| Part I A Comparative Study of the Respiratory Motion Artifacts Caused by Intravenous Administration of Gd-EOB-DTPA and Gd-DTPA in Dynamic Contrast-enhanced MR Imaging of the Liver[Purpose]:To compare the level of arterial respiratory motion artifacts related to the intravenous administration of Gd-EOB-DTPA with that of Gd-DTPA, and propose a feasible solution.[Materials and methods]:Seventy-five patients underwent dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) of the liver with Gd-DTPA and Gd-EOB-DTPA administration were collected and studied retrospectively during 1 year period. Single-breath-hold double phase acquisition was performed in both examinations. The double arterial, venous and late dynamic phases were acquired in all the patients. The same scan parameters were used. Respiratory motion artifacts on dynamic post-contrast images (arterial, venous and late dynamic) were blind-rated by 2 readers in consensus using a five-point scale. The following scale was used:score 1, extensive and images nondiagnostic; score 2, severe but images still interpretable; score 3, moderate with some effect on the diagnostic quality; score 4, minimal with no effect on diagnostic quality; score 5, none. Score of 3 or less was considered moderate to severe motion artifacts. Motion scores for each contrast media according to phase were compared by using Wilcoxon test. The frequency of patients with motion, as well as the frequency of those with moderate to severe motion, were compared by using McNemar’s Chi-square test.[Results]:The respiratory motion scores on the second arterial, portal and delayed phase in the Gd-EOB-DTPA cohort were worse than those in the Gd-DTPA cohort (Z=-5.058,-4.24,-3.625, P<0.001). A significantly higher ratio of respiratory motion artifacts (49.3% versus 6.7%,x2=28.26, P<0.001) and a higher ratio of moderate to severe respiratory motion artifacts (33.3% vs 2.7%,x2=21.04, P<0.001) were observed on late arterial phase in the Gd-EOB-DTPA cohort, in comparison to those of Gd-EOB-DTPA cohort. However, a slightely higher ratio of respiratory motion artifacts (9.3% vs 2.7%) and moderate to severe respiratory motion artifacts (4% vs 1.33%) were observed on early arterial phase in the Gd-EOB-DTPA cohort, in comparison to those of Gd-EOB-DTPA cohort (x2=1.744, P=0.180 and x2=0.25, P=0.625).[Conclusion]:Intravenous administration of Gd-EOB-DTPA causes more arterial phase artifacts than that caused by Gd-DTPA. Reducing breath-hold time of single phase and the use of arterial monitoring may improve the image quality of the arterial phase acquisition.Part II Different Imaging Patterns of Hepatocellular Carcinoma on Gd-EOB-DTPA MRI and the Correlation with the Histological Grade[Purpose]:To investigate the patterns of imaging appearance of hepatocellular carcinoma (HCC) on Gd-EOB-DTPA magnetic resonance images and correlation with the histological grade.[Materials and methods]:Eighty-two patients underwent Gd-EOB-DTPA MRI with 97 pathologically proven HCCs were analyzed. Typical enhancement of HCC during dynamic phases was defined as a combination of hyperintensity on arterial phase (AP) and hypointensity on portal venous phase (PVP) or delayed phase (DP), as decribed in the American Association for the Study of Liver Disease practice guidelines (AASLD). If a HCC did not meet the AASLD criteria, it was regarded as atypical enhancement. In addition, according to the signal on hepatobiliary phase (HBP) images, HCCs were classified as HBP hypointensity, HBP heterogenous intensity, HBP iso- to hyperintensity. As a consequence,9 imaging patterns of HCC on Gd-EOB-DTPA MRI were defined as follows:type Ⅰ, typical enhancement plus HBP hypointensity; type Ⅱ, typical enhancement plus HBP heterogenous intensity; type Ⅲ, typical