Table 1.
Patient Clinical Information and Focal Hepatic Lesion Characteristics.
Table 2.
Mean Lesion-to-liver Contrast-to-Noise Ratio (CNR) on 10-minute Delayed Hepatocyte Phase Imaging (HPI) using a 30° Flip Angle (FA) (10m-FA30), 20-minute Delayed HPI using a 10° FA (20m-FA10), and 20-minute Delayed HPI using a 10° FA (20m-FA10).
Fig 1.
Lesion-to-liver contrast-to-noise ratio (CNR) for HCCs.
Box-and-whisker plots showing the median (middle line of each box), quartiles (top and bottom lines of each box), and upper and lower adjacent (upper and lower whiskers for each box) values of the lesion-to-liver contrast-to-noise ratio (CNR) for HCCs. HCCs were imaged according to three different protocols: 10-minute delayed hepatocyte phase imaging (HPI) with a 30° flip angle (10m-FA30), 20-minute delayed HPI with a 10° flip angle (20m-FA10) or a 30° flip angle (20m-FA30). The CNR for HCCs on 10m-FA30 was significantly higher than that on 20m-FA10 and significantly lower than that on 20m-FA10, in both all patients and those just in group A (with chronic hepatitis or Child-Pugh A cirrhosis). In group B with Child-Pugh B or C cirrhosis, the CNR for HCCs on 10m-FA30 was significantly higher than that on 20m-FA10 and was not different from that on 20m-FA30.
Fig 2.
A 61-year-old man with HCC and hepatitis B-related compensated liver cirrhosis (Child-Pugh score 5).
Hepatocyte phase imaging (HPI) were acquired using 30° flip angle (FA) at 10-minute delayed time (10m-FA30) (A), 10° FA at 20-minute delayed time (20m-FA10) (B), and 30° FA at 20-minute delayed time (20m-FA30) (C) after Gd-EOB-DTPA administration. These images are viewed at the same window level (400) and width (700). HCC in segment 8 appears as low signal intensity related to the surrounding liver parenchyma on all three HPIs (arrow). Lesion-to-liver CNRs on 10m-FA30, 20m-FA10, and 20m-FA30 were 222.8, 147.5, and 343.1, respectively. The two reviewers’ mean subjective scores for the presence of focal hepatic lesions were 4 (4 and 4) on all three HPIs.
Fig 3.
A 76-year-old woman with HCC and hepatitis C-related decompensated liver cirrhosis (Child-Pugh score 8).
Hepatocyte phase imaging (HPI) were acquired using 30° flip angle (FA) at 10-minute delayed time (10m-FA30) (A), 10° FA at 20-minute delayed time (20m-FA10) (B), and 30° FA at 20-minute delayed time (20m-FA30) (C) after Gd-EOB-DTPA administration. These images are viewed at the same window level (400) and width (700). HCC in segments 5/6 appears as low signal intensity related to the surrounding liver parenchyma on all three HPIs (arrow). Lesion-to-liver CNRs on 10m-FA30, 20m-FA10, and 20m-FA30 were 35.0, 12.2, and 59.6, respectively. The two reviewers’ mean subjective scores for the presence of a focal hepatic lesion were 4 (4 and 4) on both 10m-FA30 and 20m-FA30 and 2 (2 and 2) on 20m-FA10.
Fig 4.
Lesion-to-liver contrast-to-noise ratio (CNR) for HCCs according to Child-Pugh and MELD scores.
The relationships between CNR and Child-Pugh score (A) are as follows: y = -27.2x + 300.3 (r = - 0.491) for 10m-FA30, y = -17.5x + 200.5 (r = - 0.446) for 20m-FA10; and y = -36.3x + 390.1 (r = - 0.503) for 20m-FA30. The relationships between CNR and MELD score (B) are as follows: y = -8.4x + 217.3 (r = - 0.378) for 10m-FA30; y = -4.8x + 140.5 (r = - 0.304) for 20m-FA10; and y = -11.4x + 281.1 (r = - 0.393) for 20m-FA30.
Fig 5.
The correlation of image noise between on 20m-FA10 and 20m-FA30.
There was no significant difference in noise between on 20m-FA10 (mean noise ± standard deviation, 1.33 ± 0.45) and on 20m-FA30 (1.35 ± 0.43), which suggests the image noise was not influenced when applying a 30° flip angle instead of a 10° flip angle.
Table 3.
Detection Sensitivity of Focal Hepatic Lesions for the Two Readers on 10-minute Delayed Hepatocyte Phase Imaging (HPI) using a 30° Flip Angle (10m-FA30), 20-minute Delayed HPI using a 10° Flip Angle (20m-FA10) or a 30° Flip Angle (20m-FA30).