HPLC evaluation of clinical and pharmacological factors reported to cause false-positive carbohydrate-deficient transferrin (CDT) levels

Background: Carbohydrate-deficient transferrin (CDT) is an alcohol biomarker used for detection and follow-up of excessive alcohol consumption. This HPLC study evaluated some clinical conditions and medications previously suggested to interfere with and cause non-alcohol related elevations of CDT. Methods: Serum samples were collected from patients with end-stage liver disease (n=50), type 2 diabetes mellitus (n=46), cystic fibrosis (n=24), an elevated C-reactive protein level (CRP > 100 mg/L; n=15), and from patients taking enzyme inducing or non-enzyme inducing antiepileptic drugs (n=43). Subjects with known or suspected alcohol-related problems were excluded. A sensitive and specific HPLC candidate CDT reference method was used to determine the relative amount of disialotransferrin to total transferrin. Results: Of the 178 samples, 9 (5%) had a %disialotransferrin level >= 1.8% (> 97.5th percentile) and were considered CDT positive. The highest frequency of elevated results was found in patients with end-stage liver disease (12%, n=6), including 3 with hemochromatosis, 1 with hepatitis C, 1 with autoimmune hepatitis and 1 with unspecified liver disease and cirrhosis. The other elevated %disialotransferrin results were from 2 patients taking enzyme-inducing antiepileptic drugs and 1 with type 2 diabetes. Five of 8 examined %disialotransferrin positive samples were also positive for ethyl glucuronide (EtG). Conclusion: This HPLC study found an overall low frequency of elevated %disialotransferrin levels in the clinical conditions and medications examined. Previous reports of frequent false-positive CDT results thus seem to be connected with the analytical methodology used rather than representing true clinical or pharmacological interferences. (c) 2007 Elsevier B.V. All rights reserved.