A. Borghi

Acoustic Radiation Force Impulse Elastography for fibrosis evaluation in patients with chronic hepatitis C: An international multicenter study

AIM The aim of this international multicenter study was to evaluate the reliability of Acoustic Radiation Force Impulse (ARFI) elastography for predicting fibrosis severity, in patients with chronic hepatitis C. PATIENTS AND METHODS We compared ARFI to liver biopsy (LB) in 914 patients (10 centers, 5 countries) with chronic hepatitis C. In each patient LB (evaluated according to the METAVIR score) and ARFI measurements were performed (median of 5-10 valid measurements, expressed in meters/second - m/s). In 400 from the 914 patients, transient elastography (TE) was also performed (median of 6-10 valid measurements, expressed in kiloPascals - kPa). RESULTS Valid ARFI measurements were obtained in 911 (99.6%) of 914 cases. On LB 61 cases (6.7%) had F0, 241 (26.4%) had F1, 202 (22.1%) had F2, 187 (20.4%) had F3, and 223 (24.4%) had F4 fibrosis. A highly significant correlation (r=0.654) was found between ARFI measurements and fibrosis (p<0.0001). The predictive values of ARFI for various stages of fibrosis were: F ≥ 1 - cut-off>1.19 m/s (AUROC=0.779), F ≥ 2 - cut-off>1.33 m/s (AUROC=0.792), F ≥ 3 - cut-off>1.43 m/s (AUROC=0.829), F=4 - cut-off>1.55 m/s (AUROC=0.842). The correlation with histological fibrosis was not significantly different for TE in comparison with ARFI elastography: r=0.728 vs. 0.689, p=0.28. TE was better than ARFI for predicting the presence of liver cirrhosis (p=0.01) and fibrosis (F ≥ 1, METAVIR) (p=0.01). CONCLUSION ARFI elastography is a reliable method for predicting fibrosis severity in chronic hepatitis C patients.

Accuracy of VirtualTouch Acoustic Radiation Force Impulse (ARFI) imaging for the diagnosis of cirrhosis during liver ultrasonography.

PURPOSE VirtualTouch is a new technique recently proposed to evaluate liver stiffness during B-mode ultrasonography. The goal of the present study was to analyze the diagnostic accuracy of VirtualTouch in the diagnosis of cirrhosis and its correlation with transient elastography (Fibroscan). MATERIALS AND METHODS A total of 133 patients with chronic liver disease were enrolled. 90 of 133 underwent VirtualTouch and transient elastography and 70 patients assessed with VirtualTouch were submitted to liver biopsy. Stiffness was assessed by both techniques in the right liver lobe. The diagnostic accuracy for cirrhosis was first assessed in the 90 patients submitted to transient elastography with > 13 kPa (47 % of patients) as diagnostic for cirrhosis values. The best cut-off for cirrhosis with VirtualTouch was then tested in the 70 patients with biopsy (cirrhosis in 38 % of patients). 41 patients were assessed by VirtualTouch by two different operators. RESULTS The VirtualTouch values in controls, chronic hepatitis and cirrhosis were respectively 113, 147 and 255 cm/sec. The AUROC of liver VirtualTouch for the diagnosis of cirrhosis (reference Fibroscan) was 0.941 with 175 cm/sec as the best cut-off (sensitivity 93.0 %; specificity 85.1 %). VirtualTouch confirmed good performance also in patients with bioptic diagnosis of cirrhosis (AUROC 0.908, sensitivity 81.5 %, specificity 88.4 %,). The correlation of VirtualTouch with transient elastography was strict (r = 0.891) and the correlation in VirtualTouch measurements between two operators was also good (r = 0.874). CONCLUSION VirtualTouch is able to identify the presence of cirrhosis with good accuracy, shows good interobserver reproducibility and the correlation of its values with those obtained by transient elastography with Fibroscan is good.

