Effect of a moderately energy-restricted diet on obese patients with fatty liver.

The effects of a moderately energy-restricted (25 kcal/kg) diet on liver-function tests, anthropometric measurements, mononuclear-cell phospholipid fatty acid, lymphocyte blastogenesis, and plasma prostaglandin E2 and alpha-tocopherol levels were observed at weeks 0, 8, and 24 in 14 obese patients with fatty liver. Serum aminotransferase levels were improved significantly, with decreases in the body mass index and waist circumference. Decreases in energy intake from carbohydrate and increases in intake of vitamin A, vitamin C, and vegetables were observed at week 24. In mononuclear-cell phospholipids, linoleic acid (18:2omega 6), which was significantly lower in patients than in controls at week 0, was increased at week 24. In contrast, arachidonic acid was decreased. Plasma prostaglandin E2 levels were significantly lower in patients than in controls at week 0 and increased at week 24. The mononuclear-cell response for phytohemagglutinin correlated with 18:2omega 6 in mononuclear-cell phospholipids (r = 0.692, P < 0.01). Improvement of the serum alanine-aminotransferase level correlated with an increase in the plasma alpha-tocopherol level (r = -0.667, P < 0.01) and increases in consumption of vitamin A, omega 3 polyunsaturated fatty acids, and vegetables. These findings suggest that a hypoenergetic diet rich in omega 3 polyunsaturated fatty acids and antioxidants might be beneficial for obese patients with fatty liver.

Clinical usefulness of the branched DNA assay version 2 in predicting the efficacy of Interferon treatment in a group of chronic HCV patients based on serotype.

Interferon (IFN) response depends upon pretreatment viral loads and viral genotype/serotype. This study investigated the difference in response to IFN in different viral load groups of 96 chronic hepatitis C patients and compared version 1 (bDNA1.0) and version 2 (bDNA2.0) of the branched DNA assay, according to serotype. On univariate analysis, viral load (P=0.0001, by bDNA 1.0; P=0.0020, by bDNA 2.0,), serotype (P=0.0053) and age (P=0.0073) were significant predictors of IFN response. On multivariate analysis, serotype (odds ratio, 5.44; 95% confidence interval, 1.94-15.24; P<0.01) was a stronger predictor of IFN response than age or viral load (by bDNA2.0). In very high (>6.7 log eq/ml), high (6.0 approximately 6.69 log eq/ml) and low (<6 log eq/ml) viral load groups (by bDNA2.0), complete response was 25, 55 and 92.6% in serotype 2 (sero-2), and 10, 20 and 71.4% in serotype 1 (sero-1), respectively. In sero-2, bDNA2.0 can detect high viral loads underestimated by bDNA1.0. In a low viral load group (by bDNA2.0), IFN response is dependent upon serotype; sero-2 responded better than sero-1. However, in high and very high viral load groups, sensitivities of bDNA1.0 and bDNA2.0 are not effective in clinically distinguishing CR from non-response, and aid in patient selection for IFN therapy.

Interferon therapy reduces the risk for hepatocellular carcinoma: national surveillance program of cirrhotic and noncirrhotic patients with chronic hepatitis C in Japan. IHIT Study Group. Inhibition of Hepatocarcinogenesis by Interferon Therapy.

