Search

Objective: To compare cost effectiveness of radio-frequency (RF) ablation versus percutaneous ethanol (Pe) ablation for treatment of small hepatocellular carcinoma., Methods: 57 patients with 72 hepatocellular carcinomas (HCC) treated with ethanol ablation were retrospectively compared with 60 new patients who had 72 HCC treated with RF ablation. All patients had Child-Pugh A cirrhosis and HCC less than 3.5 cm in diameter. Therapeutic efficacy was evaluated with tri-phase spiral computed tomography. Cost of hospital stay, supplies and follow-up including complications were calculated. Survival was estimated by Kaplan-Meier method and prognostic value of treatment methods by Log Rank test and multivariate Cox proportional hazard model., Results: The two groups had similar baseline characteristics. The rate of severe complications associated with RF was 15% (9/60) vs 6,9% (9/60) with Pe (P=0,11) The two-year overall survival, disease free survival and local tumor free survival in the ethanol and radiofrequency groups were 70.8% vs. 91.2% (Odd Ratio=3.7, P=0.006), 48.6% vs. 71.1% (Odd Ratio=2.2, P=0.01), and 68.5% vs. 80.7%, (Odd Ratio=1.9; P=0.09), respectively. The treatment method was the sole significant factor related to overall and tumor free survival. Costs per patient of ethanol and radiofrequency treatments were 1534 euro and 1196 euro, respectively., Conclusion: This retrospective study suggests that RF is the most cost effective strategy, with higher 2-year disease free survival rate.

Aim: The presence of esophageal varices has been reported to be a prognostic factor in patients with compensated hepatitis viral C induced cirrhosis. We studied the prognostic value of hepatic venous pressure gradient in addition to epidemiological and clinical parameters in these patients., Methods: Among patients with Child-Pugh A hepatitis C induced cirrhosis, prospectively followed in two Parisian centres, 100 had measurement of occluded and free hepatic venous pressures. We evaluated hepatic venous pressures as a predictive factor of death by Cox models (survival) and Fine and Gray models (liver-deaths)., Results: Median hepatic occluded pressure and gradient were 21.5 (15-24) and 13 mm Hg (9-15), respectively. The median duration of follow-up was 85 months (range: 70-112); 38 deaths or liver transplantation were registered. Hepatic venous pressure gradient was not significantly related to survival in the studied population but as a continuous variable was predictive of death from liver disease. On multivariable analysis serum albumin <40 g/L and platelet count <90,000 /mm(3) were the only selected prognostic factors., Conclusion: Hepatic venous pressure gradient has a limited value for assessing the prognosis of patients with Child-Pugh A hepatitis C virus induced cirrhosis; prognosis is accurately predicted by serum albumin and platelet count.

Aim: To investigate the feasibility and efficacy of hepatitis C virus screening in drug users in an addiction out-patient unit., Patients and Methods: All patients followed in an addiction out-patient unit were asked to undergo anti-hepatitis C virus antibody testing; further evaluation and treatment if indicated, were offered to positive patients. When treatment was initiated (Metavir score >=F2), patients were followed-up both by the hepatologist and the out-patient unit physician., Results: Between July 1997 and September 2000, 404 consecutive patients (310 men, mean age: 32, alcohol intake >=50 g per day in 51%, 94% in opiate substitution program) were included. Sixty-six per cent (269/404) of patients agreed to undergo HCV antibodies testing: 84% had a positive test. 68% of these patients accepted ALT serum measurement and 120 had indications for liver biopsy. Eighty-eight liver biopsies were performed, showing severe fibrosis (Metavir score F3 or F4) in 20 cases (22%). Ethanol intake was significantly correlated to fibrosis (P<0.05). Antiviral treatment was indicated in 47 patients but was only initiated in 27 due to patient refusal (n=7) or contraindication (n=13). Treatment had to be discontinued in 12 cases because of psychiatric side effects (depression: n=3; delirium: n=3; severe irritability: n=3; relapse with heroin injection: n=3). Finally, only 5 patients were sustained responders., Conclusion: Despite the high seroprevalence of HCV antibodies in this unit, the benefits of antiviral therapy are low due to high drop out rate. Ethanol withdrawal should be the highest priority in these patients.

