Background: Fetuin-A, a liver-derived glycoprotein that impairs insulin-signalling, has emerged as a biomarker for diabetes risk. Although moderate alcohol consumption has been inversely associated with fetuin-A, data from clinical trials are lacking. Thus, we evaluated whether moderate alcohol consumption decreases circulating levels of fetuin-A. Methods. We analyzed data of three separate open-label, randomized, crossover trials: 1) 36 postmenopausal women consuming 250 ml white wine (25 g alcohol) or white grape juice daily for 6 weeks, 2) 24 premenopausal women consuming 660 ml beer (26 g alcohol) or alcohol-free beer daily for 3 weeks, and 3) 24 young men consuming 100 ml vodka (30 g alcohol) orange juice or only orange juice daily for 4 weeks. After each treatment period fasting blood samples were collected. Results: Circulating fetuin-A concentrations decreased in men after vodka consumption (Mean ± SEM: 441 ± 11 to 426 ± 11 ug/ml, p = 0.02), but not in women after wine (448 ± 17 to 437 ± 17 ug/ml, p = 0.16) or beer consumption (498 ± 15 to 492 ± 15 ug/ml, p = 0.48) compared to levels after each corresponding alcohol-free treatment. Post-hoc power analyses indicated that the statistical power to detect a similar effect as observed in men was 30% among the postmenopausal women and 31% among the premenopausal women. Conclusions: In these randomized crossover trials, moderate alcohol consumption decreased fetuin-A in men but not in women. This sex-specific effect may be explained by the relatively short intervention periods or the low statistical power in the trials among women. Trials registration. ClinicalTrials.gov ID no's: NCT00285909, NCT00524550, NCT00918918. © 2014 Joosten et al.; licensee BioMed Central Ltd.Chemicals/CAS: adiponectin, 283182-39-8; alanine aminotransferase, 9000-86-6, 9014-30-6; alcohol, 64-17-5; alkaline phosphatase, 9001-78-9; aspartate aminotransferase, 9000-97-9; gamma glutamyltransferase, 85876-02-4; glucose, 50-99-7, 84778-64-3; insulin, 9004-10-8