Inflammation biomarkers and neurobehavioral performance in rural adolescents

Background

Systemic inflammation has been associated with lower neurobehavioral performance in diverse populations, yet the evidence in adolescents remains lacking. Cytokines can alter neural network activity to induce neurocognitive changes. This work seeks to investigate the association between inflammation and neurobehavior in adolescents living in a rural region of Ecuador.

Methods

We examined 535 adolescents in rural communities of Ecuador (ESPINA study), 508 of which had neurobehavioral assessments (NEPSY-II) and circulating plasma levels of inflammatory markers (CRP, IL-6, TNF-⍺, sICAM-1, sVCAM-1, SAA, and sCD14). Associations between inflammatory biomarker concentrations and neurobehavioral scores were examined using adjusted bivariate semi-parametric models with generalized estimating equations. A partial least squares regression approach was used to create composite variables from multiple inflammation biomarkers and model their association with cognitive outcomes.

Results

Higher sCD14 and TNF-α concentrations were significantly associated with lower social perception scores, by −0.465 units (95% CI: −0.80, −0.13) and −0.418 units (−0.72, −0.12) for every 50% increase in inflammatory marker concentration, respectively. Similarly, every 50% increase in the inflammation summary score was associated with a significantly lower Social Perception score by −0.112 units (−0.19, −0.03). A greater inflammatory composite variable from seven markers was associated with lower scores in language (β = −0.11, p = 0.043), visuospatial processing (β = −0.15, p = 0.086), and social perception (β = −0.22, p = 0.005) domains.

Conclusions

Higher levels of inflammation were associated with lower neurobehavioral performance in adolescents, especially with social perception. In addition, using a robust analytic method to examine an association between a composite inflammatory variable integrating seven markers led to additional findings, including the domains of language and visuospatial processing. A longitudinal follow-up of such investigations could unveil potential changes in inflammation-neurobehavior performance links through developmental stages and intervention opportunities.