Aim: To describe French practices for screening hepatocellular carcinoma., Methods: A standardized questionnaire was mailed to all French hospital hepato-gastroenterologists in June 1999., Results: 411 out of 623 practitioners responded (66%). 394 (96%) routinely screen hepatocellular carcinoma, mainly with ultrasound (98%) and mainly at 6-month intervals (77%). Screening was performed in cirrhosis (100%) or extensive fibrosis (54%), independent of the etiology (21%) or the Child-Pugh score of the chronic liver disease (41%), but based on age and treatment feasibility. If of a small hypoechogenic nodule was detected in a young patient with compensated HCV-cirrhosis, 59% of practitioners performed a histological examination. In case of non biopsy-proven hepatocellular carcinoma, a second biopsy (49%), treatment (either percutaneous alcohol injection, resection or transplantation) (24%) or an ultrasonographic follow-up (23%) was proposed. In case of biopsy-proven hepatocellular carcinoma, resection (49%), transplantation (30%) or percutaneous alcohol injection (16%) was proposed., Conclusion: Almost all French specialists routinely screen cirrhotic patients for hepatocellular carcinoma, but use somewhat different modalities. In case of small HCC without contraindications to curative treatment, surgical resection is performed in half the patients.

Objectives: Genetic alterations in the p53 protein may induce serum anti-p53 antibodies. The aim of this study was to assess the prevalence of serum anti-p53 antibodies in a large series of Western patients with hepatocellular carcinoma and the prognostic value of these antibodies on survival., Methods: Serum anti-p53 antibodies were assayed in 159 patients with hepatocellular carcinoma, at diagnosis, using an immunoenzymatic method. The initial patient characteristics were compared according to anti-p53 status. The prognostic value of these antibodies on survival was determined by univariate and multivariate analysis., Results: One hundred fifty nine patients with hepatocellular carcinoma were included in the study (129 men, mean age: 68 years). The main associated causes of chronic liver disease were alcohol (n=86), virus C (n=34), and virus B (n=18); 151 patients had cirrhosis. Median follow-up was 306 days. Among the 159 patients, 19 (12%) had serum anti-p53 antibodies. Detection of serum anti-p53 antibodies was significantly correlated with the presence of a multinodular or infiltrative tumor (P<0.03). Survival was not influenced by the presence of anti-p53 antibodies. Serum albumin, ALAT, and alfa-fetoprotein levels were independent prognostic variables., Conclusion: In our study, anti-p53 antibodies were rarely found in the serum of patients with hepatocellular carcinoma. Although they were correlated with multinodular or infiltrative tumors, they had no prognostic influence on survival. Thus, assaying serum anti-p53 antibodies does not seem to have any clinical value in those patients.

Aim: Long term treatment by fibrates could induce chronic hepatitis associated with auto-antibodies. The aim of this study was to assess the features of this hepatitis., Methods: Baseline clinical and biological features, and liver biopsy in 5 patients with fibrate-induced chronic hepatitis were studied, as well as their outcome., Results: At enrollment, patients (4 men, mean age 65 years) had highly (n = 3) or mildly (n = 2) increased serum aminotransferase activity, hypergammaglobulinemia and high titers of anti-nuclear antibodies (with homogenous fluorescence in 3 cases). Liver biopsies demonstrated a lympho-plasmocytic infiltrate in all cases. Hepatocellular necrosis was multilobular in 2 cases, and mild to moderate, located in lobular and periportal areas in 3 cases. Cirrhosis was found at presentation in 3 cases and developed within a few months in the 2 other patients. After discontinuation of fibrates, aminotransferase activity normalized within 6 weeks either spontaneously (n = 3) or under immunosuppressive treatment (n = 2). Immunosuppression was rapidly withdrawn in 2 patients (< 18 months) without relapse, while one patient was treated for 4 years because of relapse after early withdrawal. A second liver biopsy performed 6 months after discontinuation of fibrates in an untreated-patient showed no inflammation or necrosis., Conclusion: These observations suggest that fibrates could trigger chronic liver disease resembling type I auto-immune chronic hepatitis, which resolves after drug withdrawal.

