Search

Hepatitis C virus (HCV) polyprotein translation depends on direct internal entry of the 40S ribosomal subunit mediated by an internal ribosome entry segment (IRES) located in the 5′ noncoding (5′NC) region of the viral genome. HCV is genetically heterogeneous and is characterized by the existence of a quasispecies distribution of the virus population within a single infected individual. Cloning and sequencing strategies were used to characterize 5′NC quasispecies genetically. Similar to coding regions, the HCV 5′NC region was distributed as a quasispecies, but it appeared to be subjected to stronger conservatory constraints than other regions of the HCV genome, probably due to the need for structural (and functional) conservation of the IRES. Indeed, most of the quasispecies substitutions were in unpaired regions of the IRES or clustered such that base-pairing was maintained, whereas only 21% were expected to result in a loss of base-pairing. Quasispecies-related structural changes could be predicted in the core cruciform of IRES domain III composed of the RNA helices which extend from the four-way junction JIIIabc, mostly in minor variants, but sometimes in major ones. The results presented here suggest the simultaneous presence in infected patients of a mixture of genetically distinct but closely related IRES sequences that may have different structures. No significant genetic changes of 5′NC quasispecies were observed after interferon-α treatment, except in patients with mixed genotype infection who cleared one of the infecting strains during therapy, suggesting that the quasispecies distribution of IRES sequences does not play a role in HCV resistance to interferon-α therapy. In contrast, the overall quasispecies distribution of HCV genomes (including IRES sequences) might participate in regulation of hepatic and extrahepatic HCV replication.

The inducing effect of rifampicin (600 mg per day) on the hepatic microsomal drugmetabolizing system has been studied in 7 patients with normal liver after 6 days of oral administration. Using the plasma disappearance rate of antipyrine (15 mg per kg of body weight) as an index of liver microsomal metabolism, a significant decrease in the half-life of antipyrine has been observed: 11.7 ± (1 SD) 4.7 hr before treatment as compared to 6.9 ± 2.3 hr on the 7th day. Concomitantly, the half-life of rifampicin was 5.3 ± 2.1 and 2.7 ± 0.8 hr on the 1st and 7th day, respectively. These results demonstrate that rifampicin administration for only 6 days leads to the induction of the microsomal mixed function oxidase system in liver. These data are in agreement with the recent hypothesis that rifampicin may play a role in the hepatotoxicity of isoniazid by enhancing its microsomal transformation to a toxic metabolite.

Gallbladder bile collected by duodenal intubation or during surgery was examined microscopically in patients who were free of stones and in patients with proven stones. None of the 16 patients free of stones had cholesterol monohydrate crystals or calcium bilirubinate granules in bile. Among the 17 patients with proven cholelithiasis, 13 with cholesterol stones had cholesterol monohydrate crystals in their bile, but only 2 of the 4 patients with pigment stones had calcium bilirubinate granules. These data confirm that cholesterol monohydrate crystals are sensitive and specific for cholesterol stones, whereas calcium bilirubinate granules lack sensitivity for the diagnosis of pigment stones. From these results, the diagnostic usefulness of microscopic examination of bile collected from the duodenum was studied prospectively in 46 patients with symptoms suggestive of cholelithiasis but in whom stones had not been visualized at cholecystography and ultrasonography. In 15 of them, bile was found to be abnormal: cholesterol monohydrate crystals were seen in 11, cholesterol crystals + calcium bilirubinate granules in 2 and calcium bilirubinate granules in 2. To date, nine of these patients have been operated on: 6 (all with cholesterol monohydrate crystals) had small cholesterol gallstones and 3 (2 with cholesterol monohydrate crystals and 1 with calcium bilirubinate granules) had signs strongly suggestive of the recent migration of gallstones. One patient refused operation, but minute pigment stones were found to be associated with calcium bilirubinate granules at duodenal intubation. In the other 31 patients, bile contained neither cholesterol monohydrate crystals nor calcium bilirubinate granules. They were not operated on and were followed up with repeated investigations for 12 to 24 months.(ABSTRACT TRUNCATED AT 250 WORDS)

This work was supported by CIERM (Centre Inter Etablissement de Resonance Magn#{233}tique), H#{244}pital de Bic#{234}tre, 94270 Le Krernlin-Bic#{232}tre, France. I Department of Radiology, Centre H#{244}pital lhiiversit#{233} Henri Mondor, 51 Avenue du Mar#{233}chal de Lattre de Tassigny, 94010 Cr#{233}teil C#{233}dex, France. Address reprint requests to D. Mathieu. 2 Department of Hepatology, Centre H#{244}pital Universit#{233} Henri Mondor, 51 Avenue du Mar#{233}chal de Lattre de Tassigny, 94010 Cr#{233}teil C#{233}dex, France. 635

