Validity of carbohydrate-deficient transferrin (%CDT), γ-glutamyltransferase (γ-GT) and mean corpuscular erythrocyte volume (MCV) as biomarkers for chronic alcohol abuse:: a study in patients with alcohol dependence and liver disorders of non-alcoholic and alcoholic origin

Aim To test the clinical performance of carbohydrate-deficient transferrin (%CDT), gamma-glutamyltransferase (gamma-GT) and mean corpuscular erythrocyte volume (MCV) as biomarkers for alcoholism with a special focus on patients suffering from liver diseases. Design Well-characterized collectives of alcohol-dependent patients with current consumption (ALC patients, n = 101), and relevant control groups (115 social drinkers, 46 patients with unspecifically increased gamma-GT, 51 hepatitis patients and 20/31 patients with non-alcohol/alcohol-dependent liver cirrhosis) were included into the study. The Positive Alcohol Use Disorders Test (AUDIT) score, International Classification of Diseases version 10 (ICD-10)/Diagnostic and Statistical Manual version IV (DSM-IV) criteria and blood drawn within 4 days of last drinking were inclusion criteria for subjects with regular heavy drinking. %CDT was determined using an automated assay which recently had been completely modified. Findings Median AUDIT scores of patients without/with regular heavy drinking were 1-3/27. The following medians/95th percentiles were obtained for %CDT: social drinkers 2.2/3.0, patients with unspecifically increased gamma-GT 2.1/3.0, hepatitis 2.0/4.4, non-alcohol-dependent liver cirrhosis 2.4/4.8, alcohol-dependent liver cirrhosis 3.0/5.9, ALC patients 3.9/14.9. Differences between patients without and with alcohol abuse were highly significant (P < 0.001). No differences in CDT values were found between males and females. There was no correlation between %CDT values, gamma-GT, MCV and the amount of alcohol consumed in ALC patients; 3.0%CDT (95th percentile social drinkers) is proposed as cut-off for the test used (Tina-quant((R))%CDT 2nd-generation). At this cut-off, the sensitivity for ALC patients was 73.3%, whereas gamma-GT/MCV had a sensitivity of 71.3%/64.4%. Multivariate analysis performed at 95% specificity resulted in an improvement of the sensitivity by combining %CDT with gamma-GT (83.2%). A further enhancement of the sensitivity to 88.1% was obtained by combination of %CDT, gamma-GT and MCV. The diagnostic specificity of %CDT calculated at the cut-off of 3% was 93.5% in patients with unspecifically increased gamma-GT, 88.2% in hepatitis patients and 70.0% in patients with non-alcohol-dependent liver cirrhosis. %CDT was more specific in these patient collectives than MCV, and especially more than gamma-GT (specificity in hepatitis 52.9%, and 35.0% in non-alcohol-dependent liver cirrhosis). Conclusion %CDT is of high diagnostic value to support diagnosis of alcohol-use disorders. The specificity of this marker in patient groups with liver disorders is superior to the biomarkers gamma-GT and MCV.