Objectives: Giant-cell hepatitis is rare in adults and its significance has not been clarified. We report the clinical and histological characteristics and outcome in a group of adult patients with giant-cell hepatitis., Methods: Seventeen patients with giant-cell hepatitis, hospitalized in our unit between 1976 and 1992, were studied retrospectively. Giant-cell hepatitis was defined as at least two hepatocytes with four or more nuclei per cell on liver biopsy. Clinical and biochemical parameters, liver histology, and the serological profile of HAV, HBV, HCV, HIV, HSV, EBV, CMV, and paramyxovirus were evaluated. Paramyxovirus immunochemistry was performed in 6 liver biopsies., Results: There were 11 females and 6 males, an average of 48 years old (range: 29-80). Four patients had a well-defined etiology: acute hepatitis B infection with a favorable outcome in 2 cases, clometacine induced-hepatitis resulting in death from liver failure in one case, and chronic hepatitis B and C in one patient with AIDS. Among the 13 patients in which the etiology could not be determined, histologically defined acute hepatitis was observed in 8 and chronic hepatitis in 5. Nine patients were treated with immunosuppressive drugs. One patient was lost to follow-up. Eight patients responded to treatment, but 5 patients progressed to cirrhosis between 5 months and 7 years. Two of the 4 patients with unexplained liver disease who did not receive any treatment died of liver failure., Conclusion: In patient with acute or chronic hepatitis without an identified cause (with or without autoimmune abnormalities), the presence of giant-cell hepatitis seems to have a similar evolution as active autoimmune hepatitis. The poor prognosis of these patients suggests that early immunosuppressive treatment is justified.

Objectives and Methods: Extrahepatic manifestations have been reported in hepatitis C virus infection. To assess the relationship between diabetes mellitus and hepatitis C virus, we studied 152 patients with chronic hepatitis C and 152 controls hospitalized during the same period with hepatitis B virus (n = 51) or alcohol-induced (n = 101) liver diseases matched for age, sex, and the presence of cirrhosis (prevalence: 58%). Patients with jaundice, ascites, encephalopathy, prothrombin activity < 65%, or serum albumin < 35 g/L were excluded., Results: Diabetes, defined by fasting serum glycemia > 1.4 g/L on at least two separate occasions or previously treated overt diabetes, was present in 38 patients with chronic hepatitis C (24%) and in 13 patients in the control group (9%, P < 0.002). In the 51 diabetic patients, irrespective of serum anti-hepatitis C virus status, 41 (81%) had non insulin dependent diabetes and 45 (88%) had cirrhosis. Family history of diabetes or obesity was observed in 2 (5%) of the diabetic patients with chronic hepatitis C and in 6 (46%) of the diabetic controls (P = 0.002). Plasma C-peptide (855 +/- 448 pmol/L versus 1,152 +/- 491 pmol/L, NS) and insulin levels (83 +/- 40 pmol/L versus 184 +/- 86 pmol/L, NS), assayed in 17 diabetic patients with chronic hepatitis C and in 9 diabetic controls, were lower in the former. The prevalence of HLA B8, DR3 or DR4 antigens, which was searched for in 77 patients with chronic hepatitis C, was not different in diabetic and non diabetic patients, and, was similar to the reference population. Serum islet-cell antibodies were found in 5 patients with chronic hepatitis C (3 with diabetes) and in 2 controls., Conclusion: Diabetes mellitus is more prevalent in patients with chronic hepatitis C than in patients with other liver diseases, and usually occurs in the absence of predisposing factors. These results suggest a role of hepatitis C virus infection in the pathogenesis of diabetes.

Two cases of sclerosing cholangitis after oily arterial chemoembolization are reported. In one patient angiocholitis with liver abscesses, in the other patient gradual cholestasis were the main clinical features. In both cases, endoscopic retrograde cholangiogram showed a stricture of the common hepatic bile duct and, in one case, irregularities of intrahepatic biliary tree. Histologic examination of the liver in the two patients pointed out the involvement of small bile ducts and arteriolar endarteritis obliterans. Ischaemia is likely to be the main mechanism of these two cases of sclerosing cholangitis as well as in those described after FUDR intra-arterial chemotherapy. The prevalence of sclerosing cholangitis after arterial oily chemoembolization is probably underestimated because of a non specific clinical presentation and need to be precise by further study.

