Jei Hee Lee1, Bohyun Kim1, and Young Bae Kim2
1Radiology, Ajou University School of Medicine, Suwon, Korea, Republic of, 2Pathology, Ajou University School of Medicine, Suwon, Korea, Republic of
Synopsis
For visualization of capsule appearance on
HCC, GaMR is comparable with dynamic CT. And pathologic fibrous capsule can be
seen as hypointense rim in HBP.
Lower visualization of capsule appearance
in TP on GaMR seem to be parenchymal enhancement during the transitional phases.
Hyperintensity rim in T2 weighted images with
hopointense rim on HBP shows disruption of fibrous capsule with extension of
tumor cells across the fibrous capsule in pathology.
Hypointense rim on the HBP in the 2014
version of LI-RADS should not be considered as capsule appearance, and further
study is also needed.
Purpose
The presence of a capsule appearance has
often been considered a characteristic features of hepatocellular
carcinoma(HCC) with favorable prognostic factor. The presence and recognition
of a capsule appearance is important for the non-invasive diagnosis of HCC, and
Liver Imaging Reporting and Data System (LI-RADS) included capsule appearance
as a major feature. Recently, in the 2014 version of LI-RADS, various imaging
features on hepatobiliary phase(HBP) were considered as ancillary features. The
aim of this work was to establish the capability of gadoxeic acid-enhanced
MR(GaMR) imaging to detect the presence of the capsule appearance and to
correlate with dynamic CT and pathological findings.
Methods
Patients who underwent preoperative GaMR
and dynamic CT between Jan 2013 and Dec 2015 were included for analysis. And 63
patients (54 males and 9 females, mean age 55.8years) were pathologically
confirmed HCCs after surgical resection.
Two observers independently reviewed presence
or absence of capsule appearances on CT and GaMR images in the portal venous
phase(PP) and delayed phase(DP)/transitional phase(TP). Hypointense rims surrounding
nodules in HBP and T1- and T2-weighted images were considered as capsule in
GaMR and correlated with patholgy.
Interobserver agreement was assessed with
kappa statistics, and consensus opinions were reached by conference.
Pathological fibrous capsules were
classified as complete, partial, and no fibrous capsule. Pathological fibrous
capsules were compared and correlated with CT and GaMR, and sensitivity and specificity
were calculated.Results
Interobserver agreement for the presence or
absence of capsule appearance on GaMR was moderate on PP(k = 0.464; 95%
confidence interval(CI): 0.265, 0.663) and HBP(k = 0.508; 95% CI: 0.303, 0.713),
whereas fair in TP(k = 0.246; 95% CI: -0.0185, 0.511). Interobserver agreement for
capsule appearance on CT was good on DP (k = 0.617; 95% CI: 0.423, 0.811) and
moderate in PP(k = 0.555; 95% CI: 0.330, 0.781).
Table 1 and 2 summarized presence or
absence of capsule appearances on CT and GaMR with pathologic fibrous capsule. The
capsule was histologically evident in 46 of 63 nodules (73.0%), complete in 8 (12.7%),
and incomplete in 38(60.3%). Four cases(6.3%) were considered as pseudocapsule
with false-positive fibrous capsule on CT and GaMR images.
On GaMR, the capsule appearance was visible
in 22(34.9%) in PP, 16(25.4%) in TP and 23(36.5%) on HBP(Figure 1).
With CT, the capsule appearance was visible
in 43(68.3%) in PP and 33(52.4%) on DP. Sensitivity for pathologic capsule was
highest in DP on CT(54.4%) and specificity for pathologic capsule was highest in
HBP on GaMR(94.1%).
Four cases with hopointense
rim on HBP shows high signal intensity rim around the HCC in T2 weighted
images(Figure 2).Discussion
Hypointense rims surrounding nodules in HBP
were considered as capsule in our study, and that correlated with fibrous
capsule on pathology. Hypointense rim on the HBP in the 2014 version of LI-RADS
should not be considered as capsule appearance, because other cause of
hypointense rim on HBP should be considered and further study is also needed.
Four cases with hopointense rim on HBP
shows high signal intensity rim around the HCC in T2 weighted images. Pathologic
correlation shows that disruption of fibrous capsule with extension of tumor
cells across the fibrous capsule.
Lower visualization of capsule appearance
in TP on GaMR seem to be parenchymal enhancement during the transitional phases
may mask the capsule appearance, as both the tumor capsule and enhancing
adjacent hepatic parenchyma.Conclusion
For visualization of capsule appearance on
HCC, GaMR is comparable with dynamic CT. And pathologic fibrous capsule can be
seen as hypointense rim in HBP.
Lower visualization of capsule appearance
in TP on GaMR seem to be parenchymal enhancement during the transitional phases.
Hypointense rim on the HBP in the 2014
version of LI-RADS should not be considered as capsule appearance, and further
study is also needed.Acknowledgements
No acknowledgement found.References
1. Ishigami K, Yoshimitsu K,
Nishihara Y, et al. Hepatocellular
carcinoma with a pseudocapsule on gadolinium-enhanced MR images: correlation
with histopathologic findings. Radiology. 2009;250(2):435-443.
2. Dioguardi Burgio M, Picone D,
Cabibbo G, et al. MR-imaging features of hepatocellular carcinoma capsule
appearance in cirrhotic liver: comparison of gadoxetic acid and gadobenate
dimeglumine. Abdom Radiol. 2016;41(8):1546-1554.