A 20-year prospective study of cirrhosis.

A total of 512 people from a defined population in west Birmingham served by a district general hospital were found to have cirrhosis in the period 1959-76. The annual incidence rose from 5.6 per 100 000 to a peak of 15.3 per 100 000 in 1974. This was due to an increase in the incidence of alcoholic cirrhosis, which in the last six years accounted for two-thirds of cases. The proportion of patients with decompensated cirrhosis when first seen (65%) did not alter during the 18 years. This was reflected in a death rate of 78% among the 468 patients traced up to the end of 1978. Liver failure, hepatoma, and gastrointestinal haemorrhage accounted for almost three-quarters of the deaths. The proportion of patients who survived for five years was 36% for alcoholic cirrhosis, 14% for cryptogenic cirrhosis, and 60% for chronic active hepatitis, and these figures too remained constant throughout the 18 years. Modern methods of treatment for decompensated cirrhosis did not improve prognosis and only abstention in patients with alcoholic cirrhosis had a beneficial effect on survival. Since alcoholic cirrhosis is now the most common form of the disease it is important to recognise those at risk and to encourage abstinence; also, more efforts are needed to identify the causes of cryptogenic cirrhosis. Whatever the cause, cirrhosis needs to be diagnosed before decompensation occurs, if treatment is to have any effect.