The aim of this study was to evaluate the relationship between the serum N-terminal peptide of type III procollagen (PIIINP), Knodell's score and the response to treatment in 79 patients with viral C chronic hepatitis. Liver biopsy and serum PIIINP was assessed in all patients. Serum PIIINP was correlated with Knodell'score (r = 0.60; P = 10(-4)). Correlation between serum alanine amino-transferase and Knodell'score was smaller than for PIIINP (r = 0.38; P = 0.02). Forty-nine patients were treated with interferon alpha. A second biopsy and serum PIIINP determination were performed one year after treatment discontinuation. Means of serum PIIINP and Knodell'score significantly decreased after treatment (P < 0.01; P < 0.03). Among non-responder patients, serum PIIINP was initially more elevated: 37.5 +/- 12.6 ng/mL versus 22.6 +/- 3.8 (P < 10(-3)). Knodell's score decreased only among responder patients (P < 0.01 vs P < 0.7). Positive and negative predictive value for the response to treatment, evaluated by aminotransferases normalization, of serum PIIINP level < or = 24 ng/mL were 0.6 and 0.9 respectively. These findings suggest that serum PIIINP is a relevant marker of activity in patients with viral C chronic hepatitis and that it might have a predictive value of response to treatment.

The aim of study was to assess the role of hepatitis C virus (HCV) infection in 164 alcoholic cirrhotic patients. We studied the prevalence of anti-HCV antibodies using ELISA and RIBA first and second generation tests. Twenty-two % of the patients had anti-HCV antibodies detected by ELISA 2, RIBA 2 test was positive in 10% of the patients and indeterminate in 3%. We compared epidemiological, biological and histological characteristics according to the results of the tests. By comparing ELISA 2-RIBA 2 positive patients to ELISA 2 negative patients, we observed, in the former, a) a higher serum aminotransferase activity, b) a lower serum gammaglutamyl transpeptidase activity, and c) a lower histological score of alcoholic hepatitis. In addition, in a group of ELISA 2 positive RIBA 2 negative patients, the values were intermediate between those of the two former groups. However, most of these patients had a negative third generation ELISA test. The whole results suggest that HCV is likely to play a role in the pathogenesis of liver damage in a high number of alcoholic cirrhotic patients.

The aim of this study was to compare 6 potential serum markers for hepatic fibrosis in patients with alcoholic liver disease. Ninety-three patients (50 +/- 11 years old, 62 males) with biopsy-proven alcoholic liver disease were included in the study. A liver biopsy and serum assays of type I, type III and type IV collagens, N-terminal peptide of type III procollagen, laminin (by radioimmunoassays) and apolipoprotein A1 (by nephelometry) were performed in all patients. A histological score of hepatic fibrosis was established. Alcoholic hepatitis lesions and perisinusoidal fibrosis were assessed separately. A significant correlation was found between the score of hepatic fibrosis and serum levels of type I collagen (r = 0.44, P < 10(-3)), type III collagen (r = 0.36, P < 10(-2)), N-terminal peptide of type III procollagen (r = 0.50, P < 10(-3)), type IV collagen (r = 0.44, P < 10(-3)), laminin (r = 0.50, P < 10(-3)), and apolipoprotein A1 (r = 0.21, P < 0.05). After adjustment for the presence of lesions of alcoholic hepatitis and perisinusoidal fibrosis (partial correlation), serum levels of type I collagen, type III collagen, N-terminal peptide of type III procollagen, type IV collagen, and laminin remained significantly correlated with the score of hepatic fibrosis; in contrast, correlation with serum apolipoprotein A1 was no longer significant. Serum levels of N-terminal peptide of type III procollagen, type IV collagen and laminin were significantly higher in patients with perisinusoidal fibrosis.(ABSTRACT TRUNCATED AT 250 WORDS)

The authors report the cases of 4 patients with heavy chronic alcoholic intake who presented with hepatomegaly and jaundice without obvious hepatic failure and who died rapidly. In all 4 cases, histological examination of the liver showed massive microvesicular and macrovesicular steatosis involving approximately 100% of hepatocytes and, in 2 cases, minimal lesions of alcoholic hepatitis. Histochemical study, performed in 3 cases, showed that steatosis was constituted of triglycerides only, and that hepatic glycogen was completely depleted in 2 of 3 cases. No obvious cause of death was found in these 4 patients. Shortly, before their death, the 4 patients had increased their ethanol and decreased their food intake. The authors suggest that death as well as microvesicular steatosis could have be due to acute mitochondrial dysfunction.