Home Translating Report News Physicians Diseases Body Sites Lab tests Search
Home Diseases and Health Information


CDT stands for carbohydrate-depleted transferrin. This is a serum protein which has proven to be a sensitive marker of liver disease associated with alcohol abuse.


Reference Methods  
Clinical Utility  
Interfering Diseases or Substances that Alter Levels  
Commonly Used Terms  
Internet Links  


Should we use carbohydrate-deficient transferrin instead of gamma-glutamyltransferase for detecting problem drinkers? A systematic review and metaanalysis.

Scouller K, Conigrave KM, Macaskill P, Irwig L, Whitfield JB.

Drug and Alcohol Departments, Royal Prince Alfred Hospital, Missenden Rd., Camperdown, NSW 2050, Australia.


Clin Chem 2000 Dec;46(12):1894-902 Abstract quote

BACKGROUND: Carbohydrate-deficient transferrin (CDT) has been used as a test for excessive alcohol consumption in research, clinical, and medico-legal settings, but there remain conflicting data on its accuracy, with sensitivities ranging from <20% to 100%. We examined evidence of its benefit over a conventional and less expensive test, gamma-glutamyltransferase (GGT), and compared the accuracy of different CDT assay methods.

METHODS: We performed a systematic review using summary ROC analysis of 110 studies prior to June 1998 on the use of CDT in the detection of alcohol dependence or hazardous/harmful alcohol use.

RESULTS: We identified several potential sources of bias in studies. In studies examining CDT and GGT in the same subjects, subject characteristics were less likely to influence the comparison. In such paired studies, the original Pharmacia CDT assay was significantly more accurate than GGT, but the modified CDTect assay did not perform as well as the original and was not significantly better than GGT. The accuracy of the AXIS %CDT assay was statistically indistinguishable from modified CDTect. Several CDT assay methods appeared promising, in particular, liquid chromatography (chromatofocusing, HPLC, fast protein liquid chromatography) and isoelectric focusing, but there were insufficient paired studies from which to draw firm conclusions.

CONCLUSIONS: In studies published before June 1998, the results obtained with commercially available CDT assays were not significantly better than GGT as markers of excessive alcohol use in paired studies. Further high-quality studies comparing CDTect (modified) and other CDT assays with GGT in the same subjects are needed.

Comparison of different methods for detecting carbohydrate-deficient transferrin.

Sillanaukee P, Lof K, Harlin A, Martensson O, Brandt R, Seppa K.

Biomedical Research Center, Alko Ltd., Helsinki, Finland.


Alcohol Clin Exp Res 1994 Oct;18(5):1150-5 Abstract quote

Different methods for detecting carbohydrate-deficient transferrin (CDT) were compared. In addition, their efficiency for detecting alcohol abuse among men not having clinical evidence of liver disease was studied in controls (n = 26), weekend (n = 16) and daily (n = 12) heavy drinkers, and alcoholics (n = 28).

Comparisons were made between anion-exchange separation of iron-saturated transferrin (Tf) by microcolumns (CDTect) and by the Fast Protein Liquid Chromatography (FPLC% and FPLC-MG), followed by double-antibody radioimmunoassay of collected fractions. Tf fractions with pl > or = 5.7 were also measured by two different isoelectric focusing (IEF) methods, followed by immunofixation (SA-IEF-CDT and IEF-CDT-TOT), the latter method being used also for detection of asialotransferrin (IEF-CDT-AS). The cut-off was 20 units/liter for CDTect, 4.4% of total Tf for SA-IEF-CDT, and the mean +2 sd of the control group for FPLC-MG (as mg/liter of Tf), FPLC-%, IEF-CDT-TOT, and IEF-CDT-AS (all as percentage of Tf).

The overall accuracies (combining sensitivity and specificity) for detecting heavy drinkers of CDTect, FPLO (mg/liter), FPLC (%), SA-IEF-CDT, IEF-CDT-TOT, and IEF-CDT-AS were 63%, 59%, 61%, 74%, 57%, and 63%, respectively; for detecting alcoholics, 87%, 83%, 81%, 89%, 37%, and 76%, respectively.