BACKGROUND: Previous studies on the effect of interferon therapy on the incidence of hepatocellular carcinoma have not sufficiently assessed degree of liver fibrosis, a major risk factor for hepatocellular carcinoma. OBJECTIVE: To evaluate the effect of interferon therapy on incidence of hepatocellular carcinoma, adjusting for risk factors, including the degree of liver fibrosis. DESIGN: Retrospective cohort study. SETTING: Seven university hospitals and one regional core hospital in Japan. PATIENTS: 2890 patients with chronic hepatitis C who had undergone liver biopsy since 1986. Of these patients, 2400 received interferon and 490 were untreated. MEASUREMENTS: The degree of liver fibrosis was assessed from stage F0 (no fibrosis) to stage F4 (cirrhosis). Response to interferon was determined virologically and biochemically. Screening for development of hepatocellular carcinoma was performed periodically during an average follow-up of 4.3 years. Effect of interferon therapy on the risk for hepatocellular carcinoma was analyzed by using Cox proportional hazards regression. RESULTS: Hepatocellular carcinoma developed in 89 interferon-treated patients and in 59 untreated patients. Among untreated patients, the annual incidence of hepatocellular carcinoma increased with the degree of liver fibrosis, from 0.5% among patients with stage F0 or F1 fibrosis to 7.9% among patients with stage F4 fibrosis. The cumulative incidence in treated and untreated patients differed significantly for patients with stage F2 fibrosis (P = 0.0128) and for those with stage F3 fibrosis (P = 0.0011). In multivariate analysis, interferon therapy was associated with a reduced risk for hepatocellular carcinoma (adjusted risk ratio, 0.516 [95% CI, 0.358 to 0.742]; P < 0.001), especially among patients with sustained virologic response (risk ratio, 0.197 [CI, 0.099 to 0.392]), among those with persistently normal serum alanine aminotransferase levels (risk ratio, 0.197 [CI, 0.104 to 0.375]), and among those with alanine aminotransferase levels less than two times the upper limit of normal (risk ratio, 0.358 [CI, 0.206 to 0.622]). CONCLUSIONS: Interferon therapy significantly reducesthe risk for hepatocellular carcinoma, especially among virologic or biochemical responders.

Four cases with intrafamilial clustering of hepatitis C virus infection.

Four hepatitis C patients with intrafamilial clustering of hepatitis C virus (HCV) infection are reported. Antibodies to C100-3 antigen, capsid protein of HCV and GOR epitope were tested to detect histories of HCV infection. Transmission of HCV from mother to children, from father to children, and from wife to husband was implicated. Of all family members studied, three were positive for all antibodies, one for only antibody to capsid protein, two for antibodies to capsid protein and GOR epitope but negative for antibody to C100-3 antigen and one vice versa.

Immunoelectron microscopic observation of intrahepatic HBeAg in patients with chronic hepatitis B.

Immune light and electron microscopic studies using monoclonal antibodies have been applied to localize HBeAg in liver biopsy specimens of 19 patients with chronic hepatitis B. Under the light microscope, HBeAg was demonstrated in nuclei, cytoplasm and on the cell surface of hepatocytes. The number of HBeAg-positive hepatocytes correlated well with the serum levels of HBeAg (enzyme immunoassay) and DNA-polymerase. Of 11 patients in whom high numbers of HBeAg-positive hepatocytes were found at the light microscopic level, HBeAg was also studied in hepatocytes at the electron microscopic level. The HBeAg in nuclei was either found as aggregates or dispersed diffusely. In the aggregates of HBeAg, the 27-nm core particles were frequently found. In addition, the antigen was found in the cytosol of some hepatocytes as amorphous mass and in some hepatocytes in the cisternae of perinuclear space, endoplasmic reticulum and Golgi saccules. Occasionally the antigen was found on the membranes of the cell organelles and on the plasma membranes that faced the intercellular and the Disse spaces. These findings suggest that cytoplasmic HBeAg in hepatocytes may be ultrastructurally classified into two different patterns by its distribution, mainly in endoplasmic reticulum or in cytosol. The ratio varied between hepatocytes with these two types of patterns. The titer of serum HBeAg tended to be higher when the corresponding liver biopsy specimens contained more hepatocytes with HBeAg in endoplasmic reticulum than those with HBeAg in the cytosol.

Immunoelectron microscopic observation of alpha-fetoprotein synthesis in human non-malignant liver tissues using immunoperoxidase methods.

