Natural histories of hepatitis C virus infection in men and women simulated by the Markov model

The Markov model was introduced to simulate natural histories of hepatitis C virus (HCV) infection in men and women. The data set was constructed on 942 HCV carriers who were examined at least once a year without receiving antiviral therapies. Based on 2,251 patient-year data, the probabilities of transition between any two of the four clinical states, i.e., asymptomatic carrier state, chronic hepatitis, liver cirrhosis and hepatocellular carcinoma in 1 year were calculated. Hepatocellular carcinoma was defined as the absorbing state from where no transitions occur. Probability matrices thus obtained on six each subsets of HCV infection (asymptomatic carrier state, chronic hepatitis and liver cirrhosis in men and women) in their forties, fifties, and sixties, were used to simulate long-term outcomes of HCV infection. Male asymptomatic carriers aged 40 years were expected to retain the asymptomatic carrier state in 2.6%, evolve into chronic hepatitis in 48.4%, liver cirrhosis in 14.6% and hepatocellular carcinoma in 34.4% after 30 years when they reached 70 years of age, in contrast to 1.9%, 45.3%, 32.8% and 20.0%, respectively, of female asymptomatic carriers. Likewise, male patients with chronic hepatitis aged 40 years were expected to remain with chronic hepatitis in 43.8%, evolve into liver cirrhosis in 15.0% and hepatocellular carcinoma in 41.1%, contrasting with 38.9%, 32.7% and 22.0%, respectively, of female patients during 30 years. The Markov model could simulate the outcomes of 153 HCV carriers identified among blood donors after 5 years. The Markov simulation would help in assessing the long-term outcome of HCV infection and making decisions in the management of HCV carriers toward prevention of hepatocellular carcinoma. (C) 2003 Wiley-Liss, Inc.