Background
: The optimal screening tools for hepatocellular carcinoma are regular tests of alpha fetoprotein (aFP) and liver ultrasonography every 6 months in high risk group. To implement successful long-term project for reducing cancer-mortality, it is essential to know the reasons for non-compliance among the high risk group.

Methods : A telephone survey was done among the hepatitis B surface antigen positive patient group who had not received a follow-up test of aFP and liver ultrasonography. This group was selected among the patients who had received health examinations from January, 2002 to December, 2002 at a university hospital center for health promotion, and the reasons for non-compliance were inquired.

Results : There were no statistical differences in educational level, economic status, age, and sex between the patient group who had taken appropriate follow-up tests and the group who had not. But there were statistically significant differences in the admission history and transfusion history between the two groups. The reasons for non-compliance were as follows: 1. ignorance to the significance of follow-up examination (41.7%), 2. lack of time (27.8%), 3. no specific signs of cancer (22.2%), 4. high costs (5.6%), and 5. fear of discovering serious disease.

Conclusion : It is important to have patients educated on the significance of screening and early diagnosis of hepatocellular carcinoma.

Background
: Serum alpha-fetoprotein(aFP) is a useful diagnostic test on hepatocellular carcinoma(HCC) However, it is still questionable appropriate screening test or not

Methods : A total of 24,050 patients who visited the General Health Screening Center of Asan Medical Center from June, 1993 to June, 1994, were reviewed by chart and telephone survey. Among them 70 cases where serum aFP were elevated > 20ng/ml and 16 cases where HCC was suspected by abdominal ultrasonogram were selected and investigated for the existence of HCC. Four groups were divided into HBsAg(+), Anti-HCV(+), increased AST/ALT(abnormal LFT) and normal group. Each group was evaluated for sensitivity, specificity, positive, predictive value and negative predictive value of serum aFP for HCC.

Results : The prevalence of elevated aFP was 0.29%(70/24,050). There was a statistically significant difference between sexes. And the total number of HCC as 10 cases (42/100,000). With aging, the prevalence of HCC was increased. The prevalence rate pf elevated serum aFP and HCC had significant difference in anti-HCV(+), HBsAg(+) and abnormal LFT group compared to normal group(sGOT < 40U/1, HBsAg(-) and Anti-HCV(-)) (P<0.01).
In our study, sensitivity and positive predictive value of serum aFP for HCC were 100% and 14% in anti-HCV(+) group, 83.3% and 8.6% in abnormal LFT group, 80% in HBsAg(+) group, and 0% in normal group, respectively.

Conclusion : Serum aFP as a screening test for HCC in asymptomatic healthy people does not seem to be not valid. But it seems to be appropriate when it is applied for cases of abnormal LFT(sGOT >40U/1 or sGPT > 40U/1), HBsAg(+) and Anti-HCV(+) people.