Impact of gadoxetic acid (Gd‐EOB‐DTPA)‐enhanced magnetic resonance on the non‐invasive diagnosis of small hepatocellular carcinoma: a prospective study

Gadoxetic acid (Gd‐EOB‐DTPA) is a ‘hepatocyte‐specific’ contrast agent for magnetic resonance (MR) in both the vascular and the hepatobiliary phases.

Circulating microRNAs, miR-939, miR-595, miR-519d and miR-494, Identify Cirrhotic Patients with HCC

The performance of circulating biomarkers for the diagnosis of hepatocellular carcinoma (HCC) is sub-optimal. In this study we tested circulating microRNAs as biomarkers for HCC in cirrhotic patients by performing a two stage study: a discovery phase conducted by microarray and a validation phase performed by qRT-PCR in an independent series of 118 patients. Beside miRNAs emerged from the discovery phase, miR-21, miR-221, miR-519d were also tested in the validation setting on the basis of literary and tissue findings. Deregulated microRNAs were assayed in HCC-derived cells in the intracellular compartment, cell culture supernatant and exosomal fraction. Serum and tissue microRNA levels were compared in 14 patients surgically treated for HCC. From the discovery study, it emerged that seven circulating microRNAs were differentially expressed in cirrhotic patients with and without HCC. In the validation set, miR-939, miR-595 and miR-519d were shown to differentiate cirrhotic patients with and without HCC. MiR-939 and miR-595 are independent factors for HCC. ROC curves of miR-939, miR-595 and miR-519d displayed that AUC was higher than AFP. An exosomal secretion of miR-519d, miR-21, miR-221 and miR-1228 and a correlation between circulating and tissue levels of miR-519d, miR-494 and miR-21 were found in HCC patients. Therefore, we show that circulating microRNAs deserve attention as non-invasive biomarkers in the diagnostic setting of HCC and that exosomal secretion contributes to discharging a subset of microRNAs into the extracellular compartment.

Characterization of primary and recurrent nodules in liver cirrhosis using contrast-enhanced ultrasound: which vascular criteria should be adopted?

PURPOSE To assess the impact of different vascular patterns at contrast-enhanced ultrasound (CEUS) on the characterization of small liver nodules (10-30 mm) in cirrhosis and to determine whether primary nodules and recurrent nodules (after a previously treated hepatocellular carcinoma) display variations in enhancement pattern. MATERIALS AND METHODS A total of 135 cirrhotic patients were evaluated. A diagnosis of hepatocellular carcinoma (HCC) was established according to AASLD Guidelines, based on imaging (computed tomography and/or magnetic resonance) or liver biopsy. All patients underwent CEUS. Different CEUS patterns were evaluated in terms of diagnostic accuracy: HYPER-HYPO: Arterial hyperenhancement followed by washout (hypoechoic appearance compared with surrounding parenchyma) in late phase; HYPER-ISO: Arterial hyperenhancement followed by isoenhancement (isoechoic appearance) in late phase; ISO-ISO: Isoenhancement in all vascular phases. RESULTS A total of 155 consecutive primary (n = 90) or recurrent (n = 65) nodules were detected. HCC was diagnosed in 127 nodules (71 primary, 56 recurrent). A characteristic HYPER-HYPO CEUS pattern was revealed in 52/127 (40.9%) HCCs (31 primary, 21 recurrent) giving a positive predictive value (PPV) of 98% (97% primary, 100% recurrent) and an accuracy of 51% (54% primary, 46% recurrent). A HYPER-ISO pattern was noted in 46 HCCs (31 primary, 15 recurrent). Assuming this pattern to also be indicative of HCC, the PPV and accuracy were 94% (93% primary, 97% recurrent) and 77% (84% primary, 68% recurrent), respectively. An ISO-ISO pattern was present in 29 HCCs (9 primary, 20 recurrent) and 22 non-HCCs (14 primary, 8 recurrent). CONCLUSION These data confirm that the HYPER-HYPO pattern at CEUS is definitely diagnostic for HCC in cirrhosis and that the HYPER-ISO pattern has a similar PPV, indicating that this pattern is highly suspicious for HCC. The ISO-ISO pattern was found in > 50% of recurrent nodules and indicates a high risk of HCC.

