Research Presentation Session: Abdominal Viscera & GI Tract

RPS 101 - Pre- and post-treatment imaging in cirrhosis and HCC

February 28, 08:00 - 09:00 CET

7 min
Abbreviated liver MRI with second shot arterial phase image to assess the viability of treated HCC
Jaewon Han, Busan / Korea, Republic of
Author Block: S. Kim, N. K. Lee, S-B. Hong, J. Han; Busan/KR
Purpose: The purpose of this study was to evaluate the feasibility of abbreviated liver MRI (AMRI) with second shot arterial phase (SSAP) for the viability of treated HCC.
Methods or Background: We retrospectively identified 95 patients (70 men, 25 women; mean age, 68.7 years) between March 2021 and February 2022 meeting the following inclusion criteria; (1) patients with locoregional therapy for HCC who performed the modified gadoxetic acid-enhanced liver MRI protocol including the routine dynamic imaging after the first injection of 6 mL and SSAP imaging after the second injection of 4 mL and (2) available reference standard for tumour viability in treated observation. Two radiologists independently reviewed two MRI sets: full-protocol MRI and AMRI with SSAP. In the full-protocol MRI set, both reviewers assigned observations according to the liver imaging reporting and data system treatment response (LR-TR) algorithm (TR-viable, TR-equivocal, or TR-nonviable). In AMRI with SSAP set, we assigned a abbLR-TR category (abbLR-TR viable, abbLR-TR equivocal, or abbLR-TR nonviable) according to arterial hyperenhancement in SSAP and Hepatobiliary phase hypointensity, using the modified version of LI-RADS treatment response algorithm. We compared the diagnostic performance between two MRI sets.
Results or Findings: In 95 patients, 42 patients (44.2%) had viable lesions, and 53 patients (55.8%) had nonviable lesions. There was no significant difference in sensitivity for treated lesions between full-protocol MRI assigned according to the LR-TR algorithm and AMRI with SSAP assigned according to abbLR-TR category (Full-protocol, 73.8 % (31/42); AMRI with SSAP, 71.4 % (30/42); p=1.00). There was also no significant difference in specificity between two protocols (Full-protocol, 98.1 % (52/53); AMRI with SSAP, 96.2 % (51/53); p=1.00).
Conclusion: The abbLR-TR score in AMRI with SSAP showed the non-inferior diagnostic performance to the full-protocol MRI in evaluating the viability for the treated HCCs.
Limitations: No limitations were identified.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The study was approved by PNUH - 2309-013-131.
7 min
The imaging features of LI-RADS nonviable or equivocal lesions in the first follow-up evaluation after TACE for HCC to predict recurrence
Shu-Hang Zhang, Jiangsu / China
Author Block: S-H. Zhang, Y-C. Wang; Jiangsu/CN
Purpose: The purpose of this study was to investigate whether the imaging features of lesions evaluated as LI-RADS nonviable or equivocal during the first post-TACE imaging follow-up can predict recurrence in HCC patients.
Methods or Background: A total of 171 lesions from 149 patients were included in this study. All patients underwent at least three enhanced magnetic resonance imaging examinations, with the first one conducted before TACE and the second one during the first follow-up visit the one to two months after TACE. The lesions were divided into two groups, no recurrence during long-term follow-up (n = 128) and recurrence during long-term follow-up (n = 43). Imaging features evaluated included irregular shape, size, internal homogeneity, non-smooth margin, arterial peritumoural enhancement, rim enhancement in portal venous phase or delay phase, peritumoural hyperintensity in T2WI or DWI. Using univariate and multivariate logistic regression analysis to investigate which imaging features at the first follow-up can predict lesion recurrence at 6-12 months after TACE.
Results or Findings: After univariate and multivariate analysis showed that non-smooth margin (OR, 3.96, 95% CI, 1.66, 9.44, P = 0.002) and peritumoural hyperintensity in T2WI or DWI (OR, 7.74, 95% CI, 3.32, 18.06, P < 0.001) were independent risk factors for recurrence of LI-RADS nonviable lesions at 6 to 12 months. The area under the ROC curve using these two factors to predict the recurrence of LI-RADS nonviable lesions at 6 to 12 months was 0.754.
Conclusion: Non-smooth border and peritumoural hyperintensity in T2WI or DWI are imaging features to predict recurrence in HCC patients with LI-RADS nonviable lesions evaluated on the first post-TACE imaging follow-up.
Limitations: This is a retrospective study where selection bias is an inherent flaw.
