Comparative sensitivity of carbohydrate-deficient transferrin and gamma-glutamyltransferase.

Stauber et al. (1995) claim in the Abstract and Discussion of their paper to have demonstrated the superior sensitivity of carbohydrate-deficient transferrin (CDT) to gamma-glutamyltransferase (gamma-GT) for the detection of alcohol abuse in patients with liver disease. This claim is, however, derived from a study of 93 patients, only 13 of whom consumed alcohol in excess. The sensitivity of CDT is reported at 85% (equivalent to the detection of 11 of the 13 alcohol abusers), compared with 83% for gamma-GT (also equivalent to the detection of 11 of the 13), scarcely a significant difference. Stauber and colleagues also report an overall sensitivity for alcohol abuse of 70% for CDT in a total of 199 patients, including the previous 93 subjects, but with the addition of a further 106 patients (17 with excess alcohol consumption) preselected because of a high gamma-GT. On this basis, the CDT sensitivity of 70% (equivalent to 21 out of 30) would require comparison with a sensitivity of 93% (28 out of 30) for gamma-GT. The above authors recognize the quite comparable sensitivity of CDT and gamma-GT in their Results section. To suggest otherwise in the Abstract and in the opening of the Discussion is, therefore, quite misleading. In fact, the authors' data support the view that, because of the cost and time-consuming nature of the CDT assay, the more cost-effective approach at this time to the recognition of alcohol abuse is the initial determination of gamma-GT followed by CDT measurement only if the former is elevated. The concomitant finding by Stauber et al. that CDT has greater specificity than gamma-GT for alcohol abuse in patients with liver disease is already well-established.