Alpha-fetoprotein (AFP) synthesis in non-malignant liver tissue of 34 patients with chronic hepatitis or liver cirrhosis, some of whom also had hepatocellular carcinoma (HCC), was studied by light and ultrastructural immunohistochemistry using peroxidase-labeled anti-human AFP. Simultaneously, the serum level of AFP was measured in these patients by radioimmunoassay. AFP-positive cells were identified in non-malignant liver tissue of 7 patients with elevation of serum AFP. AFP was demonstrated in several hepatocytes which were clustered in hepatic lobules, and also in some bile ductular cells which were distributed in the periphery of portal tracts. In an immunoelectron microscopic study of AFP-positive hepatocytes, dense reaction products of anti-AFP were localized in the membranes and cisternae of rough endoplasmic reticulum (r-ER), perinuclear space (PNS) and Golgi apparatus. The prominent feature of AFP-positive hepatocytes was abundant r-ER encompassing many mitochondria. As to AFP-positive bile ductular cells, they had scanty cytoplasm and few intracytoplasmic organelles and were surrounded by basement membrane. AFP was focally localized in the r-ER of such bile ductular cells. These observations suggest that AFP can be produced by malignant and non-malignant liver cells and that in non-malignant liver tissues, AFP can be produced by two distinct cell types; bile ductular cells and hepatocytes themselves.

Ultrastructural observation of alpha-fetoprotein producing cells in human hepatocellular carcinoma using immunoperoxidase methods--comparison with fetal liver.

The precise sites of alpha-fetoprotein (AFP) synthesis and ultrastructural features and differences of AFP-producing cells were observed in periodate-lysine-paraformaldehyde fixed, frozen liver tissues from four human hepatocellular carcinoma (HCC) patients and three human fetuses using the direct (horseradish peroxidase-labeled Fab' fraction of anti-human AFP) immunoperoxidase method. We demonstrated that AFP was located in the membrane and cisternae of rough endoplasmic reticulum, membrane-bound ribosomes, perinuclear space and Golgi apparatus. The location and intensity of immunoreaction products of AFP in hepatoma cells varied from cell to cell and case to case, while these features tended to be regular in fetal hepatocytes. We did not observe ultrastructural differences between AFP-producing and non-producing cells adjacent to each other. These observations indicate that AFP production does not occur in morphologically distinct cell populations of hepatoma tissue and that hepatoma tissue is functionally much more heterogeneous than fetal liver.

Cellular immune response in liver of patients with chronic hepatitis B. Electron microscopic observation of lymphocyte subsets by the immunoperoxidase method using monoclonal antibodies.

The morphological association between lymphocytes and hepatocytes was studied at the light and electron microscopic levels by the peroxidase-labeled antibody method using mouse monoclonal antibodies against Leu-1, Leu-2a, Leu-3a, Leu-7 and Leu-10 antigens in liver biopsy specimens from patients with chronic hepatitis B. Leu-1 + cells (T cells), especially Leu-2a + cells (cytotoxic/suppressor T cells), infiltrated mostly in periportal areas with piecemeal necrosis and in parenchymal areas with focal necrosis. By double staining techniques, Leu-2a + cells were often seen in contact with hepatocytes containing membranous hepatitis B surface and/or core antigens in patients with chronic active hepatitis. At the ultrastructural level, Leu-2a + cells frequently occupied the sinusoid and also migrated into both the space of Disse and between hepatocytes. Furthermore, they often showed intimate surface-contact with hepatocytes having hepatitis B surface and/or core antigens, and, occasionally, injured hepatocytes were surrounded by several Leu-2a + cells. In contrast, Leu-3a + cells, Leu-7 + cells and Leu-10 + cells sometimes appeared in the sinusoid, but seldom in the space of Disse and between hepatocytes. These findings suggest that cytotoxic T lymphocytes may be associated with the necrosis of hepatocytes in chronic hepatitis B.

Clinical and histological features of sporadic non-A, non-B hepatitis.

