05d: VIRAL HEPATITIS − d) HEPATITIS B − CLINICAL (EXCEPT THERAPY) S143 under the ROC (AUROC) curve of the FIB-4 index at BL to exclude F4-F6 was 0.81 (95% CI 0.76−0.87; P < 0.001). The AUROC curve of the FIB-4 index at BL to exclude F3-F6 was 0.71 (95% CI 0.66−0.76; P < 0.001). On the follow-up liver biopsy, fibrosis decreased, plateau or increased in 27.2, 56.8 and 16% of patients, respectively. The variation of the FIB-4 index correlated with the variation of fibrosis (Spearman’s rho 0.22; P < 0.001). The FIB-4 index decreased of (mean ± SD) 0.36±0.68 units (P < 0.001), decreased of 0.14±0.85 units (P = 0.001) and increased of 0.14±1.3 (P = 0.14) when fibrosis decreased, remained stable or progressed on the follow-up liver biopsy. The FIB-4 index decreased of 0.28±0.8 units (P < 0.001) in patients treated by ADV and remained stable (P= 0.453) in control patients. The Ishak score decreased of 0.33±0.8 units (P < 0.001) in patients treated by ADV and increased of 0.1±0.9 units (P = 0.08) in controls and 72.2% of patients under ADV with a FIB-4 index  1.45 had a concordant liver biopsy. Conclusion: The FIB-4 index is a helpful tool to assess fibrosis before and during treatment of CHB. 375 HEPATITIS B SURFACE ANTIGEN CONCENTRATION IN HEPATITIS B/C CO-INFECTION A. Potthoff 1 , J. Jaroszewicz 1 , B. Calle Serrano 1 , R. Rauppach 1 , K. Wursthorn 1 , R. Flisiak 2 , M.P. Manns 1 , H.L. Tillmann 3 , M. Cornberg 1 , H. Wedemeyer 1 . 1 Gastroenterology, Hepatology und Endocrinology, Hannover Medical School, Hannover, Germany; 2 Department of Infectious Diseases and Hepatology, Medical University of Bialystok, Bialystok, Poland; 3 Duke Clinical Research Institute, Duke University Medical Center, Durham, NC, USA E-mail: potthoff.andrej@mh-hannover.de Background: Different patterns of viral dominance have been described in patients with HBV/HCV co-infection. In the majority of cases HCV is suppressing HBV-DNA. However, it is unclear, if also HBsAg levels are altered in HBV/HCV co-infected individuals. The aim of this study was to investigate quantification of HBsAg in a defined cohort of HBV/HCV co-infected patients with well documented virological patterns and clinical characteristics. Methods: Seventy patients with HBV/HCV co-infection (male/female: 47/23; mean age: 41±15 years) (80% HBeAg negative) were studied. HCV was dominant in 53% (HBV-DNA−/HCV-RNA+, group 1), HBV was dominant in 16% of patients (HBV-DNA+/HCV-RNA−, group 2), both HBV-DNA and HCV-RNA were detectable in 19% (group 3) and both nucleic acids were negative in the remaining 12% of patients (group 4). 76 HBV monoinfected patients (age and sex matched) were used as control group (75% HBeAg negative). Results: HBsAg levels and HBV-DNA were significant lower in HBV/HCV co-infection than in HBV monoinfection (mean HBsAg 9184±3457 IU/mL and 16832±4612 IU/mL, p < 0.008; mean HBV-DNA 7.6×10 6 IU/mL and 9.4×10 6 IU/mL, p < 0.001). HBV-DNA and HBsAg levels showed a good correlation in the HBV monoinfection control group (R = 0.42, p < 0.001), while this was not the case in HBV/HCV co-infected patients (R = 0.19, p = 0.11). HBsAg levels were significantly lower in group 1 than in the other groups (median HBsAg levels 649 IU/mL, 4886 IU/mL, 9002 IU/mL, 5995 IU/mL in groups 1−4, re- spectively, p < 0.03). The mean HBsAg concentration in HBeAg(+) pa- tients was higher than in those with HBeAg(-) (29180±1606 IU/mL vs. 4184±1103 IU/mL, p < 0.003). Overall, the HBsAg concentration showed a borderline association with age (R = −0.23, P = 0.04). However, we found no correlation between HBsAg levels and gender, ALT, AST, platelets, INR and a-fetoprotein. Conclusion: To our knowledge, this is the first study showing that HCV dominance, but not other HBV-DNA/HCV-RNA patterns is asso- ciated with suppressed HBsAg production in HBV/HCV co-infection. Importantly, this inhibition seems to be independent from HBV-DNA suppression. Future prospective studies need to evaluate the role of HBsAg in this special cohort. 