The therapeutic management of patients with idiopathic hemochromatosis (IH) implies the evaluation of excess hepatic iron. This work was undertaken to confirm the value of computed tomography for the assessment of liver iron overload in such patients and to evaluate this technique during the course of treatment by phlebotomy. The study included 24 patients with initially untreated IH and 7 patients previously treated by phlebotomy for 10 months to 7 years. Follow-up was obtained in 15 subjects. In patients with untreated IH, liver attenuation coefficient (LAC) was always markedly increased (92.4 +/- 7.1 Hounsfield units) as compared with LAC of subjects with normal liver (60.2 +/- 3.1 Hounsfield units) and that of patients with chronic liver disease (53.8 +/- 4.8 Hounsfield units), and was found to be specific for liver iron overload. LAC decreased progressively during phlebotomy, and this diminution was correlated with the amount of mobilized iron (r = 0.79, p less than 0.001); it returned to normal values only after complete removal of iron overload. LAC was closely correlated with liver iron concentration (r = 0.83, p less than 0.001), better than usual biochemical parameters, especially serum ferritin (r = 0.70, p less than 0.01). This study confirms that the determination of LAC on computed tomography provides a reliable index of hepatic iron stores in patients with IH, without requiring a liver biopsy, and shows that this noninvasive method is of particular interest for the follow-up of patients treated by phlebotomy.

Gilbert's syndrome was diagnosed in 37 patients with unconjugated hyperbilirubinaemia without overt haemolysis or structural liver abnormality, who had a marked reduction in hepatic bilirubin UDP-glucuronosyltransferase activity (B-GTA) (as compared with that of 23 normal subjects). No significant correlation existed in these patients between serum bilirubin level and the values of B-GTA, thus suggesting that factors other than a low B-GTA must influence the degree of hyperbilirubinaemia in Gilbert's syndrome. Studies of 51Cr erythrocyte survival and 59Fe kinetics in 10 unselected patients demonstrated slight haemolysis in eight, whereas mild ineffective erythropoiesis was suggested in all from a low 24-hour incorporation of radioactive iron into circulating red cells. This overproduction of bilirubin resulting from mild haemolysis and perhaps dyserythropoiesis might reflect only an extreme degree of the normal situation. It certainly contributes to the hyperbilirubinaemia of Gilbert's syndrome and may play a major role in the manifestation of this condition.

An ascitic fluid pH less than or equal to 7.31 has been advanced as being the best index in the early diagnosis of spontaneous bacterial peritonitis in cirrhotic patients. In order to test the validity of this criteria, 55 patients with alcoholic cirrhosis and ascites were studied. In each patient, arterial blood and ascitic fluid samples were analysed for pH, PCO2, total CO2 and PO2, and the pH gradient between blood and ascites was calculated. White blood cell and polymorphonuclear cell counts were determined in ascitic fluid, and cultures of ascites were done under aerobic and anaerobic conditions. Twelve patients had a culture proven spontaneous bacterial peritonitis. Their mean ascitic fluid pH (+/- SD) was 7.38 +/- 0.09 (range 7.21-7.49) and differed significantly (p less than 0.05) from that found in patients without spontaneous bacterial peritonitis: 7.44 +/- 0.06 (range 7.34-7.6.3). A marked overlap was observed, however, between the two groups, and only three out of the 12 patients with spontaneous bacterial peritonitis had an ascitic fluid pH less than or equal to 7.31. The pH gradient was 0.10 +/- 0.08 (range -0.01 to +0.28) in the spontaneous bacterial peritonitis group, as compared with 0.02 +/- 0.04 (range -0.09 to +0.12) in the sterile group (p less than 0.01), but a marked overlap was also noted between the two groups. In the spontaneous bacterial peritonitis group, the polymorphonuclear count was 3588 +/- 3849/microliter (range 60-11 776) versus 41 +/- 138/microliter (range 0-813) in the sterile group (p less than 0.0001). All but one patient in the spontaneous bacterial peritonitis group and only two patients in the sterile group had over 250 polymorphonuclear/ microliter. Thus, in our experience, neither the ascitic fluid pH nor the pH gradient values accurately discriminated the individual patients with and without spontaneous bacterial peritonitis. A polymorphonuclear count less than 250/ microliter remained the best criteria for the diagnosis of spontaneous bacterial peritonitis in cirrhotic patients, before having the results of ascitic fluid cultures.