Cost analysis of recall strategies for non-invasive diagnosis of small hepatocellular carcinoma

BACKGROUND Which is the least expensive recall policy for nodules in the cirrhotic liver remains unclear. AIM Aim of the study was to analyze the costs of different recall diagnostic strategies of hepatocellular carcinoma (HCC) on cirrhosis on a real series of patients. METHODS 75 consecutive small liver nodules (10-30 mm) detected at conventional ultrasonography in 60 patients with cirrhosis were submitted to contrast-enhanced ultrasound, computed tomography and gadolinium-magnetic resonance imaging with a final diagnosis established according to the latest guidelines which include different strategies for nodules 10-19 mm or > or =20mm. The actual costs required to fully characterise any nodule and staging HCC in every patient, if one or the other imaging modality had been used as the first diagnostic step, were calculated. The theoretical hypothesis that each nodule was present in each patient was also investigated from an economical point of view. RESULTS The recall strategy starting with contrast-enhanced ultrasound plus computed tomography is the least expensive strategy for patients with at least one nodule 10-19 mm in size, in nearly all combinations (single or double nodules). In patients with single 20-30 mm nodules the least expensive strategy is to start with computed tomography and to use contrast-enhanced ultrasound as a second step technique. CONCLUSIONS wider use of contrast-enhanced ultrasound has the potential to save healthcare costs after first ultrasound detection of a single small nodule in cirrhosis.

Quantification of enhancement of focal liver lesions during contrast-enhanced ultrasound (CEUS). Analysis of ten selected frames is more simple but as reliable as the analysis of the entire loop for most parameters

The aim of the study was to evaluate the reliability of the analysis of only 10 frames rather than of a whole clip in performing quantitative assessment of tumor enhancement of focal liver lesions (FLLs) following ultrasound contrast injection. Contrast-enhanced ultrasonography (CEUS) examinations of 31 FLLs (median diameter: 30mm) were performed. All clips were analyzed and quantified with an early prototype of the SonoLiver software (TomTec GmbH, Munich and Bracco Research SA, Geneva), first evaluating the entire clip then selecting only 10 frames at different time intervals. Enhancement measurements obtained from the analysis of the entire clip or of only 10 frames were closely correlated (r=0.931 and p<0.0001 for Area Under the Curve; r=0.944 and p<0.0001 for Perfusion Index). In conclusion, enhancement quantification of FLLs can be reliably obtained from only 10 frames, rather than the entire clip, at least for most parameters, making such procedure easier for potential routine use.

Relationship between hepatic haemodynamics assessed by Doppler ultrasound and liver stiffness.

AIM We tested the relationship between hepatic haemodynamics assessed by Doppler ultrasonography and liver stiffness assessed by Transient Elastography in hepatitis C related chronic liver disease. METHODS Three liver Doppler ultrasound parameters (hepatic artery resistance index, splenic artery resistance index and waveform pattern in hepatic veins) and liver stiffness measured by Transient Elastography were analysed in one hundred consecutive patients affected by hepatitis C related chronic liver disease. RESULTS Hepatic and splenic arteries resistance indexes correlate significantly (p<0.0001 for both) with liver stiffness. A hepatic artery resistance index cut-off value of 0.64 provided sensitivity and specificity respectively of 84.4% and 69.1% for predicting liver stiffness ≤or >13 kPa, whereas a splenic artery resistance index cut-off value of 0.56 provided sensitivity and specificity respectively of 81.3% and 48.5%. The coincidental finding of both resistance indexes above the respective cut-off values showed a good accuracy in identifying patients with liver stiffness values >13 kPa (accuracy=78%, +LR=2.90, -LR=0.31). A significant difference in liver stiffness values was evident between patients with triphasic and bi- or monophasic waveform pattern (p=0.005). CONCLUSIONS Hepatic and splenic arteries resistance indexes and the hepatic veins waveform pattern assessed by Doppler ultrasound may provide information similar to that of Transient Elastography in hepatitis C related chronic liver disease.

