Three hundred patients with primary Sjögren's syndrome (pSS) were investigated for liver involvement using clinical, biochemical, immunological and histological data. Seven per cent of patients showed evidence of liver disease either subclinical (2%) or asymptomatic (5%) with elevated liver enzymes. In 6.6% of patients antimitochondrial antibodies (AMA) were detected by immunofluorescence and 27% of pSS patients showed antibodies to pyruvate dehydrogenase (a-PDH) using ELISA. AMA-positive patients were further investigated with transcutaneous liver biopsy. Ninety-two per cent of patients with AMA showed liver involvement with features of chronic cholangitis similar to stage I primary biliary cirrhosis. It is concluded that liver involvement in pSS patients is rare and subclinical with histological features predominantly of stage I primary biliary cirrhosis. AMA is the most sensitive indicator of underlying liver pathology in pSS patients.

To determine the prevalence of alpha 1-antitrypsin deficiency in patients with cirrhosis or chronic active hepatitis, we performed a five-year prospective study of liver-biopsy specimens from 1055 adults. Thirty-four patients whose specimens contained hepatocyte inclusions characteristic of the deficiency were phenotyped, and 25 had phenotype MZ (2.4 per cent). The distribution of patients with this phenotype among the 185 patients with cirrhosis diagnosed histologically was three of 84 patients with alcoholic cirrhosis (3.5 per cent), seven of 34 with non-B chronic active hepatitis (20.5 per cent), six of 28 with cryptogenic cirrhosis (21 per cent), and one of 39 with other kinds of cirrhosis (2.6 per cent). The increased prevalence of MZ in patients with cryptogenic cirrhosis and with chronic active hepatitis is highly significant (P < 0.001). Because serum levels of alpha 1-antitrypsin may be unreliable for identification of the subgroup of patients with chronic active hepatitis or cryptogenic cirrhosis, analysis of serum for the MZ phenotype and meticulous examination of biopsy specimens may be necessary.

Real-time ultrasound did not improve diagnostic yield or result in fewer complications. Marking the puncture site seems adequate and has the practical advantage that it takes up less of the radiologists' time.

SUMMARY The distribution of Mallory body antigens JMB1 and 2 was examined in 82 human fresh diagnostic needle liver biopsies and 28 necropsies by the indirect immunoperoxidase technique using 2 monoclonal antibodies (anti-JMB1 and 2) against Mallory bodies. The JMB1 antigen was detectable in bile duct epithelium and in hepatocytes of histologically normal livers. It was also found in all Mallory bodies in various hepatic disorders. This antigen was markedly increased in the cytoplasm of all liver cells in acute alcoholic hepatitis superimposed on alcoholic cirrhosis, in most cases of acute alcoholic hepatitis, and in severe fatty infiltration of the liver with or without Mallory body formation. Mallory bodies contained this antigen but the cytoplasm of Mallory body containing cells lacked JMB1. In normal liver the JMB2 antigen was localised on the cytoplasmic intermediate filament network of hepatocytes and bile duct epithelium; and almost all Mallory bodies also contained this antigen but the adjacent cytoplasm of these cells lacked JMB2. In severe alcoholic liver disease these antigens could not be detected in large zones of hepatocytes even when these hepatocytes did not contain Mallory bodies. It is evident that there is disorganisation of intermediate filament constituents in severe alcoholic liver disease.

A case of Erdheim-Chester disease which affected the epiphysis and showed evidence of systemic disease is presented. Clinical and histopathological similarities with other forms of disseminated Langerhans' cell histiocytosis are noted, particularly reaction of infiltrating histiocytes for SlOO and HLA-DR.