Hepatic lipocytes are sinusoidal cells in close contact to endothelial cells. They proliferate, switch to a fibroblastic phenotype and synthetize collagen during hepatic fibrosis. Since it is known that vascular endothelial cells can influence the proliferation of neighboring cells such as smooth muscle cells, we investigated the role of hepatic endothelial cells on the growth of lipocytes and Balb/c 3T3 fibroblasts. Concentrated conditioned medium from endothelial cells inhibited both [3H]thymidine incorporation and actual growth of lipocytes and Balb/c 3T3 fibroblasts. The inhibition was lost when conditioned medium was treated with heat or trypsin, or when medium was conditioned in the presence of cycloheximide. We conclude that hepatic endothelial cells secrete a proteic growth inhibitor for lipocytes and hepatic Balb/c 3T3 fibroblasts. This inhibitor could be of importance in limiting lipocyte proliferation in the liver and possibly in preventing hepatic fibrosis.

The present study was undertaken to determine whether or not spironolactone could induce drug-metabolizing enzymes in patients with ascites due to alcoholic cirrhosis of the liver. The disposition of antipyrine was therefore studied in 15 patients with cirrhosis before and after 13 days of spironolactone treatment (200 mg/day) and in 15 comparable patients not treated by spironolactone. In spironolactone-treated patients, a 40% increase in elimination rate constant of antipyrine (P less than 0.01) was due both to a decrease in apparent volume of distribution by 11% (P less than 0.01) and to a 20% rise in metabolic clearance rate (P less than 0.01). In control patients treated with furosemide or by paracentesis no significant change in metabolic clearance rate of antipyrine was found. As with spironolactone, furosemide decreased the apparent volume of distribution (9%, P less than 0.05) in proportion to the loss in body weight. Since spironolactone treatment was not associated with improvement in the fractional clearance of bromosulfophthalein, nor in serum albumin, prothrombin time, and factor V, the data are compatible with the idea that even in cirrhosis the hepatic drug-oxidizing enzymes were induced by spironolactone. Thus, the consequences of enzyme induction should be considered when spironolactone is associated with other drugs for the treatment of cirrhotic patients.

The authors have studied the influence of rose bengal, and, by comparison, of uranin, on choleresis in the rabbit. Uranin induced an increase in bile flow and a decrease in bile salt concentration in bile. These results are consistent with an osmotic mechanism, as proposed by Sperber (1959). By contrast, rose bengal induced a marked decrease in bile flow and an increase in bile salt concentration. In rose bengal-treated animals (30 μmoles per kilogram body weight intravenously), bile flow was 65% lower and bile salt concentration 95% higher than in control animals studied during the same periods. The relationship between bile flow and bile salt excretion is consistent in the rabbit, with the existence of a bile salt-independent secretion of 60 mg per minute per kilogram body weight, that is, 60% of bile flow. In rose bengal-treated animals, this fraction was only 15 mg per minute per kilogram body weight. It is concluded that rose bengal inhibits the bile salt-independent fraction. Rose bengal, as other phtalein dyes, is excreted by the liver cells. Its effect on the bile salt-independent fraction suggests that this fraction is secreted by the hepatocytes.

The authors report two cases of fracture of a biliary endoprosthesis inserted endoscopically to treat malignant biliary obstruction. To their knowledge, such a complication has never been reported. In both cases, the fracture was revealed by cholangitis which occurred three and eight months after insertion. Successful treatment was achieved by replacement of the biliary stent.

Two successive hepatic arterial embolizations were performed in a patient with hypercalcemia secondary to hepatocellular carcinoma. The first procedure was performed on an emergency basis due to a cardiovascular episode and was immediately beneficial. The second procedure, performed five months later for a recurrence, was effective in 3 days; 13 months later, there had been no recurrence.

THERE are two well documented mechanisms of bile formation: active secretion of bile acids by hepatocytes and subsequent osmotic diffusion of water and electrolytes1,2, and secretion of an electrolyte solution by the biliary epithelium, chiefly in response to administration of secretin2,3. Several observations, however, suggest the possibility of an additional mechanism, especially in rodents, which have a very high bile flow. For example, bile flow remains high after bile salt depletion following bile duct cannulation in the rat4; and bile flow is high in the isolated perfused rat liver, with no entero-hepatic circulation of bile salts5. To explain these observations, inorganic ion secretion by the hepatocyte has been postulated6,9. Previous studies here gave results consistent-with a large hepatocytic bile salt independent fraction in the rabbit amounting to 60 µl./min/kg body weight, that is 60 per cent of the spontaneous bile flow in this animal7,8. To test the hypothesis that this fraction may result from active inorganic ion transport, we studied the effects on bile formation of known inhibitors of active sodium transport.