376 EFFICACY OF HIGHLY ACTIVE ANTI-RETROVIRAL THERAPY (HAART) IN PATIENTS WITH HBV-HIV CO-INFECTION L. Kosi 1 , T. Reiberger 1 , K. Rutter 1 , K. Pfistershammer 2 , A. Rieger 2 , M. Peck-Radosavljevic 1 . 1 Gastroenterology & Hepatology, 2 Dermatology, Division of Infectious Diseases, Medical University of Vienna, Vienna, Austria E-mail: thomas.reiberger@meduniwien.ac.at Background: Due to overlapping routes of transmission co-infection with hepatitis B virus (HBV) can be found in approximately 10% of HIV pa- tients. The efficacy of lamivudine (3TC), tenofovir (TDF) or emtricitabine (FTC) used for highly active antiretroviral therapy (HAART) have not been completely evaluated in HBV−HIV co-infected patients. Data about the incidence of hepatotoxicity of HAART in HBV−HIV co-infection is limited. Methods: Clinical and laboratory data of all HBV−HIV co-infected patients treated at the HIV outpatient clinic at the Medical University of Vienna between 1998 and 2008 were retrospectively analysed. Results: A total of 100 patients with HBV−HIV co-infection were identi- fied with 82% receiving HAART and 64% being HBV e-Antigen (HBeAg) positive. 3TC, TDF and FTC were included in 46%, 63% and 31% of used HAART regimens, respectively. Mean HBV viral load was 4.47±2.7 x10 9 IU/mL before initiation of HAART. HBeAg-positive patients had significantly higher HBV-DNA levels (7.47±5.06 x10 9 IU/mL) compared to HBeAg-negative patients (2.12±2.10 x10 9 IU/mL; p < 0.001). Over a median observation period of 68 month (range: 2–171) 73% achieved a complete suppression of HBV replication (lower limit of detection: 351 IU/mL). HBeAg seroconversion was found in 63% patients with a cumulative annual probability of 11.3%. Cumulative annual HBV surface- antigen (HBsAg) loss was 6.6% in HBeAg-positive patients and 8.7% in HBeAg-negative patients. Transient elevation of aminotransferases after initiation of HAART was found in 12% without a single case of hepato- toxicity grade III or IV. Conclusion: HAART treatment in HBV−HIV co-infected patients is well tolerated and leads to rapid suppression of HBV replication despite high baseline viremia. Rates of HBeAg seroconversion and HBsAg loss in HBV−HIV co-infected patients are comparable to or even higher than in HBV mono-infected patients. 377 LIVER STIFFNES EVALUATED THROUGH TRANSIENT ELASTOGRAPHY IN PATIENTS CHRONICALY INFECTED WITH HBV I. Sporea 1 , R. Sirli 1 , A. Deleanu 1 , A. Tudora 1 , S. Bota 1 , M. Cornianu 2 . 1 Gastroenterology and Hepatology, 2 Pathology, University of Medicine and Pharmacy, Timisoara, Romania E-mail: roxanasirli@gmail.com Background and Aim: Chronic viral hepatitis can be evaluated through invasive (liver biopsy-LB) or non-invasive (FibroScan or FibroTest) meth- ods. The aim of this paper was to assess the values of liver stiffness (LS) in patients with B chronic hepatitis, to compare them with those in patients with chronic C hepatitis, according to the degree of fibrosis, and to evaluate the LS in inactive HBV carriers. Material and Method: our study included 63 patients with chronic B hepatitis and 191 patients with chronic C hepatitis, in which LS was mea- sured and LB were performed in the same session. Also, LS was evaluated in 120 inactive HBsAg carriers. The LB was performed using Menghini type modified needles and the specimens were examined according to the METAVIR score. LS was measured by means of FibroScan (Echosens, France). Results: The mean values of LS in the two categories of patients are presented in table 1.