Page 106 - Read Online

Brunsing et al. Hepatoma Res 2020;6:59 I http://dx.doi.org/10.20517/2394-5079.2020.50 Page 9 of 16

short interval follow-up or call-back. The requirements for a power injector and for precise arterial phase
timing complicate the workflow compared to other AMRI approaches. HCC detection accuracy for
dynamic AMRI needs to be validated prospectively in a surveillance patient cohort.

HEPATOBILARY-PHASE AMRI
Imaging
HBP contrast-enhanced AMRI (HBP-AMRI), the other AMRI approach that utilizes GBCA, is performed
after administration of the hepatobiliary agent, gadoxetate disodium. The sequences include:

Hepatobiliary phase imaging
Acquired about 15-20 min following the administration of gadoxetate, when parenchymal enhancement
with this agent is expected to be maximal, the hepatobiliary phase T1-weighted images provide high
contrast-to-noise for lesion detection. In the hepatobiliary phase (HBP masses that are not of benign
hepatocellular nature (e.g., HCCs and non-HCC malignant neoplasms) are hypointense relative to the high
signal background liver, creating high liver to lesion contrast and increasing sensitivity. Hepatobiliary phase
hypointensity is not specific for malignant nodules, however, and can be seen in benign non-hepatocellular
entities, such as cysts and hemangiomas. Hence, any detected lesion must be correlated on T2-weighted
imaging. If IP/OOP images are acquired, they may permit assessment of relative fat or iron content relative
to liver, as described for the other AMRI approaches.

T2 weighted imaging
T2 weighted imaging is included to increase specificity. Benign lesions like cysts or hemangiomas have
high intrinsic T2 signal and can be readily identified, while marked T2 darkness also suggests benignity,
which helps with reducing unnecessary call-backs. In contrast, HCC tends to be mildly to moderately T2
hyperintense.

Optional: DWI
Similar to NC-AMRI, inclusion of DWI is meant to increase sensitivity for malignancy via a mechanism
distinct from HBP imaging. Some DWI features may also be used to help differentiate HCC from non-
[43]
HCC malignancy, such as intrahepatic cholangiocarcinoma (ICC), as discussed earlier .

Reporting
Reporting of HBP-AMRI is the most developed of all AMRI approaches since HBP-AMRI has been
implemented in clinical practice in selected centers in the United States. HBP-AMRI reporting mirrors
that of LI-RADS US surveillance reporting with three outcomes: Positive (suspicious nodules ≥ 1 cm),
[49]
subthreshold (suspicious nodules < 1 cm), and negative (no suspicious nodules) . Positive examinations
[31]
prompt call back for diagnostic MRI or CT. The scoring of HBP-AMRI has been reported previously ,
with an example provided in Figure 5.

Advantages
HBP-AMRI provides several advantages. The core T1-weighted HBP images have high-contrast-to-noise,
aiding in lesion detection. Importantly, hepatocytes retain gadoxetate for an extended period of time.
Thus, images can be repeated as necessary. The 20-min delay also allows hand injection of contrast while
the patient is in the waiting room, which simplifies workflow, reduces the time the patient is on the MRI
table, thus reducing the examination cost, and diminishes the chance of contrast extravasation. This also
eliminates the need for a power injector. Finally, HBP-AMRI are reported and interpreted using a simple
scoring system modeled from LI-RADS US surveillance , which many radiologists are already familiar
[54]
with, in theory facilitating implementation.