It can be included in subsequent research for analysis of patient baseline imaging features.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: This is a retrospective study. It was approved for retrospective analyses by the institutional review board of this hospital, with an informed consent waiver.
7 min
MRI evaluations of focal liver reactions (FLR) following stereotactic body radiotherapy (SBRT) for small hepatocellular carcinoma: relationship between MRI features and dose-volumetric parameters
Runqian Huang, Guangzhou / China
Author Block: R. Huang; Guangzhou/CN
Purpose: The aim of this study was to assess the MRI features of FLR for patients with small hepatocellular carcinoma (HCC) treated with stereotactic body radiotherapy (SBRT), and to investigate the relationship between MRI features of FLR and dose–volumetric parameters.
Methods or Background: Between September 2018 and April 2021, 102 patients with small HCC treated with SBRT were evaluated according to the entry criteria. After SBRT, all patients underwent regular follow-ups with blood tests and dynamic MR scans at a median follow-up of more than 18 months. Two radiologists assessed imaging features of FLR in MRI sequences at 3 to 6 months. To assess the deterioration of hepatic function, we evaluated the ALBI score. To identify the parameters of predicting the radiation-induced hepatic toxicity, univariate and multivariate analyses were used to select independent risk factors and a nomogram model was established.
Results or Findings: The MRI features of FLR were divided into three types according to T1WI, T2WI, arterial phase, portal-venous phase and delayed phase signals: low/high/high/high/high in 27 patients (Type A), low/low/low/low/low in 48 patients (Type B) and low/low/iso (or high)/iso (or high)/high in 27 patients (Type C). ALBI score at 3 to 6 months after SBRT was the significant factor that differed between Type A (worse prognosis) and alternatives (Type B and C). Multivariate analysis showed that PTV (P=0.028) and V5Gy (normal liver volumes receiving from 5Gy) (P=0.003) were the significant dose–volumetric parameters for predicting Type A. Both factors were used to construct the nomogram model, which showed good performance (AUC=0.772).
Conclusion: The MR features of FLR can reflect the deterioration of the hepatic function for small HCC by SBRT. The PTV and V5Gy were the significant parameters that predicted the risk of the deterioration of hepatic function for different MR types of FLR.
Limitations: No limitations were identified.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: No information provided by the submitter.
7 min
Ascites in patients with hepatocellular carcinoma undergoing transarterial chemoembolisation: the more, the worse?
Lukas Müller, Mainz / Germany
Author Block: L. Müller1, D. Bender1, F. Stöhr1, A. Mähringer-Kunz1, J. Mittler1, A. Weinmann1, R. Klöckner2, F. Hahn1; 1Mainz/DE, 2Lübeck/DE
Purpose: Preliminary work has shown that portal hypertension plays a key role for the prognosis in patients with hepatocellular carcinoma (HCC) undergoing transarterial chemoembolisation (TACE). Specifically, the presence of ascites appears to be a strong negative predictor for these patients. However, it remains unclear whether different ascites volumes influence prognosis. Therefore, the aim of this work was to investigate the influence of different ascites volumes on survival for patients with HCC undergoing TACE.
Methods or Background: A total of 351 patients with HCC treated at our tertiary referral centre between 2010 and 2020 were included. In patients with ascites, the fluid was segmented, and the volume quantified by slice-wise addition using contrast-enhanced CT imaging. The patients with ascites were subclassified using cut-off values previously suggested for patients with liver cirrhosis: A0 = 0ml, A1 ≤ 300ml, and A2 > 300ml ascites. Median overall survival (OS) was calculated.
Results or Findings: Ascites was present in 102 (29.1%) patients. Median OS without ascites was 16.5 months, and therefore significantly longer than in patients with ascites (6.4 months, P<0.001). Patients were further classified as n(A0)=249(70.9%), n(A1)=43(12.3%), n(A2)=59(16.8%). The corresponding median OS yielded 16.5, 9.9 and 4.0 months for A0-A2, respectively (overall P<0.001, pairwise comparison A0 vs A1 P=0.03, A1 vs A2 P<0.001).
Conclusion: Ascites in patients with HCC undergoing TACE is associated with a bad prognosis. Our results indicate an association between the amount of ascites and OS. Not only the presence but also the amount of ascites is relevant. The quantitative assessment applied here can help improve clinical decision making.
Limitations: The retrospective design and single centre nature of the study were identified as limitations.
