RSNA 2014 

Abstract Archives of the RSNA, 2014


VSGI21-08

A Review of LI-RADS Categorization in 201 Pathology Proven Hepatocellular Carcinomas

Scientific Papers

Presented on December 1, 2014
Presented as part of VSGI21: Gastrointestinal Series: Imaging of the Cirrhotic Patient

Participants

Eric Christopher Ehman MD, Presenter: Nothing to Disclose
Spencer Caton Behr MD, Abstract Co-Author: Research Grant, General Electric Company
Rizwan Aslam MBBCh, Abstract Co-Author: Research support, Bayer AG
Benjamin M. Yeh MD, Abstract Co-Author: Research Grant, General Electric Company Consultant, General Electric Company
Linda Ferrell MD, Abstract Co-Author: Nothing to Disclose
Thomas A. Hope MD, Abstract Co-Author: Speaker, Guerbet SA Research Grant, General Electric Company

PURPOSE

To explore the trends in imaging appearance and differences in findings by modality for the new LI-RADS v2014 definitions in a large group of pathology proven cases of hepatocellular carcinoma.

METHOD AND MATERIALS

Pathology reports from liver specimens (explants and partial hepatectomies) of 605 sequential patients with cirrhosis were reviewed to identify specimens with at least one focus of viable hepatocellular carcinoma, then cross-correlated with pre-operative CT and MR imaging. Patients with completely necrotic treated tumor, those without available prior pre-treatment multiphase imaging and tumors smaller than 1 cm were excluded. Each lesion was examined, the imaging features recorded, and the lesion retrospectively graded using the LI-RADS 2014 criteria.

RESULTS

147 patients with a total of 201 hepatocellular carcinomas diagnosed between 12/2008 and 10/2013 were analyzed. Average time between the most recent pre-treatment prior imaging study and surgery was 13 months. 150 (75%) lesions were imaged by multiphase CT, and 51 (25%) lesions by MRI. Overall, 64 (32%) lesions measured ≥1cm and <2cm, while 137 (68%) were ≥2cm. There were 21 (13%) LIRADS-3 lesions, 75 (37%) LIRADS-4 lesions and 102 (50%) LIRADS-5 lesions. 171 (85%) of lesions exhibited arterial hyperenhancement, 136 (68%) demonstrated washout and 29 (14%) showed evidence of capsule. At CT, the rate of LIRADS-3, -4 and -5 lesions was 13%, 37% and 50% respectively. At MR, these rates were 4%, 39% and 55%. At CT, 13% of 1-2 cm lesions were graded LIRADS-5, and at MR, 38% were graded LIRADS-5. Arterial phase hyperintensity and washout appearance rates were equivalent between MR and CT, but capsule appearance was more common on MR (29%) imaging than at CT (10%), with χ2 = 10.7 (p<0.05).

CONCLUSION

The rate of arterial enhancement and portal venous or delayed washout are similar between lesions diagnosed via CT and those diagnosed with MR. Capsule appearance was seen significantly more frequently at MR, resulting in a higher rate of LIRADS-5 lesions measuring 1-2 cm at MR compared to CT.

CLINICAL RELEVANCE/APPLICATION

Differences in sensitivity for LI-RADS 5 lesions exist for MR and CT, which may support the use of MR imaging for the evaluation of HCC over that of CT in the pre-transplant population

Cite This Abstract

Ehman, E, Behr, S, Aslam, R, Yeh, B, Ferrell, L, Hope, T, A Review of LI-RADS Categorization in 201 Pathology Proven Hepatocellular Carcinomas.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14016960.html