Establishment of Individual Prediction Model According to Risk Factors for Development of Hepatocellular Carcinoma in Korea: Establishment of Individual Prediction Model for Hepatocellular Carcinoma

Cheong, Jae Youn; Han, Kwang Hyub; Kim, Dong Kee; Ahn, Sang Hoon; Song, Ki Jun; Paik, Yong Han; Choi, Chang Hwan; Lee, Hyun Woong; Park, Young Soo; Chon, Chae Yoon; Moon, Young Myoung

Abstract

BACKGROUND/AIM: We identified risk factors for hepatocellular carcinoma (HCC) through a nine-year follow-up study, ending last year, of 4,339 patients with chronic liver diseases. The aim of this study was to establish an individual prediction model according to risk factors for the development of HCC.
METHODS
We studied a total of 994 patients who had regular check-ups from January 1990 to December 1998. We analyzed the risk factors and established the individual prediction model to predict the risk rate for HCC using logistic regression analysis. We applied the model to patients who were enrolled over the next two years.
RESULTS
90 (9.05%) out of 994 patients developed HCC during a mean of 33 months of follow-up. The risk index for individual patients was made by considering the relative risk level of statistically significant risk factors. From 1999 to 2000, 480 patients were newly enrolled and divided into three groups by their risk index and probability of HCC development. These patients were classified into a low risk group (less than 5% probability), an intermediate risk group (5% to 10% probability), and a high risk group (more than 10% probability). According to this classification, 1 of 191 patients in the low risk group (0.523%), 5 of 176 patients in the intermediate risk group (2.84%), and 21 of 113 patients in the high risk group (18.6%) were diagnosed with HCC.
CONCLUSION
We confirmed the reliability of the newly established individual prediction model for the screening of HCC. This model may help screening programs to be done effectively by focusing on high risk groups for HCC.