Associations of circulating irisin concentrations with cardiometabolic risk factors among children vary by physical activity or sedentary time levels

Whether irisin concentrations are associated with physical activity (PA) and sedentary time (ST) remains unknown. The role of irisin on cardiometabolic health among children has been contradictory and scarce. This study aimed to examine associations of PA and ST with irisin concentrations and relationships between irisin concentrations and cardiometabolic parameters among children. Additionally, we assessed the interaction between PA or ST and irisin concentrations on cardiometabolic parameters. Basing on a cross-sectional survey of 3,651 general children aged 7–12 years, 575 with different self-reported PA (moderate-vigorous intensity PA ≥ 60 min/day or <150 min/week) and ST (gender-, age-specific ST ≥ 75% or <25% percentile) levels were selected. PA and ST were assessed by the validated international physical activity questionnaires. Fasting blood glucose and lipid profile levels were measured with standard methods by biochemistry analyzer. Plasma irisin concentrations were measured by ELISA. The associations of PA and ST with circulating irisin concentrations were examined by linear regression. Linear regression analysis was also used to estimate associations of circulating irisin concentrations with cardiometabolic variables. Interactions between PA or ST and irisin concentrations on cardiometabolic parameters were calculated using multiple linear regression models with dichotomized factors (low PA and high PA; low ST and high ST). No significant association was observed between circulating irisin concentrations and habitual PA or ST. Irisin concentrations were negatively associated with body mass index (BMI) (β = −0.220), BMI z-score (β = −0.098), waist circumference (β = −0.621), diastolic blood pressure (DBP) (β = −0.561), and triglyceride (β = −0.019) in low PA subgroup, and negatively related to fasting blood glucose (β = −0.040) among high PA subgroup. Moreover, irisin concentrations were negatively associated with BMI (β = −0.157) and fasting blood glucose (β = −0.026) only in high ST subgroup (all P < 0.05). We also observed a significant interaction between PA and irisin concentrations on BMI (Pinteraction = 0.0350), BMI z-score (Pinteraction = 0.0173), and DBP (Pinteraction = 0.0068). In summary, irisin concentrations were not associated with habitual PA or ST in children. The negative associations of irisin concentrations with BMI, BMI z-score, and DBP were found only among children being inactive, implying that irisin may contribute to an improvement in health, especially among children with unhealthy lifestyles. Copyright © 2019 Cai, Tan, Tan, Zeng, Wan, Wong, O’Reilly, Sun, Yang and Chen.