RPS 301a - Upper abdominal imaging: what's new?

Author Block: A. D. Rónaszéki, B. K. Budai, R. Stollmayer, B. Csongrády, K. Hagymási, I. Kalina, G. Györi, P. Maurovich-Horvat, P. N. Kaposi-Novák; Budapest/HU
Purpose or Learning Objective: To assess the feasibility of tissue attenuation imaging (TAI) and tissue scatter distribution imaging (TSI) for ultrasound-based quantification of liver steatosis in nonalcoholic fatty liver disease (NAFLD).
Methods or Background: We prospectively enrolled 101 participants with suspected NAFLD. RS85A ultrasound scanner was used with TAI and TSI applications. Patients were divided into ≤5%, 5%-10%, and ≥10% groups based on magnetic resonance imaging proton density fat-fraction (MRI-PDFF). Spearman’s correlation, logistic regression and area under the receiver operating characteristic curve (AUROC) were calculated for both TAI and TSI. Interobserver reliability was assessed with the intraclass correlation coefficient (ICC).
Results or Findings: Both TAI (ρ=0.78, P<0.001) and TSI (ρ=0.68, P<0.001) showed a significant correlation with MRI-PDFF. TAI overperformed TSI in the detection of both ≥5% MRI-PDFF (AUROC=0.89 vs. 0.87) and ≥10% (AUROC=0.93 vs. 0.86). In linear regression analysis, MRI-PDFF proved to be an independent predictor of both TAI (β=1.03; P<0.001) and TSI (β=50.9; P<0.001; while liver stiffness (β=-0.86; P<0.001) was negatively associated with TSI. Interobserver analysis showed excellent reproducibility with TAI (ICC=0.95) and moderate reproducibility with TSI (ICC=0.73).
Conclusion: TAI and TSI could be used successfully to measure the severity of hepatic steatosis in clinical practice.
Limitations: Limitations were single-centre study design, a small number of participants, mixed aetiology of NAFLD, selection bias, lack of histological samples.
Ethics committee approval: The present study has been approved by the regional and institutional committee of science and research ethics of our University (SE-RKEB 140/2020).
Funding for this study: The authors received no specific funding.

Author Block: Q. Yu, Y. Wang, S. Ju; Nanjing/CN
Purpose or Learning Objective: Non-invasive criteria to stratify liver decompensation risk remained an unmet need in patients with compensated cirrhosis. We aimed to develop and validate a non-invasive model based on spleen volume and simple serum markers to predict decompensation in compensated cirrhotic patients.
Methods or Background: 239 compensated cirrhotic patients were enrolled from three centres in China from January 2016 to June 2020 in this retrospective study. They were followed up until the occurrence of liver decompensation. Abdominal CT and laboratory workup were collected at baseline. Spleen volume was measured automatically using an in-house AI algorithm with Dice > 0.98 in spleen segmentation. We used these data to develop a spleen volume-based model to determine the risk of decompensation in the first centre comprising 66 patients (Training cohort). We validated it in the other two centres comprising 94 and 79 patients (Test cohorts 1 and 2). And compared with Child-Pugh score, MELD, and FIB-4.
Results or Findings: 58 patients (24%) developed liver decompensation over a median follow-up of 25 months. Using a combination of spleen volume, PLT, GGT, and Hb, we developed a Spleen-Alert model (Model score > 2.14) to identify patients with compensated cirrhosis at risk of liver decompensation. HR for decompensation in patients with high risk was 13.4 in training and 6.1 and 12.2 in test, respectively. The Spleen-Alert model has good performance to predict high-risk compensated patients at risk of liver decompensation (C- indexes of 0.82 in training and 0.82, 0.77 in two test cohorts), outperforming traditional non-invasive tests (C- indexes from 0.51 to 0.74).
Conclusion: Spleen-Alert model, as a simple and non-invasive criterion, showed considerable performance in stratifying the individual risk of liver decompensation in patients with compensated cirrhosis.
Limitations: No limitations were identified.
Ethics committee approval: The ethics committee approved this study.
Funding for this study: Not applicable.

Author Block: W. Lei-di¹, P. Han², H. Meng-ting², Y. Hong-li², W. Lin-xia², X. Yue², X. Zhang³, Y. Ming²; ¹Zhongshan/CN, ²Wuhan/CN, ³Shanghai/CN
Purpose or Learning Objective: To assess the diagnostic value of spectral parameters in differentiating adrenal adenomas from metastases based on dual-layer detector spectral CT (DLCT).
Methods or Background: One hundred patients with 64 adrenal adenomas and 36 metastases were included eventually. Several spectral parameters of tumours such as the CT values of virtual non-contrast images (CTVNC), slopes of spectral HU curves (s-SHC), and iodine-to-CTVNC ratios were measured on spectral based images in each phase. Receiver operating characteristic (ROC) curves were used to compare the diagnostic values of combined or independent spectral parameters between adenomas and metastases.
Results or Findings: In the venous phase, all spectral parameters were significantly different between adenomas and metastases. Combined spectral parameters showed a better diagnostic performance in the venous phase compared with the arterial or delayed phase. In the venous phase, higher AUC was obtained for iodine-to-CTVNC value compared with the other spectral parameters in the differential diagnosis of adenomas and metastases, with diagnostic sensitivity and specificity of 90.5% and 73.8%, respectively. Higher AUC was also obtained for CTVNC value compared with the other spectral parameters in the differential diagnosis of lipid-rich adenomas and metastases, with diagnostic sensitivity and specificity of 90.9% and 81.8%, respectively. In the differential diagnosis of lipid-poor adenomas and metastases, the AUC for s-SHC value was higher than those of other parameters, with diagnostic sensitivity and specificity of 96.6% and 75.0%, respectively.
Conclusion: On DLCT images, spectral parameters in the venous phase such as iodine-to-CTVNC, CTVNC or s-SHC values have great clinical value in differential diagnosis of adrenal adenomas and metastases.
Limitations: This was a single-centre clinical trial with a relatively small sample and the metastasis group was not pathologically confirmed.
Ethics committee approval: This study was approved by IRB.
Funding for this study: There was no funding of this study.

