Dietary Inflammatory Index is positively associated with hs-CRP and inversely associated with IL-8 in Brazilian adults: a cross-sectional study

Evidence suggests that diet is associated with low-grade inflammation. The Dietary Inflammatory Index (DII) assesses the inflammatory potential of a diet; however, few studies have examined its association with hs-CRP and cytokines (IL-1β, IL-6, IL-8, IL-10, IL-12p70, and TNF-α) in Brazilian adults. This cross-sectional study, using data from 958 adults from the Viçosa Health and Nutrition Study (2012-2014), who were selected by two-stage probabilistic sampling. Dietary intake was assessed using a validated Food Frequency Questionnaire. DII scores were calculated, as were the energy-adjusted (E-DII) and residual-adjusted (E-DII_r) forms based on 25 dietary components. Inflammatory biomarkers were measured using high-sensitivity immunoturbidimetry and bead-based flow cytometry for cytokines. Linear and logistic regression models, adjusted for complex sampling design and covariates defined by a Directed Acyclic Graph (DAG), were tested in three hierarchical blocks (Stata 14.2; p < 0.05). Diets with higher inflammatory potential (E-DII, E-DII_r) were positively associated with log-transformed hs-CRP concentrations in adjusted models. In quartile analyses, individuals in the most pro-inflammatory quartile (DII, E-DII, E-DII_r) had higher hs-CRP concentrations. Logistic models showed no association between continuous scores and odds of hs-CRP ≥ 3 mg/L. However, significant associations emerged in the upper quartiles of the DII and E-DII_r. For IL-8, inverse associations appeared in both linear (DII, E-DII_r) and ordinal logistic regression (DII, E-DII_r). Findings support the association between pro-inflammatory diets and hs-CRP concentrations, with no consistent results for other cytokines. The DII may capture subclinical inflammation, but its utility could depend on methodology and timing, underscoring the need for standardized longitudinal studies.