Discrepancies among different tools evaluating Mediterranean diet adherence during pregnancy, correlated to maternal anthropometric, dietary and biochemical characteristics

Background & aims: Scientific evidence confirms the favorable impact of Mediterranean diets (MD) on maternal and neonatal outcomes. However, the assessment of this dietary pattern requires valid indexes with scoring systems adapted to pregnant females. The aim of this cross-sectional study was to compare the adherence to MD, through 5 internationally validated tools, in pregnant women following a Mediterranean eating pattern, correlated to maternal anthropometric, dietary and biochemical markers. Methods: 100 healthy pregnant females completed a sociodemographic questionnaire, a dietary recall and a food frequency questionnaire compatible with the MD.10 ml of maternal blood were drawn for the analysis of biological markers such as C-reactive protein (CRP), leptin and adiponectin. Results: We used the 50th percentile as cut-off of each scale (low or high adherence) to avoid the discrepancies noted in the literature among the large range of cut-offs points for the different tools. The % of agreement was high between the Mediterranean Food Pattern, MD Score, the MD Score and the Short MD Questionnaire. The MD Scale presented small agreement in relation to the other tested tools. All the tested indexes were significantly correlated with CRP levels, except for the MDScale. Significant correlations were reached regarding adiponectin and the MFP (p value = 0.04) and the MDScale (p value 0.03) tools. Pre-gestational body mass index was significantly correlated with all the tested biological markers. Significant correlations were seen between CRP on one hand and maternal age (p value = 0.033), adiponectin (p value = 0.028), and leptin (p value = 0.003) on the other. Fiber intake was significantly and negatively correlated to CRP (p value = 0.008) and positively to adiponectin levels (p value = 0.000). Conclusions: None of the tested tools were adapted for pregnancy, since a-priori scores were attributed for components already not consumed by pregnant females such as alcohol or recommended for daily or weekly consumptions such as whole dairy products and fish, respectively. In addition, the lack of inclusion of some traditional food ingredients of the MD implies the urge to create a new index adapted to pregnancy. (C) 2018 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.