In conclusion, the methods were in rather good agreement with each other. Diagnostic characteristics among heavy drinkers and correlations between methods differed slightly, probably depending on the ability of different methods to separate and detect asialo-, monosialo-, and disialotransferrin.



Laboratory tests for acute alcohol consumption: results of the WHO/ISBRA Study on State and Trait Markers of Alcohol Use and Dependence.

Helander A, Eriksson CJ; WHO/ISBRA Study on State and Trait Markers ofAlcohol Use and Dependence Investigators.

Department of Clinical Neuroscience, Karolinska Institute, Stockholm, Sweden.

Alcohol Clin Exp Res 2002 Jul;26(7):1070-7 Abstract quote

BACKGROUND: The WHO/ISBRA Study on State and Trait Markers of Alcohol Use and Dependence aimed partly to evaluate the overall performance and cross-national validity of traditional and new biological markers of alcohol use and abuse. This article focused on the sensitivity and specificity of ethanol and methanol concentrations in plasma, and the 5-hydroxytryptophol (5HTOL) to 5-hydroxyindole-3-acetic acid (5HIAA) ratio in urine, as laboratory tests to identify acute alcohol consumption. Comparison was made with self-reported drinking levels.

METHODS: Subjects were recruited in Australia, Brazil, Canada, Finland, and Japan. They were interviewed thoroughly about their alcohol consumption habits, by using the standardized WHO/ISBRA Interview Schedule, and were classified into four categories: nondrinkers, light/moderate drinkers, heavy drinkers (> or =210 g ethanol/week for men, and > or =140 g/week for women), or patients who were receiving treatment for alcohol dependence. Ethanol and methanol determinations in plasma were carried out by headspace gas chromatography. Urinary concentrations of 5HTOL and 5HIAA were determined by using gas chromatography-mass spectrometry and high-performance liquid chromatography, respectively.

RESULTS: The baseline levels (in nondrinkers) for methanol and the 5HTOL/5HIAA ratio did not differ markedly between the five populations, except for a considerably higher, but probably artifactual, methanol level in the Finnish plasma samples. Moreover, there were no apparent age or sex differences. The urinary 5HTOL/5HIAA ratio was the most, and ethanol the least, sensitive indicator of recent alcohol consumption, and this was true for the different drinking categories as well as for the five study populations. The highest frequency of elevated test results was observed among those classified as heavy drinkers (e.g., 38% were positive for 5HTOL/5HIAA). However, elevated values also were obtained in nondrinkers and in drinking subjects who denied any intake of alcohol within 2 days before the interview and blood/urine sampling, which suggested a low accuracy of self-reports of alcohol consumption in certain individuals.

CONCLUSIONS: The present investigation demonstrated that plasma ethanol and methanol and urinary 5HTOL/5HIAA provide useful exclusion markers for any study of biological parameters that are affected by previous acute ethanol intoxication. The major advantage of methanol and 5HTOL/5HIAA over ethanol is that they can detect recent alcohol consumption even several hours after the ethanol is no longer measurable. The results suggest that the cutoff limits to be used for these markers are not dependent on the country or population to be studied.

The combined use of the early detection of alcohol consumption (EDAC) test and carbohydrate-deficient transferrin to identify heavy drinking behaviour in males.

Harasymiw J, Bean P.

Alcohol Detection Services, LLC, 4670 Somerset Court, Brookfield, WI 53045 and. Millennium Strategies, 418 N. Westfield Road, Madison, WI 53717, USA.

Alcohol Alcohol 2001 Jul-Aug;36(4):349-53 Abstract quote

The aim of this study was to determine the efficacy of the combined use of carbohydrate-deficient transferrin (CDT) and the Early Detection of Alcohol Consumption (EDAC) test to assess heavy drinking in a population of males (n = 187) drinking an average of 20 drinks per day. Heavy drinkers (n = 138) and light drinkers (n = 49) were analysed in three ways: using the EDAC test alone, using the CDT test alone and using the EDAC and CDT tests combined. The EDAC method uses linear discriminant function to analyse a battery of routine laboratory tests that generate a score for each subject and its associated probability value. This translates into the likelihood of each individual being a heavy or a light drinker.

