Chronic unexplained hypertransaminasemia may be caused by occult celiac disease.

In a subset of patients attending liver units, a chronic increase in serum transaminases may remain of undetermined cause despite thorough investigations. On the other hand, elevated levels of serum transaminases have been reported in about 40% of adult celiac patients. To evaluate the prevalence of subclinical celiac disease in patients with chronic unexplained hypertransaminasemia in comparison with that in the general population (0.5%), 140 consecutive patients with chronic increases of serum transaminases levels of unknown cause were tested for antigliadin and antiendomysium IgA antibodies. All patients with positive antibody tests were offered upper gastrointestinal endoscopy with distal duodenal biopsy. Thirteen patients (9.3%, 95% confidence interval 5. 0-15.4) had positive antigliadin and antiendomysium antibodies. The prevalence of antibodies was 17% in women and 5.4% in men (8/47 vs. 5/93 respectively; relative risk 3.2, 95% confidence interval 1.1-9. 1). Distal duodenal biopsy performed in all but one of the patients showed mild villous atrophy with increased intraepithelial lymphocytes in three cases, subtotal villous atrophy in six, and total villous atrophy in three. The prevalence of celiac disease in the patient group was significantly higher than that in the general population (P <.001) with a relative risk of 18.6 (95% confidence interval 11.1-31.2). On the basis of the present findings, screening for celiac disease is an important tool in the initial diagnostic work-up of patients with chronic unexplained hypertransaminasemia.

Diversity of antinuclear antibody responses in hepatocellular carcinoma.

BACKGROUND/AIMS: The development of antinuclear antibodies (ANA) in malignancies has been described but its mechanism is still not understood. The aim of this study was to examine ANA specificities in hepatocellular carcinoma to further understand autoimmunity in cancer. METHODS: Two hundred and four hepatocellular carcinoma patients were compared with 68 chronic hepatitis C, with 126 chronic hepatitis B and with 30 alcoholic liver cirrhosis patients, as well as with 87 healthy donors. Indirect immunofluorescence, immunoblotting, and immunoprecipitation were used to study ANA reactivities. RESULTS: Hepatocellular carcinoma had a significantly higher frequency of ANA using HEp-2 cells as substrate (31%) than chronic hepatitis C (10%), chronic hepatitis B (9.5%), alcoholic liver cirrhosis (10%) or healthy donors (4.5%). A great diversity of ANA specificities was found in hepatocellular carcinoma. Three hepatoma sera had antibodies that co-localized with non-snRNP splicing factor SC35, suggesting that the antigenic targets might be involved in mRNA splicing. We identified antibodies to two known nuclear autoantigens: fibrillarin and p330d/CENP-F. These autoantigens are involved in the 5' processing of precursor ribosomal RNA transcripts and in mitotic functions, respectively. CONCLUSIONS: Diversity was found in the autoantibody specificity, in contrast to the specific subsets of autoantibodies seen in several systemic rheumatic autoimmune diseases. Our data suggest that immune response in hepatocellular carcinoma targets important proteins involved in cellular biosynthetic or proliferative functions.

Portal hypertension and iron depletion in patients with genetic hemochromatosis.

Clinically, portal hypertension has been considered to be less common and less severe in patients with cirrhosis resulting from iron overload in homozygotes for genetic hemochromatosis than in patients with cirrhosis of other causes. To characterize the prevalence and progression of portal hypertension in genetic hemochromatosis (GH), 120 cirrhosis and iron-overloaded patients were compared with a control group of 120 patients with postnecrotic cirrhosis (PNC) who were matched for gender, age, Child's class, and alcohol abuse. Gastroesophageal endoscopy and abdominal ultrasonography were performed at diagnosis and repeated every 12 months and every 6 months, respectively, to evaluate the presence and severity of varices, the caliber of the portal vein and its collaterals, and splenic size. At diagnosis a similar frequency of varices was observed in patients with GH (25%) and in PNC (24%), as well as of portal vein abnormalities and spleen enlargement. During the follow-up period, all but two of the patients with GH were treated by phlebotomy and depleted of excess iron. After a mean of 6 +/- 4.3 (SD) years of observations (range, 2 to 10 years), varices were improved or completely reversed in 26% of patients with cirrhosis and GH but in only 5% of those with PNC (P < .01). Bleeding from varices was observed in only one patient with GH but in five patients with PNC. Of 22 patients with GH in whom portal hypertension was unmodified or worsened, 16 had coexistent hepatic viral infection.(ABSTRACT TRUNCATED AT 250 WORDS)