The Application Of Gd-EOB-DTPA-enhanced MR Imaging For The Evaluation Of Liver Function

Objective To evaluate the feasibility of hepatocyte-specific phase liver enhancement for the evaluation of liver function by gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid(Gd-EOB-DTPA) enhanced magnetic resonance imaging(MRI), to detect the optimal method, the potential ability for the evaluation of regional liver function, and influencing factors of liver enhancement.Methods This prospective study was approved by the hospital institutional review board. All patients gave written informed consents. From October 2013 to January2015, sixty-six patients with different liver function were included in this study.According to the liver function, they were divided into four groups: normal liver function(NLF) group(n=23, 11 male, 12 female, 41±9 years), liver cirrhosis with Child-Pugh class A(LCA) group(n=25, 22 male, 3 female, 52±9 years), liver cirrhosis with Child-Pugh class B(LCB) group(n=10, 7 male, 3 female, 58±11 years),liver cirrhosis with Child-Pugh class C(LCC) group(n=8, 6 male, 2 female, 50±3years). All individuals had plain MR images before injection, and hepatocyte-specific phase images were obtained at 5, 10, 15 and 20 minutes after bolus administration of Gd-EOB-DTPA. The signal intensity(SI) of each liver segment and structures which were selected as references, plain liver, spleen, spinal cord, erector spinal muscle and saline, were measured. Relative liver enhancement of the whole liver, the left hemiliver, the right hemiliver and each liver segment were calculated by dividing the SI of liver parenchymal at t min after injection by precontrast SI of liver or SI of different references at the same time point. The blood biochemical parameters were recorded: platelet count(PLT), total bilirubin level(TBIL), albumin level(Alb),globulin level(Glb), alanine aminotransferase level(ALT), aspartate aminotransferase level(AST), alkaline phosphatase level(ALP),gamma-glutamyltransferase level(γ-GT), prothrombin time(PT), glomerular filtration rate(GFR). The one-way ANOVA was used to compare the intra- and inter-group differences of RE, SNK was used for further pairwise comparison. The receiver operating characteristic(ROC) curves was used to analyze the diagnostic efficacy of RE at each time point for liver cirrhosis or moderate-severe cirrhosis. The paired t-test was used to compare the RE of liver parenchyma between the right and left hemiliver in different groups. The one-way ANOVA was used to compare the RE among liver segments. Spearman or Pearson correlation test was used to analyze the correlation between RE and biochemical parameters, and multiple stepwise linear regression analysis was used to determine the independent factors of RE.Results(1) In NLF and LCA group, RE of the whole liver parenchymal increased gradually from 5 to 20 minutes. In LCB group, the changes of RE was moderate. In LCC group, RE decreased gradully.(2) RE of the whole liver parenchymal decreased as the severity of liver cirrhosis advanced, and significant differences were found between groups at each time point of hepatocyte-specific phase(P<0.01).(3) Plain liver as the reference, differences of REliver were found between all groups at 15 and 20 minutes of hepatocyte-specific phase(P<0.05). Spleen as the reference, difference of REliver was not found between NLF and LCA group at all time points(P>0.05). Spinal cord or erector spinal muscle as the reference, difference of REliver was not found between LCB and LCC group at all time points(P>0.05). Saline as the reference, differences of REliver were not found between NLF and LCA group,LCB and LCC group at all time points(P>0.05).(4) Intra-comparison of REliver at each time point showed that significant difference was only found between REliver of 5 minutes and 20 minutes in NLF group(P<0.05). In other cirrhotic groups, there were no differences(P>0.05).(5) REliver at 10, 15, 20 minutes had same diagnostic efficacy for liver cirrhosis or moderate-severe cirrhosis(P>0.05), and the diagnostic sensitivity, specificity,positive and negative predictive values for moderate-severe cirrhosis were all more than 80%.(6) REliver of the right and left hemilivers decreased as the severity of liver cirrhosis advanced, and REliver of the right hemiliver decreased quicker than the left hemiliver. In NLF, LCA and LCB group, significant differences were found between right and left hemiliver(P<0.05). In LCC group, there was no difference between the two hemilivers(P>0.05).(7) REliver of each liver segment decreased as the severity of liver cirrhosis advanced, and significant differences were found among the four groups in each segment(P<0.001). In NLF and LCA group, significant differences were found between all segments(P<0.001). In LCB and LCC group, there was no difference between them(P>0.05).(8) REliver at 15 minutes was correlated with PLT and Alb positively, and TBIL,Glb, AST, ALP, γ-GT and PT negatively(P<0.01). Multiple regression showed TBIL and Alb were correlated with REliver independently(P<0.001).Conclusion Liver enhancement of hepatocyte-specific phase with Gd-EOB-DTPA can be used to evaluate liver function. The plain liver is the optimal reference, and liver enhancement at 15 minutes is useful for differentiating normal-liver and patients in different Child-Pugh groups. The liver enhancement at 10 minutes can be used as index for the diagnosis of liver cirrhosis or moderate-severe cirrhosis. Liver enhancement has the potential to evaluate regional liver function. TBIL and Alb are the independent influencing factors of liver enhancement.