CDT uses ion-exchange chromatography to extract CDT in the serum and quantifies it by radioimmunoassay. The EDAC alone showed 88% (122/138) sensitivity rate when identifying heavy drinking males and 98% (48/49) specificity rate when assessing light drinkers. The CDT test alone showed a sensitivity rate of 58% (80/138) and a corresponding specificity rate of 96% (47/49). When analysed in parallel, 92% (127/138) of heavy drinkers showed abnormal EDAC and/or CDT tests and 94% (46/49) of light drinkers were negative for both tests. When analysed sequentially, the CDT test confirmed 61% (75/122) of the heavy drinkers identified by the EDAC test. Specificity rate for this testing strategy was 100%, because the only false positives for EDAC tested negative for CDT.

This preliminary study shows that EDAC and CDT may react independently to alcohol intake and can be combined for maximum diagnostic accuracy.

Carbohydrate-deficient transferrin for detection and monitoring of sustained heavy drinking. What have we learned? Where do we go from here?

Anton RF.

Medical University of South Carolina, 67 President Street, P.O. Box 250861, Charleston, SC 29425, USA.

Alcohol 2001 Nov;25(3):185-8 Abstract quote

Carbohydrate-deficient transferrin (CDT) has been widely investigated as a biological marker of heavy alcohol consumption. In general, it has been found to be at least as sensitive and probably more specific than gamma-glutamyltransferase (GGT). Because the two analytes are not highly correlated, their use in parallel enhances the sensitivity of detection of heavy alcohol consumption, especially in clinical populations.

Women as a group produce more CDT under natural conditions and may produce less CDT in response to heavy drinking. Carbohydrate-deficient transferrin has found a place in the monitoring of alcoholics during treatment. Changes in CDT levels from individualized baseline values seem to be more sensitive to lower level relapse drinking than is the use of raw cut-off values. There are some conditions such as severe liver disease in which higher than normal levels of CDT are produced, thereby reducing the specificity of this marker for detecting heavy drinking under certain conditions.

uture directions for the use of CDT include standardization and automation of measurement technology, evaluation of how to use it wisely in myriad medical and institutional settings, understanding more thoroughly the gender issues in its production, and greater evaluation of its performance as a monitoring tool during treatment and follow-up situations. How to combine CDT with both verbal tools of alcohol assessment and newer biological markers will also need more extensive evaluation.

Carbohydrate-deficient transferrin as compared to other markers of alcoholism: a systematic review.

Salaspuro M.

Research Unit of Alcohol Diseases, University of Helsinki, Finland.


Alcohol 1999 Nov;19(3):261-71 Abstract quote

This is a systematic review of the studies in which carbohydrate-deficient transferrin (CDT) has been compared to other laboratory markers in different experimental conditions, clinical settings, and populations.

Only the studies (n = 54) in which CDT was compared either to the conventional or new biological markers of alcoholism, heavy drinking, or alcohol use were selected for further evaluation. Two prospective studies indicate that in men CDT is slightly more sensitive than gamma-GT in reflecting changes in these markers caused by drinking of a moderate and fixed amount of alcohol during three to four weeks.

In one prospective study, in which the drinking history of male heavy drinking volunteers was as close the golden standard as possible; that is, obtained by a prospective anonymous drinking diary, CDT was slightly but not significantly better marker than conventional laboratory markers (ASAT, ALAT, gamma-GT and beta-Hex) in the identification of men drinking more than 400 g of alcohol daily.

Similar prospective studies concerning women have not been done. Six prospective treatment outcome studies indicate that CDT may be a significantly more sensitive marker than gamma-glutamyltransferase (gamma-GT) in the detection of relapses in male alcoholics. However, these two tests can also be considered to be complementary markers. Furthermore, in the detection of relapses the baseline values of CDT and gamma-GT should be measured and compared on individual basis to the pretreatment values. Comparable data are not available from female alcoholics. In selective materials comprising male alcoholics and heavy drinkers, CDT was found to be a slightly more sensitive marker than gamma-GT in seven retrospective studies. In five studies, gamma-GT was slightly better. However, the differences between CDT and gamma-GT in general were not statistically significant. In three studies, the combined use of CDT and gamma-GT improved the sensitivity but with the expense of specificity. Only four studies included women and in three of these the sensitivity of gamma-GT was better than that of CDT, whereas in one study CDT was better than gamma-GT in the detection of female heavy drinkers. Seven studies performed in primary health care settings and among young populations demonstrate that the performance of CDT in the identification of heavy and problem drinkers in this type of populations is very low, although comparable to the poor performance of the conventional laboratory markers, too. According to seven studies, the sensitivity of gamma-GT is slightly better than that of CDT in the identification of excessive alcohol consumption among hospitalized male and female patients. However, in this type of hospital setting, the specificity of CDT is markedly higher than that of gamma-GT. There is some evidence indicating that the performance of the tests can be improved with the combined use of both tests.