• The vascular lesions of the liver described in association with drug and toxic substance exposure are reviewed, with special reference to the abnormalities of the following: (1) the hepatic venous efferent system (ie, large and small hepatic vein obstruction); (2) the sinusoids (ie, sinusoidal dilatation, peliosis, perisinusoidal fibrosis); (3) the portal vein and its branches; and (4) the hepatic arterial tree. Drug-induced vascular tumors of the liver and vascular changes observed within nonvascular tumors of the liver are also envisaged. Finally, the pathophysiologic mechanisms of all these lesions are considered. The awareness of this type of hepatotoxicity and the knowledge of its epidemiology seem to be essential for both its early detection and prevention. ( Arch Intern Med 1983;143:495-502)

• We report an association between idiopathic hypereosinophilic syndrome and obstruction of the hepatic veins (Budd-Chiari syndrome). Budd-Chiari syndrome was assessed by liver biopsy and hepatic phlebography and documented by computed tomography. Postmortem examination revealed fibrous occlusion of the hepatic venous tree, as well as fibrosis of the endocardium and of myocardial and pulmonary vessels. To our knowledge, the association between idiopathic hypereosinophilic syndrome and Budd-Chiari syndrome has never previously been reported. Since it has been suggested that hypereosinophilia might cause endothelium damage, a link between these two entities is postulated. ( Arch Intern Med 1985;145:751-753)

In 11 patients, sera displaying a ringlike nuclear immunofluorescent staining on sections of rat liver tissue were shown by Western blotting to contain antibodies to lamins. Sera from 8 patients contained autoantibodies reacting with lamin B, whereas sera from the other 3 patients reacted with lamins A and C. All patients (9 women and 2 men) had a chronic autoimmune disorder, which rarely fulfilled the usual criteria for a diagnosis of systemic lupus erythematosus. The disorder was characterized by acute or chronic (active or granulomatous) hepatitis; steroid-responsive blood cytopenia, often associated with a circulating anticoagulant, or anticardiolipin antibodies, or both; and cutaneous leukocytoclastic angiitis or probable brain vasculitis. Eight patients had at least two of these three conditions. Antilamin autoantibodies may thus be a marker for an unusual subset of autoimmune diseases.

A woman, who had lived in equatorial Africa, complained of pain in the right hypochondrium. Roentgenography revealed a calcified mass in the liver. The mass was removed surgically and was found to contain eggs ofFasciola gigantica, a fluke that rarely infests humans. The roentgenographic appearance of the lesion (its large size, uneven density, and polycyclic configuration) was distinct from that produced by other causes of hepatic calcification.

Hepatocellular adenomas are benign tumors that can be difficult to diagnose. To refine their classification, we performed a comprehensive analysis of their genetic, pathological, and clinical features. A multicentric series of 96 liver tumors with a firm or possible diagnosis of hepatocellular adenoma was reviewed by liver pathologists. In all cases, the genes coding for hepatocyte nuclear factor 1alpha (HNF1alpha) and beta-catenin were sequenced. No tumors were mutated in both HNF1alpha and beta-catenin enabling tumors to be classified into 3 groups, according to genotype. Tumors with HNF1alpha mutations formed the most important group of adenomas (44 cases). They were phenotypically characterized by marked steatosis (P < 10(-4)), lack of cytological abnormalities (P < 10(-6)), and no inflammatory infiltrates (P < 10(-4)). In contrast, the group of tumors defined by beta-catenin activation included 13 lesions with frequent cytological abnormalities and pseudo-glandular formation (P < 10(-5)). The third group of tumors without mutation was divided into two subgroups based on the presence of inflammatory infiltrates. The subgroup of tumors consisting of 17 inflammatory lesions, resembled telangiectatic focal nodular hyperplasias, with frequent cytological abnormalities (P = 10(-3)), ductular reaction (P < 10(-2)), and dystrophic vessels (P = .02). In this classification, hepatocellular carcinoma associated with adenoma or borderline lesions between carcinoma and adenoma is found in 46% of the beta-catenin-mutated tumors whereas they are never observed in inflammatory lesions and are rarely found in HNF1alpha mutated tumors (P = .004). In conclusion, the molecular and pathological classification of hepatocellular adenomas permits the identification of strong genotype-phenotype correlations and suggests that adenomas with beta-catenin activation have a higher risk of malignant transformation.