enhacement plus HBP iso- to hyperintensity; type Ⅳ, atypical enhancement with arterial enhancement (no washout) plus HBP hypointensity; type Ⅴ, atypical enhancement with arterial enhancement (no washout) plus HBP heterogenous intensity; type Ⅵ, atypical enhancement with arterial enhancement (no washout) plus HBP iso- to hyperintensity; type Ⅶ, isointensity on AP and hypointensity on PVP or DP plus HBP hypointensity; type Ⅷ, isointensity on AP and hypointensity on PVP or DP plus HBP heterogenous intensity; type Ⅺ, isointensity on AP and hypointensity on PVP or DP plus HBP iso- to hyperintensity. The signal intensity ratio (SIR) of liver to lesion on HBP was measured. Mann-Whitney U test was used in the consideration of tumor size between the typical and atypical enhancement HCCs. Among the different histologic grades, the signal intensity and SIR of HCCs on HBP were compared using x2 test and Kruskal-Wallis test respectively.[Results]:All the 97 HCCs were defined into 5 imaging patterns, as follows:90.7%(88 of 97) of HCCs with type Ⅰ, Ⅱ, and Ⅲ, and 9.3%(9 of 97) with type Ⅳ and Ⅶ. Type Ⅰ, Ⅱ, and Ⅲ had typical enhancement and HBP hypointensity, heterogenous intensity, hyperintensity, retrospectively. Type Ⅵ and Ⅶ had atypical enhancement and HBP hypo intensity. HCCs with atypical enhancement pattern (type Ⅳ and Ⅶ, mean size,0.9 cm) were smaller than HCCs with typical enhancement pattern (type Ⅰ, Ⅱ and III, mean size,3.2 cm) (P=0.000). The signal intensity and SIR of liver to lesion on HCCs’ HBP were not correlated with histologic grades (P=0.669,0.638).[Conclusion]:HCCs can be classified into 5 imaging patterns on Gd-EOB-DTPA MRI. The combination of DCE MRI and HBP images are useful to the diagnosis of smaller HCCs. The signal intensity of liver to lesion on HCCs’ HBP were not correlated with histologic grades.Part Ⅲ Hepatobiliary Phase Imaging with Gd-EOB-DTPA for Detection and Characterization of Hepatocellular Carcinoma in Patients with Chronic Liver Disease[Purpose]:To evaluate hepatobiliary phase imaging (HBP) with Gd-EOB-DTPA for detection and characterization of hepatocellular carcinoma (HCC) in patients with chronic liver disease.[Materials and methods]:Ninety-nine patients with chronic liver disease and underwent Gd-EOB-DTPA MRI were analyzed in our hospital between September 2011 and May 2014.89 patients were pathologically or clinically diagnosed as HCC.10 patients were clinically diagnosed as normal to serve as a control. All the clinical diagnoses were made by two senior radiologists in consensus. Two senior radiologists compared the ability of HBP in detection of HCC with DWI in consensus. Another three junior and media grade radiologists independently reviewed each examination with and without HBP. Lesion detection, confidence scores and receiver operating characteristic (ROC) analysis were compared. The ability of HBP and DWI in detection of HCC were compared by using the McNemar test. Lesion detection and confidence scores using the "with HBP" and "without HBP" image sets were compared by using the Wilcoxon test.[Results]:130 nodules were diagnosed as HCC (n=111) and benign (n=19).100 HCCs were proved by pathology and 11 were clinically diagnosed. HBP has a higher sensitivity for HCCs than DWI (99.1% vs 90.1%, P=0.012). With the inclusion of the HBP, for HCCs≤1 cm in size, lesion detection was improved (93.2% vs 70.0%, P=0.016) and mean confidence scores significantly increased (P=0.001). For lesions≤1 cm in size, the diagnostic performance improved with the addition of the HBP (0.949 vs 0.744, P=0.0023).[Conclusion]:HBP with Gd-EOB-DTPA has a superior sensitivity than DWI. It may improve the detection and the diagnosis confidence of HCCs≤ 1 cm in patients with chronic liver disease. |
PDF Full Text Request