BACKGROUND: Carbohydrate-deficient transferrin (CDT) levels have rarely been determined in an Asian population. We evaluated the analytical performance of a test for measuring CDT levels by using capillary electrophoresis (EP). METHODS: We determined the precision of CDT measurement by using capillary EP and nephelometry and compared the CDT values obtained using both the methods. We included healthy control subjects, abstinent patients with liver disease, and individuals consuming varying amounts of alcohol. RESULTS: The CDT measurement by using capillary EP were correlated well with those CDT measurement by using nephelometry, N Latex CDT assay, Y=0.5706X+1.581, R=0.930. The results obtained from both methods showed good qualitative agreement with each other (kappa coefficient=0.61). Genetic variants of transferrin isoforms were detected in 4.1% of the tested population. Both the CDT and gamma-glutamyl transpeptidase (GGT) levels in the abstinent patients with liver disease were significantly higher than those in healthy abstinent individuals (0.9% vs. 0.5%, 109.5 mg/dL vs. 28.5 mg/dL, respectively), but the difference in CDT values in the 2 groups was less pronounced for the CDT values. Individuals who had a mean daily alcohol intake of more than 60 g/day showed significantly higher CDT levels than those who had a mean daily alcohol intake of less than 60 g/day (1.9% vs. 0.7%, P=0.03). CONCLUSIONS: The CDT test using capillary EP showed good performance, and this method has several advantages such as automation and detection of variant forms. Thus, CDT can be a more useful marker than GGT for monitoring alcohol abstinence, especially in patients with liver disease.