Abstract

To evaluate the role of dysmorphic intratumoral vessels as imaging marker for the prediction of high lung shunt fraction (LSF) in patients with hepatocellular carcinoma (HCC). This retrospective study included 403 patients with HCC who underwent a planning arteriography for transarterial radioembolization with administration of 99mTc-macroaggregated albumin to calculate LSF. The LSF was measured by using planar body scans. Two radiologists evaluated the pre-treatment contrast-enhanced CT findings, including tumor number, size, margin, distribution, tumor burden, portal and hepatic vein invasion, early hepatic vein enhancement, and dysmorphic intratumoral vessels. The logistic regression analysis was performed to determine significant predictors for high LSF > 20%. Using the identified predictors, diagnostic criteria for high LSF were proposed. Among 403 patients, 52 (13%) patients had high LSF > 20%, and dysmorphic tumor vessels were present in 115 (28.5%) patients. Predictors for LSF > 20% were tumor size > 11 cm, hepatic vein invasion, early hepatic vein enhancement, and dysmorphic intratumoral vessel. If the patient had three or more of the four predictors for LSF > 20% on imaging, the accuracy and specificity for diagnosing LSF > 20% were 88.8% and 96.3% respectively. Dysmorphic intratumoral vessel in HCC is an imaging marker suggesting a high LSF, which may be applicable to treatment modification or patient exclusion for radioembolization with combined interpretation of tumor size and hepatic vein abnormality.

Details

Title
Association of dysmorphic intratumoral vessel with high lung shunt fraction in patients with hepatocellular carcinoma
Author
Choi, Tae Won 1 ; Joo, Ijin 1 ; Kim, Hyo-Cheol 1 

 Seoul National University Hospital, Department of Radiology, Seoul, Republic of Korea (GRID:grid.412484.f) (ISNI:0000 0001 0302 820X) 
Publication year
2022
Publication date
2022
Publisher
Nature Publishing Group
e-ISSN
20452322
Source type
Scholarly Journal
Language of publication
English
ProQuest document ID
2704767562
Copyright
© The Author(s) 2022. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.