Liver disease in ulcerative colitis: an epidemiological and follow up study in the county of Stockholm.
In an epidemiological study of the incidence of ulcerative colitis (UC) in the county of Stockholm between 1955 and 1979, 1274 patients with UC were discovered. Almost all these patients had regularly been investigated with liver function tests; 142 (11%) of them showed signs of hepatobiliary disease. A follow up study on all 142 patients with abnormal liver function and UC was made between 1989 and 1991 to evaluate the cause of the liver abnormality and to find out if the liver disease had affected the survival rates. At follow up, eight patients were reclassified as having Crohn's disease, 60 had developed normal liver function as judged from test results, while the remaining 74 still had signs of hepatobiliary disease. The most common explanation for a transient abnormality in liver function was active colitis. The temporary signs of liver injury were not associated with changes in survival rates for these patients. Infections, especially those because of hepatitis B and C virus transmitted by blood transfusions accounted for the abnormalities in liver function in 21 patients, nine of which had a chronic, but non-fatal course. Twenty nine (2.3%) of the patients developed primary sclerosing cholangitis (PSC), and 12 of them died during the study period four because of cholangiocarcinoma and eight because of hepatic failure; one patient had a transplant. The estimated median time of survival from the first presentation of evidence of a liver function, compatible with the diagnosis of PSC, to death or liver transplantation was 21 years. A comparison of survival rates in patients with UC and patients with UC and concurrent PSC showed, a significant reduction in survival in the PSC group (p<0.0001). The number of patients with UC who developed PSC remained constant during the study period. Thus, although evidence of abnormal liver function is a common finding in UC, a spontaneous return to normal levels is common. In this study, which did not have a selection bias, the median time of survival among PSC patients was far longer than previously described although development of PSC among patients with UC does significantly reduce the estimated median time of survival.