Efficacy of Individual Prediction Model for the Early Diagnosis of Hepatocellular Carcinoma / 대한내과학회지

ABSTRACT

BACKGROUND: Screening tests for hepatocellular carcinoma (HCC) in the high risk population can detect tumors at an earlier stage and thus confer a higher chance of receiving treatment. However, the usefulness, frequency and cost-effectiveness of screening for HCC may differ in different areas, possibly reflecting differences in risk factors. Last decade, we have identified risk factors for HCC in 4339 Korean patients. The aim of this study was to investigate the efficacy and usefulness of individual prediction model for the early diagnosis of HCC. METHODS: We studied a total of 833 patients who visited Yonsei University Medical Center for regular check-up including ultrasonography and alpha-fetoprotein from January 1999 to December 2000. The patients were classified into a low risk group ( 15%) by the probability of HCC development according to individual prediction model (IPM). The patients who developed HCC during the follow-up periods were analyzed using IPM. All the detailed data of clinical parameters were obtained by our self-exploited data base system prospectively and analyzed by SAS program. RESULTS: 44 (5.3%) out of 833 patients developed HCC during mean follow-up periods of 36 months. According to IPM, 2 (0.62%) of 324 patients in the low risk group, 20 (4.84%) of 413 patients in the intermediate risk group, and 22 (22.9%) of 96 patients in the high risk group were diagnosed as HCC. In 29 of 44 HCC patients (65.9%), initial presentation of tumor size was less than 3 cm in diameter. CONCLUSION: We confirmed the reliability of established IPM for screening of HCC and this model may help screening program to be done effectively by focusing high risk groups for HCC.