This research investigated the association between facial flushing after drinking and alcohol-induced biomarker abnormalities.

This retrospective study included 374 male drinkers who visited the department of Family Medicine of Chungnam National University Hospital between January and December of 2010. The participants were classified into two groups: the flushing group (n = 107) and the non-flushing group (n = 267). The biomarkers assessed were % carbohydrate-deficient transferrin (CDT) and gamma glutamyl transferase (rGTP). The upper limits of %CDT and rGTP were set as 2.47 and 50, respectively. The receiver operating characteristic (ROC) curve was used to obtain the cut-off value for the amount of drinking that caused abnormal %CDT and rGTP levels in the two groups. The sensitivity and specificity of the cut-off drinking amount for %CDT and rGTP abnormalities were analyzed in each group.

In the flushing group, the cut-off value for alcohol-induced %CDT abnormality was 3.38 drinks (1 drink: 14 g of alcohol) per week, with sensitivity of 77.8% and specificity of 70.4%. In the non-flushing group, the cut-off value was 11.25 drinks per week, with sensitivity of 62.2% and specificity of 69.6%. The cut-off value for the amount of alcohol that induced rGTP abnormality was 3.38 drinks per week in the flushing group, with sensitivity of 68.0% and specificity of 76.8%, whereas it was 8.75 drinks in the non-flushing group, with sensitivity of 71.1% and specificity of 66.7%. The area under the ROC of the drinking level was 0.726 in the flushing group and 0.684 in the non-flushing group for %CDT. For rGTP, the value was 0.738 in the flushing group and 0.718 in the non-flushing group.

The weekly drinking amount required to induce biomarker abnormalities was lower in the flushers than in the non-flushers.