Search

Purpose: To identify the diseases that are associated with a high plasma concentration of vitamin B12 and to measure the strength of this association., Patients and Methods: Retrospective study including all admissions between 1st May, 2005 and 30th April, 2008 in the UMAG pole departments (emergency, internal medicine, acute geriatrics and medical intensive care) with a test for plasma vitamin B12. The association between each of medical information system codes (solid tumors, malignant hematologic process, and renal disease) and a high or low vitamin B12 concentration was measured by odds ratios (OR) from logistic models taking into account repeated admissions, with adjustment for age and the weighted Charlson index., Results: Among 3702 admissions, 12% had a B12 more than 820pg/ml, 10.4% a B12 less than 180pg/ml and 77.6% a normal B12 concentration. After adjustment for age and the weighted Charlson index, high concentration of vitamin B12 was associated with interstitial renal diseases (OR 2.7; 95% CI: [1.7-4.2]), and cirrhosis or hepatitis (OR 4.3; [2.9-6.4]). After additional adjustment for these parameters, it was still associated with tumors (OR 1.8; [1.2-2.6]), malignant hematologic diseases (OR 2.1; [1.3-3.5]), metastasis (OR 2.9; [1.5-5.9]), liver metastasis (OR 6.2; [2.7-14.5]), liver carcinoma (LC) (OR 3.3; [1.1-10.4]), liver tumors other than LC (OR 4.7; [1.2-17.9]) and lymphoma (OR 3.2; [1.6-6.4]) but not with myeloma (OR 1.9; [0.6-1.4]). Low concentration of B12 was associated with myeloma (OR 2.9; [1.3-6.6])., Conclusion: Finding a high plasma concentration of vitamin B12 should lead to a systematic search for a hepatic disease or a tumor, and particularly for a hepatic localization of a tumor., (Copyright © 2012 Société nationale française de médecine interne (SNFMI). Published by Elsevier SAS. All rights reserved.)

Background/aims: The incidence of hepatocellular carcinoma is higher in males, presumably due to the influence of sex steroids. Therefore, to further assess the role of sex steroids in the occurrence of hepatocellular carcinoma, this study investigated the predictive value of different sex hormones and their binding protein., Methods: Among 101 male patients with cirrhosis included in a prospective screening study, 29 developed hepatocellular carcinoma. We assessed the predictive value of 17 clinico-biological and 4 serum hormonal variables collected at enrollment, by the log-rank test and the Cox model., Results: Age (p = 0.003), bilirubin (p = 0.04), sex-hormone-binding-globulin (p = 0.006) and albumin (p = 0.08) were predictive using the log-rank test, while estradiol and total and free testosterone were not. The Cox model showed age (p = 0.0003; relative risk = 7.52), sex-hormone-binding globulin (p = 0.001, relative risk = 3.37) and albumin (p = 0.02, relative risk = 2.94) as the most predictive parameters., Conclusion: We conclude that high serum sex-hormone-binding-globulin levels have an independent predictive value for the occurrence of hepatocellular carcinoma. Serum sex-hormone-binding-globulin could be used to define patients at high risk for hepatocellular carcinoma and could hypothetically play a role in hepatocarcinogenesis.

Upper gastrointestinal bleeding (GIB) is a major complication in cirrhotic patients. Endoscopy and oesophageal sclerosis are reference treatments and must be done as soon as possible. However, such treatment is not possible unless the patient is admitted to hospital. In a prospective, randomised, double-blind trial, we compared the efficacy of terlipressin combined with glyceryl trinitrate (TER-GTN), administered as early as possible to 76 patients with cirrhosis who had active GIB (84 bleeding episodes). Infusion was done at the patient's home by the physician on the emergency team (a mobile intensive care unit) if the patient had GIB and a history and clinical signs of cirrhosis. Patients received either an intravenous injection (1 to 2 mg) of TER-GTN or a double-placebo injection, and then another injection at 4 and 8 h. Control of bleeding, rebleeding, and mortality rate at days 15 and 42 were evaluated. In most patients, endoscopy confirmed the rupture of oesophageal varices (75.7%). Bleeding control was significantly better in the TER-GTN group (n = 41) than in the double-placebo group (n = 43) (p = 0.034). Mortality due to bleeding episodes was significantly lower in the TER-GTN group than in the double-placebo group at day 15 (p = 0.035) and at day 42 (p = 0.06). There were no serious side-effects. Early administration of TER-GTN lowers the deleterious consequences of prolonged hypovolaemia on the hepatic function of these patients.

