Posters
82
by regatlating insulin secretion. However, the role of GLP-1 for HCVinduced insulin resistance remains unclear. Aim: The aim of this study is to examine an association between HCVinduced insulin resistance and GLP-1. Patients and Methods: Fifty-four Japanese males with HCV- or HBVinduced liver disease (HCV group; n = 3 6 , HBV group; n = 19) were enrolled. Blood samples were collected from each subject after overnight fasting and levels of glucose, immunoreactive insulin (IRI), and liver function tests were routinely measured. Subjects' insulin resistance levels were determined by homeostasis model assessment (HONDk-IR). GLP-1 was measured by ELISA. All data are expressed as mean:t:SE. M a n n Whitney U test and Spearn~an correlation test were used for statistical analysis. P values <0.05 were considered significant. Results: There were no significant differences in age, platelet count, and levels of AST, ALT, albumin, and total bilirubin between HCV and HBV groups. HCV group showed a significant increase in HOMA-IR levels compared to those of HBV group (3.9:t=1.5 vs. 1.8-t:0.3; p <0.05). GLP-1 levels were significantly increased in HCV g-roup compared to those of HBV group (6.94-0.8 vs. 4.74-0.4; p <0.05). In HBV group, a correlation between GLP-1 levels and IRI levels was not seen. On the other hand, in HCV group, there was a si~tificant correlation between GLP-1 levels and IRI levels (p < 0.05). In addition, HOMA-IR levels were significantly elevated in subjects with increased GLP-1 levels (3.14-0.5 vs. 2.14-0.4; p < 0.05) in HCV group. GLP-1 levels were not significantly associated with both albumin levels and platelet count in HCV group. Conclusions: We first demonstrated that 1) increased GLP-1 levels were seen in HCV group, but not in HBV group. 2) a significant association between GLP-1 and HOMA-IR levels in HCV group. Therefore, GLP-1 may play a crucial role for HCV-induced insulin resistance.
[
•
EFFECTS OF SOMATOSTATIN, TERLIPRESSIN AND SOMATOSTATIN PLUS TERLIPRESSIN ON PORTAL AND SYSTEMIC HEMODYNAMICS AND RENAL SODIUM EXCRETION IN PATIENTS WITH CIRRHOSIS
G. Kalambolds 1, M. Economou 2, P. Kosta 3, K. Papadimitriou 3, C.H. Pappas 4, A. Katsarald 5, E.V Tsianos 1'2. ;Department of Internal
Medicine, Medical School of Ioannina, Ioannina, Greece," 2Department of Gastroenterology, Medical School of Ioannina, Ioannina, Greece," 3Departrnent of Radiology, Medical School of Ioannina, Ioannina, Greece," 4Department of Cardiology. Medical School of Ioannina, Ioannina, Greece," 5Department of Statistics, Medical School of Ioannina, Ioannina, Greece Background and Aims: Terlipressin and somatostatin are the most preferable agents for the control of variceal bleeding in cirrhotic patients. This study evaluated the hemodynamic effects of somatostatin, terlipressin, and somatostatin plus terlipressin in cirrhotic patients with portal hypertension and the effect of each treamlent on renal sodium excretion. Methods: Twenty-four patients with esophageal vafices were randomly assigned to receive either an intsavenous infusion of placebo (n= 12) or somatostatin 250btg/h after an initial bolus of 250btg (n=12) for 60rain. Thereafter, each patient received an intravenous injection of terlipressin 2 mg while the infusion of either somatostatin or placebo was maintained. Portal and systemic hemodynamic parameters, assessed by Doppler sonography, and urinary sodium excretion were evaluated at baseline, 60 min after placebo or somatostatin, and 30rain after terlipressin. Results: Data is expressed as medians. Placebo had no effect on the patients studied. Terlipressin alone significantly reduced portal vein velocity (0.09 vs. 0.15rrds, p=0.01) and portal flow volume (0.56 vs. 1 l/rain, p = 0.007) and increased mean arterial pressure (MAP; 103.3 vs. 89.9mmHg, p = 0.02) and systemic vascular resistance (1541 vs. ll08dyn.s/cm 5, p=0.008) together with a decrease in cardiac output (CO; 6.4 vs. 7.61/min, p=0.007). Fractional sodium excretion (FeNa) significantly increased in patimlts without ascites (0.43 vs 0.16%, p = 0.02)
and did not change in patients with ascites. Somatostatin did not alter portal hemodynamics whereas it significantly reduced MAP (86.9 vs. 98.6mmHg, p=0.03), heart rate (65 vs. 73bpm, p=0.03), CO (8.4 vs. 9.11/min, p=0.02) and, in patients with ascites, FeNa (0.13 vs. 0.23%, p = 0.03). The addition of terlipressin induced similar changes in hemodynamic parameters and sodium excretion to those observed after terlipressin alone. The mag-nitude of increase in MAP was sig-nificanfly higher in patients receiving terlipressin alone than in those receiving somatestatin plus terlipressin (15 vs. 5.3 %, p = 0.004) while CO was conversely affected (-28.5 vs. -20.9%, p=0.01). Conclusions: A combination of somatostatin and terlipressin does not exert an additive portal hypotensive effect in drrhotic patients compared with terlipressin alone. In contrast to somatostatin, terlipressin alone or added to somatostatin ameliorates systemic hernodynarnics and, in patients with ascites, improves renal sodium excretion.
~IN
PATIENTS WITH CIRRHOSIS, LIVER STIFFNESS MEASUREMENT (LSM) PREDICTS THE PRESENCE AND SIZE OF OESOPHAGEAL VARICES (OV) MORE ACCURATELY THAN PORTAL GRADIENT (PG)
F. Kazemi, E. Pinto, O. N'Kontchou, N. Oanne-Carrit, J.C. Tfinchet, M. Beaugrand. Service d'H@atologie, H@ital Jean Verdier, Bondy,
France Background and Aim: LSM is a new method reflecting of liver fibrosis. It has been suggested that in patients with cirrhosis results correlate with the severity of liver disease and predicts the presence and size of OV We studied the relation between LSM and parameters reflecting the severity of liver disease and compared tile value of LSM for predicting tile presence and size o f OV to the one of PG. Methods: One hundred and sixty six patients (aged 56± 13, 112 men) with histologically proven cirrhosis (alcoholic n = 35, HCV n = 98, other causes n 33) had simultaneously upper tract endoscopy and LSM, measured by Fibroscan®, a device measuring liver elasticity (stiffness) by transient elastography, from the velocity of a shock wave inside the fight liver. Seventy patients had a measurement of wedged hepatic pressure and portal gradient. Results: LSM was correlated by Spearman corrdation test with ChildPugh stage (r=0.42), presence of oesophageal varices (r=0.6), grade of varices (r 0.6), wedged hepatic pressure (r 0.64) and portal gradient (r=0.64), p < 10-4 for each. In patients with haemodynamic measurement LSM had a better predictive value for the presence of OV than PG (AUROC 0.89 for LSM, AUROC 0.78 for PG). The cut off value of 23 kPa determined by the ROC curves allows to split the population in two groups. LSM value <23kPa was highly predictive of the absence of OV grade II or III (PPV: 42%, NPV: 95%). False negative results were encountered in patients with macronodular cirrhosis or patients previously treated by Interferon. Conelusion: LSM is correlated to the severity of liver disease and predicts more accurately than PG the presence and size of oesophageal vatices.