Posters – Monday May 22, 2006 8:00AM Hepatitis C

Complications of Liver Disease

Abstract M1214 – Nonabsorbed (< 0.4%) Antibiotic Rifaximin Improves Hepatic Encephalopathy Symptoms in Patients With Cirrhosis Due to Hepatitis C

M. Palmer

Background:

Lactulose, administered for the treatment of hepatic encephalopathy (HE), is often poorly tolerated. Data suggest antibiotics may be a useful treatment option for HE. Oral rifaximin (Xifaxan^®, Salix Pharmaceuticals, Inc., Morrisville, NC) is a gut-selective, nonabsorbed (<0.4%) antibiotic with broad-spectrum activity in vitro. We prospectively investigated whether rifaximin in an outpatient setting would be effective and well tolerated for the treatment of stage 1 HE in patients with cirrhosis due to hepatitis C virus (HCV).

Methods:

Consecutive patients with cirrhosis due to HCV diagnosed with stage 1 HE (determined by West Haven criteria) were treated with rifaximin 400 mg 3 times daily for 14 days. All patients were assessed 24 hours prior to the start of therapy and 14 days after completion of therapy for multiple parameters, including ability to perform mental tasks, asterixis, and a quality of life (QoL) composite score (i.e., altered sleep patterns, personality change, short-term memory loss, attention span, reaction time, slow/slurred speech, and general QoL).

Results:

A total of 37 consecutive patients were enrolled and no patient was receiving therapy for HE at study entry. Twenty-three patients were receiving pegylated interferon (IFN) plus ribavirin for chronic HCV, and 17 of these 23 patients were receiving a selective serotonin reuptake inhibitor for mild IFN-induced depression. Type 2 diabetes mellitus (DM) was reported in 12 patients. Rifaximin treatment lowered serum ammonia to normal levels in all patients, and overall, HE symptoms improved. Rifaximin was also well tolerated, with all patients completing treatment. One patient missed 2 rifaximin doses on day 9 due to complaints of flatulence and abdominal bloating, which resolved without medical intervention. There was a low incidence of adverse events (AEs): 3 patients (8%) reported mild bloating, and there was 1 report each of mild abdominal pain, diarrhea, nausea, and headache. No hypo- or hyperglycemic episodes were reported during the study.

Conclusion:

This pilot study indicated that rifaximin is well tolerated and effective in improving symptoms of stage 1 HE in patients with cirrhosis due to HCV. In addition, no glycemic episodes were reported in patients with DM. Given that neomycin is associated with a risk for nephrotoxicity and lactulose administration in patients with DM is not ideal, rifaximin may be the more appropriate treatment for patients with DM. In addition, symptoms of HE can be mistaken for pegylated interferon and/or ribavirin AEs, and patients treated with these medications should be periodically evaluated for HE.

Abstract M1210 – Diabetes mellitus worsens the clinical outcome of acute variceal bleed in cirrhotic patients

H. A. Shah; S. Majid; M. Salih; F. W. Ismail; K. Mumtaz; S. S. Hamid; W. Jafri

Introduction:

Association of Diabetes Mellitus (DM) and hepatic cirrhosis is well recognized and hyperglycemia induces splanchnic hyperemia with increase in portal pressures. There is scanty data regarding the clinical outcome in cirrhotic diabetics following variceal bleed.

Aim:

To compare the clinical outcome of gastro-esophageal variceal (GOV) bleed in cirrhotic diabetic patients with those without diabetes.

Patients and Method:

We reviewed the case notes of patients with GOV bleed admitted from June 2000-2005, under gastroenterology service at Aga Khan University Hospital. Diagnosis of cirrhosis was made on clinical, laboratory, radiological findings. Liver biopsy was done in selected patients. Acute Variceal bleeding, failure to control bleed and rebleeding were defined according to Baveno III consensus report.

Results:

A total of 839 admissions with variceal bleed occurred in 382 patients. Diabetes was present in 148(39%) patients.Comparison between diabetic and non diabetic cirrhotics with variceal bleed is presented in Table 1.

Conclusion:

The presence of Diabetes mellitus in patients with cirrhosis signifies a worse prognosis for control of acute gastro-esophageal bleed and incidence of rebleed in hospital. The risk of rebleed following discharge is also increased in diabetic cirrhotics as compared to non diabetics.

Variable	Diabetics n (%)	Non diabetics n (%)	P value
1) Age (yrs.) 2) Gender Male Female 3) Child Pugh class. A B C 4) Bilirubin <2 2-3 >3 5) Ascites 6) No. of admissions with GI bleed 7) Failure to control bleed 8) Rebleeding during hospital stay	53.4±10.2 95(64%) 53(36%) 7(4.7) 64(43.2) 77(52) 101(68.2) 26(17.6) 21(14.2) 88(59.5) 2.49±1.3 17(11.5) 21(14.2)	52.2±13.4 145(62%) 89 (38%) 11(4.7) 73(31.2) 150(64.1) 116(49.6) 43(18.4) 75(35.1) 138(59) 2.03±1.0 13(5.6) 17(7.3)	NS NS NS 0.01 NS NS 0.001 0.001 NS 0.001 0.05 0.02

Abstract M1206 – Predictive factors of survival after TIPS insertion for refractory ascites.

S. Evrard; F. Nevens; A. Mroue; M. Adler; J. Deviere; O. Le Moine

Background:

TIPS may improve the control of refractory ascites (RA) in patients with cirrhosis. However, it may induce severe liver failure and therefore jeopardise their chance to benefit from liver transplantation. Currently, no criteria exist to adequately select patients for TIPS and assess their short-term prognosis after the procedure.

Aim:

To identify pre-procedural clinical and biological variables associated with survival after TIPS placement in a large cohort of patients with RA.

Methods:

117 consecutive patients, from 2 belgian referral centers, who underwent non covered TIPS for RA between January 1992 and August 2004 were retrospectively analysed. Univariate and multivariate analysis (Cox regression) were performed to find the independent variables associated with 1 and 6 months survival.

Results:

The population included 75% of alcoholic cirrhosis and 62% of Child B patients. No procedure related deaths were encountered and survival was 85%, 66% and 58% at 1, 6 and 12 months, respectively. Ascites control was observed in 87% and 91% of surviving patients after 1 and 12 months, respectively. Thirty eight % of patients developed encephalopathy after TIPS. Alcoholic etiology (ETIO), pre-TIPS low bilirubin level (BILI) and high PTT were independent predictors of 1-month survival (OR: 2,5; 1,08 and 0,96 respectively). A prognostic index (pi) could be calculated as following: pi=exp {(0,085xBILI)+(-0,038xPTT)+(0,903xETIO)} where BILI is expressed in mg/dl, PTT in % and ETIO as 1 if alcoholic or 2 non-alcoholic. By plotting pi on a Youden curve, the best cut-off value was 1,2. Values of pi < 1,2 predicted 96% of the patients who will survive 1 month after TIPS (specificity) whereas values > 1,2 were less sensitive to predict death (44%). When analysing 6 months survival, ETIO and PTT were still independently predicting outcome.

Conclusions:

Alcoholic cirrhosis and preserved liver function are predictive of survival after TIPS for RA. The prognostic index (pi) may help select the patients who will benefit the most of the procedure without jeopardising further liver transplantation. Prospective validation of this index is now mandatory.