Screening is widely used to detect early hepatocellular carcinoma in Asian patients with cirrhosis. Its effectiveness in Caucasian patients has been suggested, but remains to be proven. Therefore we prospectively studied 118 French patients (68 males, 50 females, age 55 +/- 12) with Child-Pugh A or B cirrhosis (alcoholic in 82) and without detectable hepatocellular carcinoma. The screening program consisted of ultrasound examination of the liver and determination of blood alpha-fetoprotein and des-gamma-carboxyprothrombin levels every 6 months. The median follow up was 36 months (range 4-48). Only four patients were lost to follow up. Fourteen hepatocellular carcinomas were detected, in six cases by ultrasonography alone, in four by alpha-fetoprotein alone, in three by ultrasonography and alpha-fetoprotein and in one case by ultrasonography and des-gamma-carboxyprothrombin, but never by des-gamma-carboxyprothrombin alone. The tumor presented as a unique nodule in nine patients. The tumor was less than 3 cm in diameter without portal thrombosis or metastasis in three cases. Surgery was performed in only one case. In this study, the annual incidence of hepatocellular carcinoma was high (5.8%), but the screening methods used did not effectively identify potentially resectable tumors in Caucasian patients with cirrhosis.

We report the case of a 82-year-old man, living in institution, hospitalized for a severe anaemia due to scurvy. Scurvy is rare in Occident. A multifactorial anaemia is usually associated with scurvy, but is rarely symptomatic. Alcoholism favours scurvy and anaemia. Treatment consisted of parenteral vitamin C supplementation associated with blood transfusion.

Viral infection has often been suggested as a possible cause of Sjögren's syndrome or chronic lymphocytic sialadenitis, and Epstein-Barr virus has been found in the salivary glands of patients with this condition. After we had noted Sjögren's syndrome in several patients infected with hepatitis C virus (HCV), a virus also excreted in saliva, we set up a prospective study to investigate the association of chronic lymphocytic sialadenitis, with or without symptoms, to chronic HCV liver disease. The histological appearances of labial salivary glands in patients with proven HCV hepatitis or cirrhosis were compared with those in dead controls. Histological changes characteristic of Sjögren's syndrome were significantly more common in HCV-infected patients (16 of 28, 57%) compared with controls (1 of 20, 5%). Focal lymphocytic sialadenitis characteristic of Sjögren's syndrome (though only 10 patients had xerostomia and none complained of xerophthalmia) appears to be common in patients with chronic HCV liver disease; if this association is confirmed, identification of the underlying mechanism may improve our understanding of both disorders.

The aim of this study was to compare serum N-terminal peptide of type III procollagen to aminotransferases and gamma-globulins as a marker for histological activity in patients with chronic hepatitis and to assess the role of type I collagen, a new serum marker, as a marker of fibrosis in these patients. Sixty patients with biopsy-proven chronic hepatitis were included in this study. Liver disease was virus B-related in 29, autoimmune in five, drug-induced in five, and of unknown etiology in 21. Each biopsy was independently assessed by two liver pathologists. Two histological scores, a score of activity and a score of fibrosis, were established. Serum N-terminal peptide of type III procollagen and type I collagen were assayed by liquid phase RIA. Significant correlations were noted between serum N-terminal peptide of type III procollagen and scores of activity (r = 0.70, p less than 10(-4)) and fibrosis (r = 0.45, p = 0.0005), and between serum type I collagen and scores of activity (r = 0.46, p = 0.0004) and fibrosis (r = 0.67, p less than 10(-4)). When the correlation between scores of activity and fibrosis (r = 0.52, p = 10(-4)) was considered by partial correlation, serum N-terminal peptide of type III procollagen was correlated with the score of activity (r = 0.63, p less than 10(-3)) but not with the score of fibrosis, and serum type I collagen was correlated with the score of fibrosis (r = 0.58, p less than 10(-3)), but not with the score of activity.(ABSTRACT TRUNCATED AT 250 WORDS)