Controlled attenuation parameter predicts steatosis in alcoholic liver disease and correlates with poor metabolic phenotype and cardiovascular risk: a biopsy-controlled multicenter study

POSTER PRESENTATIONS SAT-475 Controlled attenuation parameter predicts steatosis in alcoholic liver disease and correlates with poor metabolic phenotype and cardiovascular risk: a biopsy-controlled multicenter study M. Thiele1,2, V. Rausch3, G. Fluhr3, F. Piecha3, J. Mueller3, B.S. Madsen1,2, B. Straub4, S. Detlefsen2,5, M. Lupsor-Platon6, V. de Ledinghen7, H.K. Seitz3, A. Krag1,2, S. Mueller3. 1Dpt. of Gastroenterology and Hepatology, Odense University Hospital; 2Institute of Clinical Research, University of Southern Denmark, Odense, Denmark; 3 Salem Medical Center and Center for Alcohol Research, University of Heidelberg, Heidelberg; 4Institute for Pathology, University of Mainz, Mainz, Germany; 5Dpt. of Pathology, Odense University Hospital, Odense, Denmark; 6Dpt. of Ultrasonography, University of Medicine and Pharmacy, Regional Institute of Gastroenterology and Hepatology, ClujNapoca, Romania; 7Centre d’investigation de la fibrose hépatique, Hopital Haut-Leveque, Pessac cedex, France E-mail: [email protected] Background and Aims: Fatty liver is a risk factor for liver fibrosis and cardiovascular disease (CVD). By definition, non-alcoholic and alcoholic fatty liver disease are mutually exclusive, but the metabolic syndrome (MetS) is common in alcoholic patients, where it may aggravate steatosis and act in synergy with alcohol as a “second hit”. Liver steatosis assessment and metabolic profiling in alcoholic patients should therefore be performed. We aimed to determine the value of controlled attenuation parameter (CAP) for diagnosis of steatosis ≥S2 in patients with an excessive alcohol overuse. We furthermore investigated the correlation between CAP, MetS and CVD, and the response of CAP to alcohol detoxification. Methods: We recruited 225 alcoholic patients from four European centres with concomitant liver biopsy, metabolic profiling and measurement of CAP and liver stiffness (FibroScan Echosens, France). Another 340 patients referred for detoxification had CAP measurements at baseline and after abstinence. We used International Diabetes Federation criteria to define MetS. The Framingham risk score denoted 10-year risk of CVD. Liver histology was assessed according to Kleiner for steatosis (S0-3) and fibrosis (F0-4). Bright liver echo pattern (BLEP) suggested steatosis on ultrasound. We used literature based cut-off values for CAP of 250, 270 and 300 dB/m2. Results: Eighty of the 225 patients (36%) had S2 or S3 steatosis; 90 (40%) had advanced liver fibrosis. CAP was fairly good to diagnose ≥S2 steatosis (AUC 0.78, 0.72–0.84), with a better performance than BLEP (AUC 0.65, AUC comparison P = 0.001). CAP below 250 dB/m2 ruled out S2 steatosis (sens 89%, NPV 88%), while CAP above 300 dB/m2 ruled in ≥S2 steatosis (spec 81%, PPV 61%). 270 dB/m2 was the best overall cut-off (sens 80%, spec 60%, NPV 85%, PPV 53%, 67% correct classifications). In multivariable regression, liver steatosis, MetS and 10-year CVD risk were strong predictors of CAP, rather than amount of alcohol intake. Irrespective of histological steatosis stage, patients with MetS on average had 40 dB/m2 higher CAP than patients without MetS. Every 5% increase in CVD risk corresponded to a 4 dB/ m2 CAP increase. In 340 patients admitted to detox, CAP decreased from 291 to 258 dB/m2 during 6 ± 2 days (P < 0.001). Conclusions: Controlled attenuation parameter is an important tool for alcoholic fatty liver disease. Alcoholic patients with high CAP after short-term detoxification have a poor metabolic profile and a high risk of cardiovascular disease. SAT-476 Transient elastography correlates with liver fibrosis and markers of portal hypertension in PSC: laennec score-based analysis of explanted livers M. Krawczyk1,2, J. Ligocka3, M. Ligocki4, M. Zawadzki3, M. Milkiewicz5, G. Szparecki6, B. Górnicka6, M. Krawczyk3, F. Lammert7, P. Milkiewicz4,8. 1Department of Medicine II, Saarland University Medical Center, Saarland University Medical Center, Homburg, Germany; 2Laboratory of Metabolic Liver Diseases, Department of General, Transplant and Liver Surgery; 3Department of General,

