BACKGROUND/AIMS: Serum alpha fetoprotein (alpha-FP) measurement has a limitation to detect hepatocellular carcinoma (HCC) because it is elevated in various liver diseases. Therefore, we studied the sensitivity and specificity of high alpha-FP in the diagnosis of HCC. METHODS: We studied 253 patients with HBsAg positive liver cirrhosis prospectively. We analyzed incidence of HCC related cut-off values of serum alpha-FP levels. During the follow-up period, we analyzed sensitivity and specificity of cut-off values of alpha-FP for the diagnosis of HCC, and alpha-FP elevation rate in relation to mass size. RESULTS: One hundred and twenty-five patients had a transient elevation of alpha-FP levels above 20 ng/mL. The corresponding incidences of HCC were 27.2% (34/125) and 15.6% (20/128 patients without elevation of alpha-FP), respectively with a statistically significant difference (p=0.03). Among 54 patients with HCC, 18 patients (33.0%) had levels of alpha-FP below 20 ng/mL on the time of diagnosis of HCC. When we defined cut-off values of serum alpha-FP as 20, 100 and 500 ng/mL, the corresponding sensitivity and specificity for HCC were 62.9% and 24.0%, 7.4% and 54.2%, 77.3% and 91.9%, respectively. We studied sensitivity according to cut-off values of alpha-FP defined as 20, 100, 200, 500 ng/mL in patients with small HCC below 2 cm. The corresponding sensitivity were 50.0%, 43.7%, 25.0%, 18.7%, respectively. In patients with levels of serum alpha-FP below 20 ng/mL, percentages of mass size less than 2 cm, 2~3 cm, 3~5 cm and more than 5 cm were 50.0%, 25.0%, 28.5% and 25.0%, respectively. CONCLUSIONS: We suggested that in order to detect HCC, careful periodic monitoring with alpha-FP, ultrasonography and abdominal computed tomography is needed in patients with HBsAg positive liver cirrhosis and whose serum level of alpha-FP is above 20 ng/mL.
BACKGROUND/AIMS: Serum alpha fetoprotein (alpha-FP) measurement has a limitation to detect hepatocellular carcinoma (HCC) because it is elevated in various liver diseases. Therefore, we studied the sensitivity and specificity of high alpha-FP in the diagnosis of HCC. METHODS: We studied 253 patients with HBsAg positive liver cirrhosis prospectively. We analyzed incidence of HCC related cut-off values of serum alpha-FP levels. During the follow-up period, we analyzed sensitivity and specificity of cut-off values of alpha-FP for the diagnosis of HCC, and alpha-FP elevation rate in relation to mass size. RESULTS: One hundred and twenty-five patients had a transient elevation of alpha-FP levels above 20 ng/mL. The corresponding incidences of HCC were 27.2% (34/125) and 15.6% (20/128 patients without elevation of alpha-FP), respectively with a statistically significant difference (p=0.03). Among 54 patients with HCC, 18 patients (33.0%) had levels of alpha-FP below 20 ng/mL on the time of diagnosis of HCC. When we defined cut-off values of serum alpha-FP as 20, 100 and 500 ng/mL, the corresponding sensitivity and specificity for HCC were 62.9% and 24.0%, 7.4% and 54.2%, 77.3% and 91.9%, respectively. We studied sensitivity according to cut-off values of alpha-FP defined as 20, 100, 200, 500 ng/mL in patients with small HCC below 2 cm. The corresponding sensitivity were 50.0%, 43.7%, 25.0%, 18.7%, respectively. In patients with levels of serum alpha-FP below 20 ng/mL, percentages of mass size less than 2 cm, 2~3 cm, 3~5 cm and more than 5 cm were 50.0%, 25.0%, 28.5% and 25.0%, respectively. CONCLUSIONS: We suggested that in order to detect HCC, careful periodic monitoring with alpha-FP, ultrasonography and abdominal computed tomography is needed in patients with HBsAg positive liver cirrhosis and whose serum level of alpha-FP is above 20 ng/mL.