Carbohydrate deficient transferrin in alcoholic and non-alcoholic liver disease: a comparison of two assay methods.

Carbohydrate-deficient transferrin (CDT) was assayed in 105 patients with non-alcohol-related liver diseases, 50 patients with alcohol-induced liver disease and 40 alcohol misusers with minimal hepatic dysfunction. The patients with liver disease were hospitalized for assessment of suitability for orthotopic liver transplantation. CDT was measured by two commercially available micro anion exchange methods; CDTect (Pharmacia) and %CDT (AXIS). Additionally, total transferrin was measured to allow expression of the CDTect results as 'relative' measurements. Overall, 41/105 (39%) of patients with non-alcohol-related liver diseases had a raised CDTect result, 13/105 (12%) had a raised %CDT result and 26/105 (25%) an abnormal CDTect/total transferrin ratio. In general, patients with cholestatic liver disease, primary biliary cirrhosis or primary sclerosing cholangitis had more 'false positive' results by CDTect than did those with viral or autoimmune hepatitis. The sensitivity of the two methods for the detection of alcohol misuse was similar; CDTect 77%, %CDT 65% and CDTect/total transferrin ratio 70%. This was, however, lower than the sensitivity of the conventional marker GGT (82%). In patients with severe liver disease, where up to 30% of the serum total transferrin concentrations fall outside the reference range, 'relative' CDT methods have greater specificity than 'absolute' measurements.