Transplant and Liver Surgery; 4Liver and Internal Medicine Unit, Department of General, Transplant and Liver Surgery, Medical University of Warsaw, Warsaw; 5Department of Medical Biology, Pomeranian Medical University, Szczecin; 6Department of Pathology, Medical University of Warsaw, Warsaw, Poland; 7Department of Medicine II, Saarland University Medical Center, Homburg, Germany; 8Translational Medicine Group, Pomeranian Medical University, Szczecin, Poland E-mail: [email protected] Background and Aims: Previous studies demonstrated a close correlation between transient elastography (TE) and liver biopsy, especially in patients with chronic viral hepatitis. However, biopsy restricts the analysis to only a minimal part of the liver parenchyma. Here we investigated the association between TE, serum markers of portal hypertension (sCD163, VCAM-1) and histology of explanted livers in patients with primary sclerosing cholangitis (PSC). Methods: Prospectively we recruited 30 patients with PSC (12 women, age 35.6 ± 12.2 years) who were transplanted at our centre. TE (Fibroscan) examination and blood sampling was performed during evaluation for liver transplantation (LT); a second blood sample was taken at the time of LT. Explanted livers were assessed according to the recently proposed, extended Laennec staging system, which comprises seven stages of liver fibrosis and measures mean thickness of fibrotic septa (Kim/Park J Hepatol 2012). Liver scarring was also quantified using measurement of collagen contents. Serum markers of portal hypertensions (sCD163 and VCAM1) were measured using ELISA. Results: In total, 54% of patients presented with cirrhosis on histology. The mean septum thickness was 267.2 ± 137.4 μm and the thickest septum diameter was 458.8 ± 237.0 μm. The mean liver stiffness was 21.4 ± 18.2 kPa, and >50% patients presented with at least 15 kPa. TE showed significant correlations with the stages of fibrosis in explanted livers as assessed with Laennec (and METAVIR) scores (p = 0.001 and p = 0.006, respectively). TE correlated significantly with mean septa diameter ( p = 0.022) and with the diameter of thickest septa ( p = 0.045) as well as with collagen contents ( p < 0.001) in explanted livers. Overall, patients with cirrhosis at histology had significantly ( p = 0.002) increased mean liver stiffness as compared to non-cirrhotic livers (28.56 ± 19.44 vs. 9.00 ± 3.81 kPa). Finally, liver stiffness correlated with serum markers of portal hypertension, i.e. CD163 and VCAM1, assessed both at the time of inclusion ( p = 0.002 and p = 0.028, respectively) and at the time of LT ( p = 0.004 and p = 0.031, respectively). Conclusions: We demonstrate for the first time that in patients with PSC transient elastography correlates with liver fibrosis quantified in explanted livers using the Laennec staging system. Moreover, it correlates with serum surrogate markers of portal hypertension. Hence, we postulate that elastography is a reliable tool for noninvasive monitoring of PSC patients awaiting liver transplantation. SAT-477 Effect of meal ingestion on liver stiffness and controlled attenuation parameter M. Silva1, H. Cardoso1, A. Peixoto1, S. Lopes1, R. Gonçalves1, S. Rodrigues1, A. Albuquerque1, P. Pereira1, G. Macedo1. 1 Gastroenterology, Centro hospitalar São João, Porto, Portugal E-mail: [email protected] Background and Aims: Despite the increasing use of noninvasive methods for the assessment of liver fibrosis and steatosis, it is not yet clear the effect of fasting and food intake on these parameters. Our aims were to evaluate the effect of food intake on LS (measured by transient elastography) and CAP in patients with different degrees of liver disease and healthy volunteers, and secondarily, to assess possible anthropometric, clinical and biological parameters associated with variations of LS and CAP. Methods: Prospective single-center study including patients with liver disease and healthy volunteers. LS and CAP were evaluated using FibroScan® (Echosens, Paris, France) before (fasting ≥8 h) and 30 minutes after intake of a standardized breakfast. Anthropometry,

Journal of Hepatology 2017 vol. 66 | S543–S750

S665