The incidence of hepatitis A (HA), hepatitis B (HB), and non-A, non-B hepatitis (NANBH) was 27%, 30% and 43% among 73 patients with sporadic hepatitis. Epidemiological data (geographical distribution, seasonal variation, age, sex, and occupation) were not distinguishing of the type of hepatitis. Neither intrafamilial infection nor previous contact with viral hepatitis patients could be demonstrated in the NANBH cases. Fever and jaundice were less frequent in NANBH than in HA. Maximum levels of SGPT, serum bilirubin, ZTT, and gamma-globulin were significantly lower in NANBH than in HA and HB. Ten of 29 NANBH patients (35%) presented abnormal SGPT activities for more than 6 months, and four (14%) more than 12 months. In the ten patients with prolonged courses, jaundice was more frequent and maximum levels of SGPT were higher than in patients with transient courses. Histopathologic findings were not markedly different from those of HA and HB. Bile duct damage, fatty deposition, and giant multi-nucleated cells were recognized in 6, 12, and 2 NANBH patients, respectively. There were no characteristic ultrastructural changes in NANBH.

Non-B hepatocellular carcinoma: influence of age, sex, alcohol, family clustering, blood transfusion and chronic liver disease.

In 144 cases of hepatocellular carcinoma (HCC), 166 cases of cirrhosis without HCC and 142 cases of chronic hepatitis, we examined HBsAg, anti-HBs and anti-HBc in sera and compared the following factors between hepatitis B virus marker-negative and -positive patients: age, sex, alcohol consumption, family clustering of liver diseases, and histories of blood transfusion and post-transfusion hepatitis. Results of this study demonstrated several distinct differences in clinical backgrounds between non-B (negative for HBsAg, anti-HBs and anti-HBc) and B (positive for HBsAg) patients with HCC. Non-B patients were significantly older, had a lower frequency of familial tendencies for liver diseases, and more frequently had cancers other than HCC in their families. Some of these differences were also observed between non-B and B patients with cirrhosis and chronic hepatitis. Among patients with chronic hepatitis, the non-B patients had received blood transfusion or had post-transfusion hepatitis more frequently than the B patients. However, this difference was not apparent in patients with liver cirrhosis or HCC, suggesting that progression of non-A, non-B post-transfusion hepatitis to cirrhosis and HCC may not be as frequent as progression to chronic hepatitis.

Full and empty particles of hepatitis B virus in hepatocytes from patients with HBsAg-positive chronic active hepatitis.

Complete and defective hepatitis B virus (HBV) particles in sera and hepatocytes were observed by electron microscopy for an understanding of the maturation process of hepatitis B virus. To distinguish Dane particles with or without DNA on the basis of staining density with uranyl acetate, Dane particles, purified from sera of asymptomatic carriers, and Dane particle cores, separated by ultracentrifugation in a metrizamide density gradient, were observed by electron microscopy. Complete cores at a low density (1.19 to 1.23 gm/cm3) were electron dense and incomplete cores at a high density (1.23 to 1.27 gm/cm3) were partially electron dense or empty. These findings demonstrated that the presence or absence of DNA is reflected by the electron density of the core. Our result, in which less than 10% serum Dane particles have full cores in ultrathin sections of the pellet, is in agreement with Gerin's finding that defective Dane particles are predominent in sera. Naked core particles and Dane particles in two biopsy specimens from patients with hepatitis B surface antigen, hepatitis B e antigen, and DNA polymerase-positive, chronic active hepatitis were classified into complete versus incomplete particles on the basis of electron density. Nine hundred and fourteen naked core particles were detected in nuclei and cytosol of 68 hepatocytes. One hundred and five core particles (11.5%) were electron dense and 809 core particles (88.5%) were partially electron dense or empty. Furthermore, 488 Dane particles were observed in the cisternae of the endoplasmic reticulum of these hepatocytes. Fifty Dane particles (10.2%) had full cores, and 438 Dane particles (89.8%) had partially full or empty cores. These findings suggest that DNA may be incorporated into about 1 to 10% of core particles when they are assembled in nuclei of hepatocytes. Morphologic differences in damage to hepatocytes containing various frequencies of full Dane particles were also studied, but no significant correlation was found between damage in hepatocytes and frequency of full Dane particles.