Validade relativa de um questionário de frequência alimentar, padrões alimentares e fatores associados à biomarcadores inflamatórios em adolescentes

Abstract

Eating practices represent one of the main components of the lifestyle to be favor the development of obesity and its comorbidities and metabolic disorders important factors that trigger the inflammatory process. The first article of this thesis aimed at identifying the main dietary patterns (BPs) of adolescents and verifying association with inflammatory biomarkers. It was a cross-sectional study with 391 adolescents aged 17 and 18 from public schools in São Luís-MA. Were build latent variables “Socioeconomic Conditions” (SES) and “inflammation”, consisting of the biomarkers interleukin-1β, interleukin-6 and interleukin-8. Dietary patterns were identified by factor analysis by main components. Modeling of structural equations to assess factors associated with inflammatory biomarkers. Three PAs were extracted (western, basic Brazilian and healthy). Not found association between PAS and “inflammation”. However, excess weight was positively associated with “inflammation” (CP = 0.281; p = 0.002). The second article objective to evaluate the relative validity of the Food Frequency Questionnaire (FFQ) used in the RPS Cohort to assess food intake of adolescents in São Luís- BAD. Study developed with 152 adolescents aged 17 and 18 years old. To the validation of the FFQ, the average of three 24-hour recalls (R24h) was used as reference method. The paired t-Student test was used to verify differences between the instruments. To measure concordance, coefficients of Pearson correlation, intraclass correlation coefficient (ICC), weighted Kappa and Bland-Altman graphics. Significance level <5% was adopted. When compared with R24h, the FFQ evaluated overestimated the consumption of most nutrients. You Pearson's correlation coefficients after adjustment for energy consumption and attenuation ranged from 0.06 - 0.43 and, despite the weak / moderate correlation, were significant for the nutrients iron, calcium, riboflavin, sodium, saturated fatty acids, niacin and vitamin C. Adjusted and diluted intraclass correlation coefficients (ICC) varied from 0.01 - 0.31 and the weighted kappa of the energy-adjusted variables ranged from 0.01 - 0.46. Agreement analyzes were significant for Vitamin C, fiber, calcium, riboflavin, niacin and sodium, lipids and iron. This FFQ was relatively valid acceptable for lipid nutrients, saturated fatty acids, fibers, calcium, iron, riboflavin, niacin, vitamin C and sodium. The third article aimed to identify adolescents' PAs and assess factors associated with inflammatory biomarkers, through a cross-sectional study aligned with the RPS cohort, with data from São Luís-MA. 511 adolescents aged 18 and 19 years were evaluated. PAs were identified by factor analysis by main components. To assess nutritional status, we used Body Mass Index (BMI) and body fat percentage (% BF), assessed by the air displacement plethysmography method. Modeling was done hierarchized using Linear Regression to estimate the Beta coefficient (β) of independent variables with the dependent variables, Interleukin-6 (IL-6) and Ultra-sensitive C-reactive protein (hs-CRP). Five dietary patterns were identified: energy-dense, sugary drinks and breakfast cereals, prudent, common Brazilian and alcoholic and energy drinks. The “prudent” pattern pointed to the association with the subclinical inflammation assessed using IL-6, this pattern being considered a protective factor for inflammation (β = -0.11; p value = 0.040). PAs “Common Brazilian” and “alcoholic and energy drinks” were indirectly associated the increase in IL-6, mediated by nutritional status. CRP levels were directly associated with BMI (β = 0.36; p value = <0.001) and% BG (β = 0.02; p value = 0.014). Subclinical inflammation was associated with nutritional status and prudent BP, in which adolescents with higher BMI and% BF had higher concentrations of CRP-us and those with greater adherence to prudent AP had lower concentrations of IL-6, thus being a protector of subclinical inflammation.