Detection of relapses in alcohol-dependent patients using carbohydrate-deficient transferrin: improvement with individualized reference levels during long-term monitoring.

The present study examined whether the sensitivity of carbohydrate-deficient transferrin (CDT) in serum, a biochemical marker of recent excessive alcohol consumption, could be improved during long-term monitoring by introducing individualized cut-offs between normal and elevated CDT levels. Alcohol-dependent male outpatients (n = 22), trying to abstain from alcohol for 6 months, were monitored by comparing weekly measurements of CDT with self-reports of alcohol consumption three times/week and daily urinary levels of 5-hydroxytryptophol (5-HTOL), a new marker of recent alcohol intake. The method used to calculate cut-offs was based on the intraindividual variation in CDT not dependent on excessive alcohol consumption or analytical variations. An increase in CDT exceeding the minimum level for each patient by 3 and 4 times the mean coefficient of variation for healthy social drinkers (i.e., by 30% and 40%) was compared as an indication of alcohol consumption, even if the value did not exceed the conventional cut-off. By using individualized CDT cut-off points, 68 and 41 episodes of drinking were detected in the patients with the cut-offs of > 30% and > 40%, respectively, as compared with 25 with the conventional limit. Most episodes could be verified clinically and/or by elevated urinary 5-HTOL levels during the 2-week period preceding each serum sampling. The results suggest that the possibility to detect relapses by CDT can be improved during long-term monitoring of alcohol-dependent outpatients by introducing individualized cut-off points between normal and elevated CDT levels.