The carbohydrate-deficient transferrin test in hospital practice

We report an experience in two hospital populations of the use of a commercially available kit for the detection of carbohydrate-deficient transferrin (CDT). Patients from a drug and alcohol unit and a gastroenterology clinic at two hospitals were selected for the study. Sera were used from blood samples collected for routine biochemical assays. All patients had a specific alcohol history taken by one clinician and CDT results were correlated with reported alcohol intake by the patient and where relevant by their relatives. Sensitivity and specificity of the CDT assay were calculated using an alcohol intake of 60 g/day as the cut-off point for detection of heavy drinking. The CDT assay had a specificity of 95%; a sensitivity of 80% and a 90% positive and 89% negative predictive value. The severity and type of liver disease had little influence on the CDT result and a high alcohol intake was the only predictor of a raised CDT concentration. The assay provided information not available from routine investigations in some patients and also proved useful in monitoring patients over periods of up to 4 years. The test has a role in the evaluation of patients in a hospital practice where routine histories of alcohol intake may lack sensitivity and where other diseases may cause routine liver tests to be unreliable.