PURPOSE: Dietary fatty acids affect several pregnancy outcomes including fetal growth and development. We compared self-reported intakes with concentrations of fatty acids in adipose tissue in pregnant women. METHODS: The study was nested within Geração XXI, a birth cohort assembled in Portugal. Intake was assessed by nine food diaries (FDs) completed throughout pregnancy and an FFQ administered in the immediate postpartum period. A gluteal adipose tissue sample was obtained from 23 women. RESULTS: FDs and FFQ estimated similar percentages of saturated (SFA), monounsaturated (MUFA), and polyunsaturated fatty acids (PUFA), but the adipose tissue yielded a lower percentage of SFA and higher percentages of MUFA and PUFA. Correlations between FDs and adipose tissue ranged from r = 0.50 for trans fatty acids to r = -0.19 for linolenic acid. The proportion of women categorized in opposite tertiles by these two methods ranged from 4.3% to 30.4%. Correlations between FFQ and adipose tissue were even weaker and levels of misclassification higher. CONCLUSIONS: The correlations observed in this study between self-reported intakes and tissue concentrations are weaker than those observed in a similar study conducted among nonpregnant women, suggesting that adipose tissue levels of fatty acids may be a poor biomarker of dietary intake in pregnancy. 2010 Elsevier Inc. All rights reserved.
PURPOSE:Dietary fatty acids affect several pregnancy outcomes including fetal growth and development. We compared self-reported intakes with concentrations of fatty acids in adipose tissue in pregnant women. METHODS: The study was nested within Geração XXI, a birth cohort assembled in Portugal. Intake was assessed by nine food diaries (FDs) completed throughout pregnancy and an FFQ administered in the immediate postpartum period. A gluteal adipose tissue sample was obtained from 23 women. RESULTS: FDs and FFQ estimated similar percentages of saturated (SFA), monounsaturated (MUFA), and polyunsaturated fatty acids (PUFA), but the adipose tissue yielded a lower percentage of SFA and higher percentages of MUFA and PUFA. Correlations between FDs and adipose tissue ranged from r = 0.50 for trans fatty acids to r = -0.19 for linolenic acid. The proportion of women categorized in opposite tertiles by these two methods ranged from 4.3% to 30.4%. Correlations between FFQ and adipose tissue were even weaker and levels of misclassification higher. CONCLUSIONS: The correlations observed in this study between self-reported intakes and tissue concentrations are weaker than those observed in a similar study conducted among nonpregnant women, suggesting that adipose tissue levels of fatty acids may be a poor biomarker of dietary intake in pregnancy. 2010 Elsevier Inc. All rights reserved.