Eight studies indicate that both in men and women CDT is a better marker than gamma-GT in the identification of alcohol abuse among patients with alcoholic and nonalcoholic liver diseases. This is mostly due to the higher specificity of CDT as compared to that of gamma-GT.

Superiority of carbohydrate-deficient transferrin to gamma-glutamyltransferase in detecting relapse in alcoholism.

Schmidt LG, Schmidt K, Dufeu P, Ohse A, Rommelspacher H, Muller C.

Department of Psychiatry, Benjamin Franklin University Hospital, Free University of Berlin, Germany.

Am J Psychiatry 1997 Jan;154(1):75-80 Abstract quote

OBJECTIVE: The usefulness of carbohydrate-deficient transferrin is widely accepted in screening (male) population samples for heavy alcohol consumption, but its role in relapse detection is not convincingly established. The authors therefore compared the diagnostic value of carbohydrate-deficient transferrin with the commonly used gamma-glutamyltransferase in identifying relapsed alcoholics during outpatient aftercare.

METHOD: The patients were 101 male alcoholics who entered a 6-month rehabilitation program after hospital detoxification. Drinking status was assessed by means of self- and collateral reports obtained during regular contacts with the rehabilitation team; relapse was defined as consumption of any alcohol. Visits occurred weekly during month 1, biweekly during month 2, and every 4 weeks during months 3-6. At every visit a blood sample was taken for measurement of carbohydrate-deficient transferrin and gamma-glutamyltransferase.

RESULTS: The proportion of men who reported relapse was 25.6% per scheduled contact on average. Positive predictive values indicated that relapse was identified with a 76.2% probability by carbohydrate-deficient transferrin values above the upper normal limit, in contrast to a 32.9% chance with gamma-glutamyltransferase. Carbohydrate-deficient transferrin was especially useful in detecting early relapses during the initial rehabilitation phase, when gamma-glutamyltransferase values had not normalized. Because of the longer half-life of gamma-glutamyltransferase, it had some value with a 4-week monitoring schedule in detecting new drinking episodes in alcoholics whose previous results had been normal.

CONCLUSIONS: Carbohydrate-deficient transferrin proved to be superior to gamma-glutamyltransferase in relapse detection in an outpatient care setting for alcoholics.

Longitudinal comparison of carbohydrate-deficient transferrin and gamma-glutamyl transferase: complementary markers of excessive alcohol consumption.

Helander A, Carlsson AV, Borg S.

Karolinska Institute, Department of Clinical Neuroscience, Psychiatry Section at St Gorans Hospital, Stockholm, Sweden.


Alcohol Alcohol 1996 Jan;31(1):101-7 Abstract quote

The utility of carbohydrate-deficient transferrin (CDT) and gamma-glutamyl transferase (GGT) as biochemical markers of excessive alcohol consumption was studied in alcohol-dependent subjects.

Serum samples were collected once weekly from 10 male out-patients undergoing a 6-month alcohol treatment programme. Frequency of relapse into drinking (defined as any intake of alcoholic beverage) was assessed by self-reports during patient interviews three times per week and by daily determination of the 5-hydroxytryptophol level in urine. A marked decrease in mean CDT and GGT values was observed during the initial month. Only one patient remained totally abstinent throughout the observation period, while four had sporadic relapses (2-5 days with alcohol consumption). Both CDT and GGT remained below the respective reference limits in those patients. The other five patients drank more frequently (range 22-57 days) and increased their mean levels of CDT and GGT after the initial decrease.

