RESUMEN
BACKGROUND: Alcohol consumption is commonly accepted in Western societies and is a known risk factor in pregnancy, which could lead to fetal alcohol spectrum disorders (FASDs). Prevalence of alcohol consumption during pregnancy is mostly unknown. Prevalence estimates in publications based on questionnaires are limited by possible underreporting due to social stigmatization. The aim of this study was to estimate the prevalence of harmful alcohol consumption in a large cohort of pregnant women using different biomarkers related to alcohol consumption and compare the findings with those of non-pregnant women METHODS: Routine parameters known to be influenced by alcohol consumption (γ-glutamyltransferase, GGT; carbohydrate-deficient transferrin, CDT/%CDT; mean corpuscular/cell volume, MCV; combined parameter of GGT and %CDT, GGT-CDT) were analyzed in serum samples of 2,182 pregnant women and 743 non-pregnant, age-matched females. Data were tested for (i) differences between pregnant and non-pregnant women and (ii) changes across the 3 trimesters of pregnancy. RESULTS: Prevalence rates differ greatly according to the parameter and cutoff, which reflects the limitations of assessing alcohol consumption with biomarkers. The prevalence of harmful alcohol consumption on the basis of a single or several elevated parameters was 13.8% (95% CI: 12.4 to 15.2) in pregnant women and 18.6% (95% CI: 15.8 to 21.4) in non-pregnant women, though 85.0% of the elevated measurements were attributable to an isolated elevation in %CDT only. Using GGT-CDT as the parameter with the highest specificity according to the literature, the estimated prevalence of harmful alcohol consumption in pregnancy is 0.5% (95% CI: 0.2 to 0.7). CONCLUSION: Estimated prevalence rates differ greatly with respect to the biomarkers and cutoffs used. The use of CDT/%CDT alone appears to overestimate harmful alcohol consumption during pregnancy.