Abstract
OBJECTIVES: Several sets of criteria have been suggested for clinical diagnosis of hepatocellular carcinoma (HCC) without biopsy but there are no comprehensive data to support the usefulness of these criteria. Here, we sought to validate the accuracy of our clinical criteria for HCC diagnosis in a cohort of patients, and further tested the effect of HBV and clinical cirrhosis status on diagnostic accuracy. METHODS: A total of 232 patients with liver nodules >1 cm in diameter who underwent surgical resection or liver biopsy, and had fulfilled all required examinations for clinical non-invasive diagnosis of HCC were reviewed retrospectively. RESULTS: Hepatitis B virus (HBV) was positive in 170 patients (73.3%). One hundred and eighty-nine cases were diagnosed as HCC using the clinical criteria and 186 cases of HCC were confirmed by pathologic examination. The overall sensitivity, specificity and positive predictive value of the clinical criteria were 95.1, 73.9 and 93.7%, respectively. The accuracy was not significantly affected by lesion size (1-2 cm vs. >2 cm) or the presence of clinical cirrhosis. The sensitivities were 97.3 and 86.8% in the HBsAg positive group and non-HBV group, respectively (P<0.001), and the specificities were 56.5 and 91.3%, respectively (P<0.001). CONCLUSIONS: The clinical criteria for the diagnosis of HCC showed an acceptable accuracy irrespective of lesion size or the presence of clinical cirrhosis in an HBV-endemic population. However, the presence of HBV affected the sensitivity and specificity of the clinical criteria for HCC diagnosis in an HBV endemic area.