Background The aim of this study was to compare the results of Phosphatidylethanol (PEth) and carbohydrate‐deficient transferrin (CDT) in blood as biomarkers of alcohol consumption in a large clinical cohort and to evaluate concentrations in relation to age and sex. Methods Results of PEth 16:0/18:1 in blood and CDT in serum were included, together with information of age and sex, which were extracted from a clinical chemistry database containing samples mostly from patients of primary care physicians and social care institutions. PEth concentrations were determined using Ultra Performance Convergence chromatography mass spectrometer. CDT was quantified by electrophoretic Capillary System. CDT values ≥ 1.7 %‐units and PEth values ≥ 0.31 µmol/L were considered to indicate heavy alcohol consumption. Results Samples from 6705 patients were included. The median age was 54.5 years, and 34 % were females. Only 47 % of the patients with PEth ≥ 0.31 µmol/L had increased CDT ≥ 1.7 %‐units examined in the same specimen (Cohen’s kappa was 0.43, p < 0.001). Patients above 50 years had significantly higher concentrations for both CDT (1.0 %‐units vs. 0.9 %‐units, p < 0.001) and PEth (0.340 µmol/L vs. 0.200 µmol/L, p < 0.001) compared with younger patients. Concentrations of CDT were significantly higher in males compared with females (p = 0.002), while no significant sex differences were seen for PEth (p = 0.465). Conclusions A high fraction of the patients had PEth values above the suggested cutoff for heavy drinking and normal CDT values, verifying the superior sensitivity of PEth compared with CDT. The effect of age seems to be minor for both markers. Higher concentrations of CDT, but not PEth, were seen in males, indicating that PEth, as opposed to CDT, might be formed equally in men and women. Therefore, the bias due to sex is possibly present only for CDT, not for PEth.

Aims Falsely lower or even negative phosphatidylethanol (PEth) levels may theoretically be seen in patients with haemolytic diseases, and the present study aimed to elucidate this hypothesis. Methods PEth and carbohydrate‐deficient transferrin (CDT) from 9893 serum and whole blood samples were included along with markers of haemolysis (i.e. haptoglobin, HbA1c, reticulocytes, LD and Hb). Cases showing discrepancy between PEth and CDT, that is, a low PEth value and a high CDT value, were considered to be possibly caused by falsely lowered PEth despite high alcohol consumption. These cases (N = 233) were compared to the control group without PEth and CDT mismatch. Results The levels of haptoglobin were significantly lower in the cases showing low PEth and high CDT (estimate = −0.62, p = 0.002). The levels of HbA1c (estimate = −3.26, p = 0.001) and Hb (estimate = −0.507, p < 0.001) were also significantly lower in this group. These findings indicate haemolytic diseases in the low PEth/high CDT group. There were no significant differences for reticulocytes and LD concentrations between the low PEth/high CDT group and the control group. Conclusions These results indicate that falsely low PEth values could be associated with markers of haemolytic diseases, although more research is needed to highlight this further.

No abstract