SUMMARY Lymphocytes from 39 patients with HBsAg positive chronic liver disease were incubated with their own hepatocytes to investigate mechanisms of lymphocyte-mediated liver damage. Cytotoxicity was significantly increased in 46% overall, and in 73% of those with chronic active hepatitis. Unlike HBsAg negative chronic active hepatitis where only non-T cells were cytotoxic, HBsAg positive patients had both cytotoxic T and non-T cells. A purified liver membrane complex (LSP) and aggregated IgG both blocked non-T cytotoxicity without affecting T cell cytotoxicity; this suggests that the former is probably an antibody-dependent cell-mediated reaction against normal membrane components. This was confirmed in preliminary studies which demonstrated that preincubation of hepatocytes with the F(ab)' fragment of an anti-human IgG reduced non-T lymphocyte cytotoxicity. T-cell cytotoxicity was restricted to HBeAg-positive

A coagulation screen has been performed on 12 patients with acute liver failure. Six died and six recovered. All six fatal cases developed a haemorrhagic state with haemostatic failure. An attempt has been made to delineate the various mechanisms for the production of the coagulation defect. The significance of the different haematological parameters in assessing prognosis has been assessed. The study emphasizes the importance of the synthetic ability of the liver in determining survival prospects. A good correlation between the factor-VII level, which is a guide to liver synthesis, and recovery has been shown. The value of a specific factor-VII assay in acute liver failure appears considerable. Where this assay cannot be performed the clot opacity fibrinogen technique provides a reasonable guide to the prognosis. The presence or absence of DIC was not a determinant factor in survival in this series.

669The results on the overnight urine specimens in all the subjects are shown in the figure. The mean overnight 1 1-hydroxycorticosteroid levels and the standard deviations in the control, obese, and hirsute groups were similar, being 119 +47 nmol (43 + 17 [g), 121 ± 50 nmol (44 ± 18 Lg), and 138 +55 nmol (50 +20 :'g) respectively. The highest value of 271 nmol (98 ug) occurred in an obese patient. By contrast, only one of the patients with Cushing's syndrome had a urinary 11-hydroxycorticosteroid excretion of less than 276 nmol (100 ug) during the night. On the night in question her midnight plasma 11-hydroxycorticosteroid was normal at 189 nmol/l (6-8 ,ug/100 ml), and her overnight urine contained 254 nmol (92 ,tg); on a subsequent occasion a repeat overnight collection gave a raised value of 389 nmol (141 ILg). The diagnosis of bilateral hyperplasia was confirmed at operation. The overnight specimens from all the other patients with Cushing's syndrome contained more than 276 nmol (100 bLg) of 1 1-hydroxycorticosteroids, the highest containing 2663 nmol (965 ,ug). DiscussionOur results confirm that there is a circadian rhythm in the urinary excretion of adrenal steroids in normal people, the lowest levels occurring during the night. In patients with Cushing's syndrome this rhythm is not so apparent, and the nocturnal excretion is much higher.2 3 We therefore thought that patients with adrenocortical overactivity might be identified by their higher overnight urinary excretion of 1 1-hydroxycorticosteroids, and this was found to be so.Our data were obtained solely from women because only one man with Cushing's syndrome was admitted during the study. The overnight 11-hydroxycorticosteroid level in this man with an adrenal carcinoma was grossly raised at 5382 nmol (1950 [Lg).Thus this test is probably applicable to both men and women.Clinical suspicion of Cushing's syndrome is often aroused by obesity and hirsutism in women, particularly if it is associated with hypertension or diabetes mellitus, but it is not practicable to investigate all such patients in hospital. This simple screening test, requiring only the collection of an overnight urine specimen for 1 1-hydroxycorticosteroid estimation, has proved useful in selecting those outpatients who require more extensive investigation. Our data indicate that levels over 276 nmol (100 ,ug) that were examined by electron microscopy aggregates of HBsAg and HBsAb were seen. In contrast, HBsAb was never detected by RIA in those with non-fulminant hepatitis, and in only one serum specimen (5%) were aggregates seen on electron microscopy. A significant sex difference between fulminant and non-fulminant hepatitis was observed, 65%' of patients with fulminant hepatitis but only 15% of patients with non-fulminant hepatitis being women (P <0 01).

SummaryThe clinical relevance of the e antigen-antibody system was investigated in 61 people persistently positive for hepatitis-B surface antigen, including 22 healthy carriers. The e antigen was not detectable in any of the healthy carriers, whereas it was found in 15 out of 28 patients with chronic aggressive hepatitis and two out of 11 with chronic persistent hepatitis. Its presence therefore indicates chronic liver disease but its absence does not exclude it. It may prove to be a particularly useful prognostic aid in chronic persistent hepatitis, since one of the two patients in whom it was found later developed aggressive hepatitis. In contrast, e antibody is of little diagnostic help, for, though it was found mostly in healthy carriers (18; 820 ), it was also detectable in 9 (230o) of the patients with chronic hepatitis.In 13 (76°' ) of the patients positive for e antigen Dane particles were seen on electron microscopy, but these were also present in 5 (19%) of the patients positive for e antibody. These findings are consistent with other evidence suggesting that e antigen is not a surface component of the Dane particle, but rather an independent soluble protein manufactured by the host in response to infection with the hepatitis-B virus.

Plasma from patients with both acute and chronic liver disease has been examined for evidence of acquired dysfibrinogenaemia, using electrophoretic methods and coagulation tests. An examination of isolated fibrins upon SDS polyacryamide gel electrophoresis failed to demonstrate any molecular or structural defect associated with the polypeptide chains of the patients' fibrinogen or fibrinogen derivatives produced by thrombin or plasmin. However, purified fibrin monomers isolated from plasma using both Reptilase and thrombin exhibited delayed polymerization rates and the occurrence of acquired dysfibrinogenaemia in liver disease is therefore confirmed.