A phase I study of continuous hepatic arterial infusion of Irinotecan in patients with locally advanced hepatocellular carcinoma

PURPOSE Aim of this phase I study was to identify the maximum tolerated dose and dose limiting toxicity of continuous infusion of Irinotecan through a port-a-cath placed in the hepatic artery in patients with hepatocellular carcinoma and cirrhosis to explore new strategies in advanced hepatocellular carcinoma. Response rate and time-to-progression were analysed. METHODS Irinotecan was delivered as a five-day continuous infusion every 21 days, with increases of 2.5mg/m(2)/day every three patients, starting from 7.5mg/m(2)/day. Dose limiting toxicity corresponded to one patient in each triplet developing G4 haematological or G3 non-haematological toxicity, confirmed in two triplets. Twenty-eight patients (17 Child-Pugh A, 11 B) received treatment and tumour response was assessed after three courses completed by 22 patients. RESULTS Dose limiting toxicity was G3 diarrhoea in two patients, reached at 27.5mg/m(2)/day and the recommended dose was set at 25mg/m(2)/day. Nineteen of 30 patients experienced adverse events related to porth-a-cath placement and one died from liver ischemia and sepsis. Median time-to-progression was 11.3 months. CONCLUSION Intrarterial infusion of Irinotecan is feasible in patients with hepatocellular carcinoma on cirrhosis at a recommended dose of 25mg/m(2)/day, with no major adverse drug-related events, but with some concerns about the insertion and management of the intra-arterial device.

79 INFLUENCE OF AMINOTRANSFERASES LEVEL ON THE CORRELATION OF LIVER STIFFNESS ASSESSED BY ACOUSTIC RADIATION FORCE IMPULSE (ARFI) ELASTOGRAPHY WITH LIVER FIBROSIS-AN INTERNATIONAL MULTICENTER STUDY

79 INFLUENCE OF AMINOTRANSFERASES LEVEL ON THE CORRELATION OF LIVER STIFFNESS ASSESSED BY ACOUSTIC RADIATION FORCE IMPULSE (ARFI) ELASTOGRAPHY WITH LIVER FIBROSIS-AN INTERNATIONAL MULTICENTER STUDY S. Bota, I. Sporea, R. Sirli, H. Tanaka, H. Iijima, R. Badea, M. Lupsor, C. Fierbinteanu-Braticevici, A. Petrisor, H. Saito, H. Ebinuma, M. Friedrich-Rust, C. Sarrazin, H. Takahashi, N. Ono, F. Piscaglia, A. Borghi, M. D’Onofrio, A. Gallotti, M. Peck-Radosavljevic, A. Ferlitsch, A. Popescu, M. Danila. Gastroenterology and Hepatology, University of Medicine and Pharmacy, Timisoara, Romania; 2Division of Hepatobiliary and Pancreatic Disease, Department of Internal Medicine, Hyogo College of Medicine, Nishinomiya, Japan, IIIrd Medical Clinic, University of Medicine, Cluj-Napoca, IInd Medical Clinic and Gastroenterology, University Hospital, Bucharest, Romania; Department of Internal Medicine, Keio University, Tokyo, Japan; Department of Internal Medicine 1, J.W. Goethe University, Frankfurt/Main, Germany; Department of Internal Medicine, Gastroenterology, Saga Medical School, Saga, Japan; Div. Internal Medicine, Dept. Clinical Medicine, University and General Hospital S. Orsola-Malpighi, Bologna, Department of Radiology, University Hospital G.B. Rossi, University of Verona, Verona, Italy; Internal Medicine III, Div. of Gastroenterology and Hepatology, Medical University of Vienna, Vienna, Austria E-mail: bota_simona1982@yahoo.com