were analyzed with CDUS-measured hemodynamics using Spearman rank correlation coefficient(rs).Patients were divided into 2 groups depending on main portal venous velocity.The sum of branching grades of LPV and RPV at all BSP TI settings and the BSP TI values applied to reach a same branching grading standard were compared between the 2 groups.Results:At all BSP TI settings,branching grade scores of LPV and RPV were of statistically significant correlation with CDUS-measured flow velocity of LPV and RPV respectively.The correlation coefficient of LPV at each BSP TI settings(600ms,900 ms,1200ms,1500ms)were 0.534(p=0.007),0.408(p=0.048),0.563(p=0.004),0.621(p=0.001)respectively.The correlation coefficient of RPV at each BSP TI settings were 0.655(p=0.001),0.578(p=0.004),0.699(p<0.001),0.641(p=0.001)respectively.For patients of obviously decreased portal flow velocity group,branching grade scores were significantly lower than patients of normal or slightly decreased portal flow velocity group.To reach the same portal venous branching visualization,patients of obviously decreased portal flow velocity need to be applied with higher BSP TI settings.While most of patients of normal or slightly decreased portal flow velocity were applied 900 ms to reach grade 6,42.9% of patients of obviously decreased portal flow velocity could not reach it even applied a BSP TI setting of 1500 ms.Conclusion:NCE-MRA has the potential to suggest portal venous flow changes with high-quality morphological visualization of portal vein imaging simultaneously.It is feasible for NCE-MRA to provide hemodynamical information. |
PDF Full Text Request