In our sample of overweight and obese Chilean adolescents, we found that inflammation significantly mediated the relationship between two measures of adiposity, WC and FMI, and increasing levels of HOMA-IR. The association between adiposity and hs-CRP has been extensively studied [13, 14]. Likewise, inflammation has been associated with IR . However to our knowledge, there is no other human study showing inflammation as a potential pathophysiologic link between obesity and metabolic derangement, an association well-described in cell-culture and in animal models [16, 17].
Gender differences in the relationship between adiposity and inflammation have been previously described in adults with waist-to-hip ratio related to inflammation in men and FM having a stronger association with inflammation among women . Our results make an important contribution, as they indicate that gender differences, described in adults, may also occur in adolescents. During adolescence important sex differences in body composition emerge. Females have more peripheral fat; therefore, WC may not best reflect total fat. On the other hand, males have lower FM, but are likely to have more visceral fat . Measures of adiposity are imperfect and are differentially related to visceral fat. Thus, it is important to choose the measure of adiposity most associated with later adverse outcomes.
Several limitations should be noted. Our findings cannot be generalized to normal weight adolescents or to adolescents from other backgrounds (not Chilean). Future studies should include more direct measures of visceral and subcutaneous fat content (e.g. magnetic resonance or ultrasound imaging), and additional measures of inflammatory status (e.g. interleukin 6, tumor necrosis factor alpha, adiponectin) and insulin sensitivity. It would also be important to assess the magnitude of other factors that might be related to inflammatory status and metabolic performance, for example non-alcoholic steatohepatitis which often occurs with obesity and IR and can progress to inflammation and fibrosis [19, 20]. The strengths of the study include careful anthropometric assessment and reliable measurement of FM. Moreover, our subjects were recruited from a community population, rather than using a clinical sample.
In conclusion, our sample of overweight and obese Chilean adolescents showed gender-specific associations between adiposity and systemic inflammation. Additionally, we found that hs-CRP statistically mediated the association between adiposity and IR. This suggests that a systemic inflammatory state is initiated by FM excess even during adolescence and that inflammation influences the metabolic consequences of overweight and obesity. Future research should include 1) detailed assessment of adiposity topography, 2) additional assessments of inflammation (e.g., inflammatory infiltration of adipose tissue) and 3) studies in other populations.