Author Block: H. J. Lee, T. M. KIM, J. H. Kim, H. Chang, M. H. Choi, J. Y. Cho, s. y. kim; Seoul/KR
Purpose or Learning Objective: Monitoring adults with classical congenital adrenal hyperplasia (CAH) is challenging due to variability in clinical and laboratory settings. Moreover, guidelines for adrenal imaging in CAH are not yet available. We evaluated the relationship between adrenal morphology and disease control status in classical CAH.
Methods or Background: This retrospective, cross-sectional study included 90 adult CAH patients and age and sex-matched healthy controls (n=270). Adrenal volume, width, and tumour presence were assessed using abdominal computed tomography. Correlations of adrenal volume and width with hormonal status were evaluated. The diagnostic performance of adrenal volume and width of patients with CAH for identifying disease control was assessed.
Results or Findings: Adrenal morphology of CAH patients showed hypertrophy (45.6%), normal size (42.2%), and hypotrophy (12.2%). Adrenal tumours were detected in 12 patients (13.3%). The adrenal volume and width of CAH patients were significantly larger than that in the control group (18.2±12.2 vs. 7.1±2.0cc, 4.7±1.9 vs. 3.3±0.5mm, P<0.001 for both). 17α-hydroxyprogesterone (17-OHP) and androstenedione were the highest in patients with adrenal hypertrophy, followed by normal adrenal gland and adrenal hypotrophy (P<0.05 for both). The adrenal volume and width correlated positively with adrenocorticotropic hormone, 17-OHP, 11β-hydroxytestosterone, pregnenolone sulfate, and dehydroepiandrosterone sulfate in both male and female patients (r = 0.330.95, P<0.05 for all). We obtained the optimal cutoff values of 10.7cc and 4mm for classifying well-controlled patients using adrenal volume and width, respectively (AUC 0.820.88, P<0.001 for both).
Conclusion: We may use both adrenal volume and width as reliable quantitative parameters for monitoring patients with classical CAH.
Limitations: Measuring the adrenal volume is time-consuming and laborious compared with measuring the width. To apply in clinical practice, automatic or semi-automatic volumetric assessment of the adrenal gland is warranted.
Ethics committee approval: IRB approval.
Funding for this study: Not applicable.

Author Block: A. Bonanomi, P. A. Bonaffini, P. N. Franco, C. Valle, P. Marra, A. Falanga, S. Sironi; Milan/IT
Purpose or Learning Objective: To report ischaemic and haemorrhagic abdominal complications in COVID-19 patients. To correlate these complications with lung involvement, laboratory tests, comorbidities and eventual anticoagulant treatment.
Methods or Background: Thirty COVID-19 patients underwent an abdomen CECT for not lung-related clinical symptoms (i.e. abdominal pain) between March 10th and April 26th 2020, were retrospectively evaluated. Ischaemic and haemorrhagic complications were recorded. For lung involvement, a parenchymal stage (early, progressive, peak, absorption) was assigned. Blood coagulation values, eventual anticoagulant therapy, comorbidities and eventual presence of PE were assessed.
Results or Findings: Ischaemic complications had been demonstrated in 10 patients: 6/10 small bowel ischaemia (thickened, poorly-enhanced intestinal wall and/or pneumatosis intestinalis) with 1 concomitant small bowel obstruction and 1 perforation; 4/10 ischaemic colitis (layered wall and submucosal oedema). Main mesenteric vessels were patent but 1 case with superior mesenteric vein thrombosis. Two ischaemia cases also presented concurrent splenic infarctions. Bleeding complications were found in 20 patients: spontaneous haematomas in soft tissues in 15, retroperitoneal hematomas in 2 and gastro-intestinal bleeding in 3. Ten presented foci of active bleeding. Platelet and lymphocyte counts were in their normal range. D-Dimer levels demonstrated a significant difference between the two groups (p=0.005), being higher in ischaemic cases. Most of the patients had severe lung disease (45% peak, 29% absorption). Only 2 patients had evidence of PE. 22 patients (hospitalised) were under anticoagulant treatment.
Conclusion: Ischaemic and haemorrhagic abdominal complications may occur in COVID-19 patients, particularly during extended lung disease. CT plays a key role in the diagnosis of these potentially life-treating conditions.
Limitations: Retrospective nature of the study. Small population study.
Ethics committee approval: Not applicable.
Funding for this study: Not applicable.