As determined from the values at admission and during the course of the study, CDT appeared to be the most sensitive marker in six out of the 10 patients. In one patient, both markers were affected in a parallel way, whereas two of those with frequent relapses responded to alcohol consumption with a marked increase in GGT, but with no or only a slight increase in CDT. One patient did not show any abnormal CDT or GGT values. In 54 female and 60 male serum samples collected at random from patients during admission at an alcohol detoxification unit, 35% and 58% of the CDT values exceeded the reference limits for females and males, respectively. For GGT, 59% of the female and 67% of the male values were above cut-off. Carbohydrate-deficient transferrin and GGT were not significantly correlated.

Taken together, the present results indicate that measurement of both CDT and GGT will increase the possibility of identifying excessive alcohol consumption. By following changes in CDT and GGT values during a period of alcohol withdrawal, the most sensitive individual marker can be determined. This in turn allows for improved detection of relapse into heavy drinking during long-term monitoring of out-patients.



Cord blood carbohydrate-deficient transferrin levels are markedly higher than maternal.

Whitty JE, Dombrowski MP, Martier SS, Subramanian MG, Sokol RJ.

Department of Obstetrics and Gynecology, Hutzel Hospital/Wayne State University, Detroit, MI 48201, USA.

J Matern Fetal Med 1997 Jan-Feb;6(1):45-8 Abstract quote

Regular, heavy alcohol intake results in transferrin that is deficient in carbohydrate moieties. Carbohydrate-deficient transferrin (CDT) has been used as a biologic marker of heavy alcohol exposure in nonpregnant humans. There have been no reports of CDT levels in pregnancy.

Our objective was to determine maternal and cord blood levels of CDT. Parturients were recruited at delivery based on graded representative alcohol consumption, from abstainers to heavy drinkers, as determined by screeners skilled at eliciting drug and alcohol histories. Maternal and cord blood serum samples were obtained at delivery. A double antibody radioimmunoassay was used to determine CDT in each sample. There were 83 paired specimens analyzed by paired t tests and stepwise regression analysis. Cord blood CDT units/liter (44.0 +/- 29.5) were significantly (P < 0.0001) higher than maternal (18.4 +/- 7.0). Maternal and cord CDT did not correlate with race, perinatal risk score, gestational age at delivery, birth weight, Apgar scores, or reported alcohol intake. Maternal CDT levels had a significant negative correlation with cigarette smoking.

Cord blood CDT levels are significantly higher than maternal. While regular, heavy alcohol consumption by adults results in serum transferrin deficient in carbohydrate moieties, the reason for elevated fetal CDT is unknown.


Carbohydrate-deficient transferrin levels in a female population.

La Grange L, Anton RF, Garcia S, Herrbold C.

Department of Psychology, New Mexico Highlands University, Las Vegas 87701, USA.

Alcohol Clin Exp Res 1995 Feb;19(1):100-3 Abstract quote

Female college students (n = 49) from a small southwestern United States university participated in the 9-month study. Data collected included the assessment of drinking habits and other related substance use habits and serum levels of carbohydrate-deficient transferrin (CDT). Each subject provided an interview and blood sample on three occasions at 90-day intervals.

Appended to the interview were a series of questions regarding stage of menstrual cycle and the diagnoses of certain diseases. The CDT values obtained were consistent with those obtained in other studies. Moderate-drinking subjects had significantly higher CDT values than did the abstainers and light drinkers. Females using oral contraceptives had significantly higher CDT values than those who were not taking oral contraceptives.

Finally, although CDT values varied over time, they did not appear to vary as a function of menstrual cycle stage.


Carbohydrate-deficient transferrin and other markers of high alcohol consumption: a study of 502 patients admitted consecutively to a medical department.

Bell H, Tallaksen CM, Try K, Haug E.

Medical Department, Aker University Hospital, Oslo, Norway.


Alcohol Clin Exp Res 1994 Oct;18(5):1103-8 Abstract quote

An isoform of transferrin, carbohydrate-deficient transferrin (CDT) is increased in a high percentage of abusing alcoholics and has been found superior in its specificity compared with other biological markers.

