Abstracts of the A.I.S.F. Annual Meeting 2010 / Digestive and Liver Disease 43S (2011), S65–S108 S93 F-2 Pattern of drug-induced liver injury: The impact of amoxicillin-clavulanate on clinical course A. Licata, C. Randazzo, G. Butera, V. Calvaruso, A. Calì, M. Cappello, A. Craxì, P.L. Almasio Sezione di Gastroenterologia, Di.Bi.M.I.S. Università di Palermo. Background and aim: Drug-induced liver injury (DILI) is increasingly rec- ognized as a cause of liver disease. We aimed to evaluate the rate of severe cases of DILI on total hospital admissions and to identify drugs commonly responsible. Methods: Between 1996 and 2010, all records of consecutive patients diag- nosed as DILI at our Unit were analyzed. Data were collected from in- and outpatient visit charts. Diagnosis of DILI was based on the presence of Inter- national Consensus. Liver damage was defined as hepatocellular, cholestatic or mixed, according to clinical and laboratory data. All patients had regular follow-up visits every three months to update outcomes. Results: Out of 9,372 patients, 115 (age 52.6, range 11-88, 68% were older than 40 yrs) fulfilled the diagnostic criteria of DILI. Twenty five patients (21%) had pre-existing compensated chronic liver disease (CLD). Hepatocellular pat- tern was more commonly observed (47%) followed by cholestatic (30%) and mixed pattern (23%). Pre-existing CLD was associated with the hepatocellular pattern of injury (P 0.004). The most frequent drugs were non-steroidal anti-inflammatory drugs (NSAIDs; 30%), followed by antibiotics (20%), im- munosuppressant (13%), anti-platelet agents (11%), anti-diabetic drugs and statins (15%). In 24 (21%) of cases, two or more drugs were involved. Among antibiotics, Amoxicillin-Clavulanate was the most frequently associated with hepatocellular pattern (P 0.037) and jaundice. Liver stiffness measurement by transient elastography (Fibroscan ® ) was carried in 38 (33%) patients. Antibi- otics group showed higher liver stiffness values as compared to NSAIDs group (17.5 Kpa vs 15.7 Kpa respectively) (P 0.004). Acute liver failure developed in 4 (3.4%) and 2 died (1.7%). Overall, complete recovery occurred in 54.8% of patients, with no difference between patients with and without CLD. Conclusions: DILI is a relevant cause of hospitalization for liver disease. Amoxicillin-Clavulanate was more frequently associated with severe liver damage as showed by highest liver stiffness values. F-3 Characterization, back-up and follow-up of occult hepatitis B infection (OBI) among blood donors B. Foglieni, M. Spreafico, A. Berzuini, L. Raffaele, I. Guarnori, D. Prati Department of Transfusion Medicine and Haematology, Alessandro Manzoni Hospital, Lecco, Italy Background: OBI is characterized by the presence of low HBV DNA level in the liver and/or serum of HBsAg negative individuals with or without serological markers of previous infection. OBI can be potentially transmitted by blood transfusion; thus OBI identification among blood donors is needed to further reduce the risk of transfusion-transmitted infections. Since April 2008, 17 OBI carriers out of 12000 screened donors were identified in our centre by the use of the NAT Roche System. Aim: To characterize OBI carriers both serologically and molecularly and to monitor these markers in available plasma samples collected before OBI identification and during a follow-up period. Methods: According to Taormina criteria, NAT positive samples were tested by quantitative real-time PCR (qPCR) assays designed on HBV-DNA Surface and Core regions, and quantified by the COBAS TaqMan, both on neat serum samples and after HBV-DNA enrichment (6mL ultracentrifuged:28.000g, 3h,4°C). Pre-transfusion and follow-up samples were tested individually by either the commercial and/or the home-made qPCR techniques. Results: 29% of patients were HBcAb+, 12% HBsAb+ and 53% were both HBcAb+ and HBsAb+. Only one OBI was negative for all HBV serological markers. Serological markers did not change during a one to 38 months follow-up (median: 7). HBV-DNA enrichment improved virus detection, in- creasing HBV identification from 57% to 81% by using the commercial assay and to 100% of samples by the home-made assays. In 8/13 OBI patients (62%), back-up samples gave positive results when individually tested by more sensitive assays. The follow-up revealed a fluctuation of viral loads. All OBI patients except from one had at least one positive HBV detection. On a median of three follow-up samples per patient, 85% of them was positive (range: 25–100%). Conclusions: HBV-DNA enrichment is of great importance to detect viremia and to confirm the diagnosis of OBI carriers, even when viral load falls during time. Analysis of back-up and long-term follow-up samples is important to exclude recent infections and false-positive results and to better characterize OBI. F-4 Hemodynamic response to beta-blockers for portal hypertension: A single center experience C. Tiani, A. Berzigotti, L. Vizioli, V. Zorzi, D. Magalotti, M. Zoli Dipartimento di Medicina Interna, dell’Invecchiamento e Malattie Nefrologiche. Università di Bologna Background: Portal hypertension is a common clinical syndrome which almost invariably complicates liver cirrhosis. The most important and direct consequence of portal hypertension is the development of gastro-oesophageal varices. Currently beta-blockers are the best therapy in patients with varices at risk of bleeding and the gold standard to prove their efficacy is HVPG measurement. A decrease in HVPG <12 mmHg or ≥20% from baseline significantly reduces the risk of first variceal bleeding, of other complications of cirrhosis and death, but there is scarce information on whether this remains unchanged over the treatment period. Patients and methods: We evaluated 57 patients with cirrhosis and gastro- oesophageal varices at risk of first variceal bleeding, in whom HVPG was measured prior to beta-blockers treatment. After baseline HVPG all beta- blockers were started and titrated up to the maximal tolerated dose. 54 patients were submited to a second HVPG measurement at 1-3 months and were followed-up for a median of 32 months. Patients showing a HVPG decrease ≥20% on second measurement were considered Responders. We evaluated the incidence of complications and survival both in responders and non-responders on follow-up. In a subgroup of 16 Responders we assessed the modification of HVPG at long term (18 and 36 months). Results: 41% of patients were responders and 59% non-responders (NR). As compared to NR, responders had a lower incidence of complications of portal hypertension (18% vs. 25% in NR) and an increased survival (71% vs. 59% in NR, p=0.04). 77% of responders assessed at 18 and 36 months lost the hemodynamic response. These patients had an elevated incidence of clinical events, death and liver transplantation. Conclusions: Our experience confirms that NR to beta-blockers are at higher risk of complications of cirrhosis and death as compared to hemodynamic Responders. In the small subgroup of Responders evaluated long-term, most Responders lost the hemodynamic response over follow-up, and behaved as NR after loosing the hemodynamic response. F-5 Hepatocellular carcinoma (HCC) in cirrhosis: Long-term results of percutaneous radiofrequency ablation of both the nodule and the portal venous tumor thrombus A. Giorgio, G. de Stefano, A. Di Sarno, N. Farella, U. Scognamiglio, A. De Rogatis, E. Trapanese, V. Giorgio Infectious Disease and Interventional Ultrasound Unit, “D. Cotugno” Hospital, Naples, Italy Aim: To report 5-year survival of percutaneous radiofrequency ablation (RFA) of both medium-sized Hepatocellular carcinoma (HCC) accompanied by portal venous tumor thrombus (PVTT) in cirrhotic patients. Patients and Methods: From January 2005 and January 2010 among 2847 consecutively patients with HCC seen at our Institution, 672 had HCC and PVTT; 57 of the 672 had a single HCC accompanied by main portal vein tumor thrombus (MPVTT). Thirty-five patients (mean age 73 years) with