REPLY: SUPERIOR DIAGNOSTIC ACCURACY OF CARBOHYDRATE-DEFICIENT TRANSFERRIN OVER GAMMA-GLUTAMYLTRANSFERASE

In his letter, Dr Rosalki (1996) criticizes our claim (Stauber et al, 1995) of superior sensitivity of carbohydrate-deficient transferrin (CDT) over that of gamma-glutamyltransferase (gamma-GT). It was not our intention to claim that sensitivity of CDT alone is superior to that of gamma-GT, but that sensitivity and specificity of CDT together, and thus its overall diagnostic accuracy, are superior. For comparison of CDT with gamma-GT, we analysed only data obtained from outpatients who had various kinds of chronic liver disease, but were not preselected because of an elevated gamma-GT. As can be seen in Table 2 of our article (Stauber et al., 1995), the sensitivities of CDT (85%) and gamma-GT (83%) were comparable, while the specificity of CDT (83%) was superior to that of gamma-GT (16%). These findings are consistent with a previous study by Bell et al. (1993), who reported sensitivities of 61% for CDT and 85% for gamma-GT and specificities of 92% for CDT and 18% for gamma-GT. Based on these data, it may be clearly concluded that overall diagnostic accuracy of CDT is superior to that of gamma-GT in patients with liver disease.