We used serum CDT as a screening parameter in 502 patients consecutively admitted to our medical department during a 4-week period. The intake of ethanol during the last 4 weeks was registrated by personal interviews and the mean daily consumption calculated. Serum CDT was measured at admission (CDTect) and compared with gamma-glutamyltranspeptidase (GGT), AST, ALT, and mean corpuscular volume (MCV). Serum CDT detected 18 of 26 (69%) patients who consumed > 50 g ethanol daily.

The clinical sensitivity of CDT of detection ethanol consumption > 50 g daily was 69%, compared with 73%, 50%, 35%, and 52% for increased values of GGT, AST, ALT, and MCV, respectively. Altogether, 38 of 476 patients (8%) with a daily ethanol consumption < 50 g also had increased serum CDT levels. The specificity of CDT was 92%, compared with 75%, 82%, 86%, and 85% for GGT, AST, ALT, and MCV, respectively. In the 60 patients who consumed > 10 g ethanol daily, we found a significantly positive correlation between CDT and ethanol consumption (r = 0.52, p < 0.001). A positive correlation was also found between serum transferrin and CDT (r = 0.51, p < 0.001).

In conclusion, the specificity of CDT is much higher compared with GGT in detecting alcohol abuse. Some acute and chronic illnesses may increase the serum level of CDT. False-positive CDT levels may be caused by changes in serum transferrin concentration.


Sex differences of carbohydrate-deficient transferrin, gamma-glutamyltransferase, and mean corpuscular volume in alcohol-dependent patients.

Mundle G, Munkes J, Ackermann K, Mann K.

Addiction Research Center, University of Tubingen, Germany

Alcohol Clin Exp Res 2000 Sep;24(9):1400-5 Abstract quote

BACKGROUND: Biological markers like carbohydrate-deficient transferrin (CDT), gamma-glutamyltransferase (GGT), and mean corpuscular volume (MCV) are used widely to screen for alcoholism. Most research has focused on male alcoholics, and there are few studies on female patients. The results are inconsistent; in general, they show lower sensitivities for all markers for women.

METHODS: We compared the diagnostic value of CDT, GGT, and MCV in 126 alcohol-dependent patients (91 men, 35 women) who entered an inpatient treatment program. For the receiver operating characteristic (ROC) analyses, we investigated a control group of 112 patients (64 men, 38 women) from the Department of Psychiatry at the University of Tubingen with no diagnosis of substance abuse or substance dependency.

RESULTS: Mean levels of CDT and MCV were significantly different in male and female patients. CDT showed higher test results in men (4.4% vs. 2.8%, p < 0.05), whereas mean levels of MCV were higher in women (99.7 fl vs. 96.4 fl,p < 0.01). The sensitivities of CDT and GGT were higher in men than in women (CDT: 76% vs. 54%,p < 0.1; GGT: 68% vs. 43%,p < 0.05), and the sensitivity of MCV was significantly higher in women (71% vs. 41%,p < 0.01). The superiority of MCV in women also was supported by ROC analyses (p < 0.01). The combined use of markers showed satisfactory sensitivity rates of > or = 80% not only in men but also in women. Yet, the specificity rates were partly below the recommended 90% for identifying alcohol abuse; therefore, these markers must be combined with caution.

CONCLUSIONS: If combined, the biological markers CDT and GGT are useful diagnostic instruments for both alcohol-dependent men and women. According to our results, the "forgotten" marker MCV is superior in women and is a marker of second choice in men. The combination GGT and MCV is the most cost-effective choice for men and women.

Henry JB. Clinical Diagnosis and Management by Laboratory Methods. Twentieth Edition. WB Saunders. 2001.

Commonly Used Terms

Basic Principles of Disease
Learn the basic disease classifications of cancers, infections, and inflammation

Commonly Used Terms
This is a glossary of terms often found in a pathology report.

Diagnostic Process
Learn how a pathologist makes a diagnosis using a microscope

Surgical Pathology Report
Examine an actual biopsy report to understand what each section means

Special Stains
Understand the tools the pathologist utilizes to aid in the diagnosis

How Accurate is My Report?
Pathologists actively oversee every area of the laboratory to ensure your report is accurate

Internet Links
Last Modified 4/29/2003

Send mail to The Doctor's Doctor with questions or comments about this web site.
Copyright © 2004 The Doctor's Doctor