Funding for this study: This study was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation); project number: 518477942.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The study has been approved by the local ethics committee responsible (Rhineland-Palatinate, Germany).
7 min
Adding estimates for central venous pressure boosts performance of non-invasive assessment of portosystemic gradient prior to TIPS implantation
Felix Hahn, Mainz / Germany
Author Block: F. Stoehr1, L. Müller1, T. Loew1, C. Labenz1, J. Mittler1, R. Klöckner2, M. B. B. Pitton1, F. Hahn1; 1Mainz/DE, 2Lübeck/DE
Purpose: We aimed to evaluate the accuracy of non-invasive scores to predict markedly increased portosystemic gradient (PSG) in patients with clinically significant portal hypertension (CSPH) undergoing TIPS procedure and to further investigate patients without markedly increased PSG at the time of intervention.
Methods or Background: We included patients who underwent TIPS implantation at our tertiary care center between 2010 and 2022 with CT imaging and complete laboratory workup prior to TIPS and without history of splenectomy. Liver and spleen volume were automatically assessed using a commercially available in-house trained segmentation software. Scores using CT data by Iranmanesh et al. and Kihira et al. were assessed with regard to classification accuracy. Moreover, differences between the subgroups of patients with and without PSG>10mmHg were analysed.
Results or Findings: Preliminary results after assessing 200 patients show a mean PSG prior to TIPS of 16.5±4.8mmHg. Out of these patients, 22 with CSPH had a PSG≤10mmHg at the time of the TIPS procedure. Both scores showed high sensitivity (89% and 87% resp.), but poor specificity (36% and 41% resp.), in predicting a markedly increased PSG. ROC-analysis yielded an AUC of 0.67 and 0.70. In the subgroup of patients with low PSG, central venous pressure (CVP) was significantly elevated compared to patients with high PSG (13.4±5.9mmHg vs 7.7±3.6mmHg, p=0.003), while distribution of portal venous pressure was not significantly different (p=0.51). Adding IVC diameter as an estimator for CVP in linear regression resulted in a significant AUC increase to 0.78 and 0.79 resp. (p=0.04 and p=0.01).
Conclusion: Including IVC diameter boosted the accuracy of CT morphological scores for assessing elevated PSG, especially increasing specificity.
Limitations: This was a single-centre, retrospective study.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The study was approved by the local ethics committee (Medical Association of Rhineland Palatinate, Mainz, Germany). Informed consent was waived due to the retrospective nature of the study.
7 min
The comparison of preoperative and intraoperative graft volumes in living donor liver transplantation: effect of graft type on accuracy
Hande Özen Atalay, Istanbul / Turkey
Author Block: H. Özen Atalay, T. Kanmaz, A. Durur Karakaya; Istanbul/TR
Purpose: The purpose of this study was to conduct a retrospective evaluation of the difference between the preoperative estimated and the actual intraoperative graft volume detected in living liver transplantation donors in terms of graft types.
Methods or Background: The present retrospective study was performed at a single centre, involving a total of 219 donors. The analysis of total, right, left lobe and left lateral segment liver volumes were conducted in the venous phase abdominal CT by using Myrian software. The preoperative graft volumes were compared to the intraoperative graft volumes. Intraclass correlation coefficient, Blant Altman and one-sample tests were used to calculate correlation coefficient and to detect error of estimation.
Results or Findings: 128 donors underwent right hepatectomy, 81 donors had left lateral segmentectomy and 10 donors had left hepatectomy. The estimated mean volumes of the right lobe, left lobe, and left lateral segment were 885.39 mL, 402.2 mL, and 241.14 mL, respectively. The mean intraoperative weights of the right lobe, left lobe, and left lateral segment were 820.65 mL, 504.2 mL, and 250.82 mL, respectively. The automated preoperative right lobe volume assessment had a higher correlation with intraoperative graft weight (r=0,753) than the left lobe and left lateral segment volume assessment (r=0.468 and r=0.007, respectively). The mean percentages of error in volume estimation were calculated at 12,20%, 23,79% and 68,53% for the right lobe, left lobe and left lateral segment, respectively.
Conclusion: This study underscores the need for more accurate preoperative volume estimations, particularly for the left lobe and left lateral segment. These areas showed lower correlation, higher error ratios, suggesting that extra caution is needed when planning transplantation procedures involving these segments.
Limitations: Our retrospective study included a limited number of patients from a single institution.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: This study was approved by Koc University Biomedical Research Ethics Committee, Istanbul/Turkey.

This session will not be streamed, nor will it be available on-demand!