Food frequency questionnnaire as an indicator of the serum composition of essential n3 and n6 polyunsaturated fatty acids in early pregnancy, according to body mass index

Background: We investigated whether food frequency questionnaire (FFQ) may be indicative of the serum composition of essential n 3 and n 6 polyunsaturated fatty acids (PUFAs) in early pregnancy and if correlations are affected by body mass index (BMI).
Methods: The present study comprised a prospective cohort conducted in Rio de Janeiro, Brazil. The sample was composed of 248 women, aged 20 to 40 years, between 6 and the 13 weeks of gestation. Dietary intake was assessed using a validated FFQ. Fatty acid serum compositions were determined in fasting serum samples, employing a high-throughput robotic direct methylation coupled with fast gas liquid chromatography. Spearman’s correlation (rs) was used to assess the relationship between fatty acid intake and corresponding serum composition. Women were classified according to BMI (kg m minus 2) as underweight per normal weight or excessive weight .
Results: In the total sample, dietary report was significantly correlated with the serum composition of total polyunsaturated fatty acid, linoleic acid, eicosapentaenoic acid and docosahexaenoic acid. When analyses were stratified by BMI, significant correlations between FFQ and serum composition among underweight/normal weight women were observed for total PUFA, LA , EPA and DHA. Among women of excessive weight, significant correlations were observed only for alpha linolenic acid and DHA.
Conclusions: FFQ in early pregnancy may be used as a possible indicator of serum concentrations of fatty acids. Higher correlations were observed among underweight/normal weight women. (Authors abstract)

Brazilian studies reveal low dietary intake and inadequate n 3 PUFA status among pregnant women, especially with regard to DHA.  In nutritional investigations, it is very important to evaluate the accuracy of the methods for estimating the dietary intake of specific nutrients, such as DHA and EPA during pregnancy. Food frequency questionnaires (FFQ) are the most common method used in epidemiological studies. However, this method has a number of limitations that affect both the accuracy and the precision of the dietary intake estimation. One of these limitations at the individual level is the possibility of under- or overestimation by women and overweight subjects. This limitation can be overcome by using biological markers of specific dietary nutrients. Although there is no biological marker for total fat intake, the fatty acid composition of serum and plasma reflect the fatty acid intake over the past few hours and months and may be used to validate the results from a dietary assessment instrument. The present study aimed to assess whether FFQ may be indicative of the serum composition of essential n 3 and n 6 PUFAs in early pregnancy and to determine whether the correlations were affected by body mass index (BMI). Significant Spearman’s correlations were observed for the concentrations of total PUFAs, EPA, DHA and LA. The correlation coefficients for PUFAs, EPA and LA were higher for the underweight/normal weight group compared to the excessive weight group.  A novel finding of the present study is that correlation coefficients were affected by BMI.
The results showed a weak correlation between fatty acid intake as estimated by the FFQ and maternal serum composition. These results are in agreement with those observed in other studies of FFQs among non pregnant and pregnant women.  Moderate correlations were noted between dietary intake of EPA and DHA estimated with an FFQ and the concentrations of EPA and DHA in serum.  Studies have reported significant positive correlations between the fatty acid composition of erythrocytes and serum samples; therefore, serum may also be considered as a reliable biological marker.

The sensitivity of the FFQ for classifying women into the same or adjacent quartiles for serum fatty acid and dietary intake was 78.22 per cent for EPA and 68.14 per cent for DHA, whereas misclassification into extreme quartiles was 2.82 per cent for EPA and 7.25 per cent. Similar results were observed and reported an efficacy of the FFQ for DHA and EPA of 83 per cent and 78.1 per cent, respectively.  In conclusion, the findings reported in the present study demonstrate that FFQ may be used as a possible indicator of the serum concentrations of fatty acids (especially EPA, DHA and LA) in early pregnancy. FFQ is a non invasive instrument that can be used in routine prenatal for providing nutritional counselling early in pregnancy with respect to types of fats, dietary sources of DHA and EPA, as well as the importance of these nutrients for foetal development and health. (Editors comments)