Literature screening was performed at four levels. Level 1, reviews, letters, case reports, editorials, and comments were excluded from the papers identified using the above search strategy. Level 2, articles in which biomarkers were not evaluated for their utility in detecting hepatocellular carcinoma (HCC) were excluded. The full texts of reports that met the above criteria were obtained with duplicate articles excluded. Level 3, the content of the articles was analyzed to ensure that the serum biomarkers in the study included Alpha-fetoprotein (AFP), des-gamma-carboxy prothrombin (DCP), Golgi protein-73 (GP73), or circulating AFP isoform AFP-L3, and that these biomarkers were used just for diagnosing HCC. The data pertaining to other biomarkers were excluded from further analysis. Articles were further screened to ensure that the studies included data pertaining to patients with HCC and appropriate control populations. At Level 4, only reports with sensitivity and specificity data for the biomarkers were selected. A total of 40 reports met the inclusion criteria and were selected for meta-analysis.