Does cardiorespiratory fitness moderate the prospective association between physical activity and cardiometabolic risk factors in children?

BACKGROUND/OBJECTIVES: Physical activity (PA) and cardiorespiratory fitness (CRF) are independently associated with reduced cardiometabolic risk in children, and may affect risk through different pathways. This study aims to examine if CRF moderate the prospective association between PA, sedentary time, and cardiometabolic outcomes in 10-year-old children.

SUBJECTS/METHODS: In total, 718 children of 1129 (drop out n = 7) had valid measures of PA (accelerometry), CRF (the Andersen running test), and a cardiometabolic risk profile measured at baseline and follow-up 7 months later. Cardiometabolic outcomes were systolic blood pressure, waist circumference (WC), total cholesterol, high-density lipoprotein, triglycerides, glucose, and insulin (HOMA-IR). The cardiometabolic risk factors were analysed individually, and as a clustered risk score (z score). A linear mixed model was used to examine the prospective associations between different PA exposures (overall PA, sedentary time, moderate-to-vigorous PA (MVPA), vigorous PA) and cardiometabolic outcomes, including the interaction term PA × CRF in the model to assess moderation by CRF.

RESULTS: CRF modified the association for baseline overall PA (P < 0.039) and MVPA (min/day) with clustered cardiometabolic risk at follow-up (P < 0.023). Moreover, CRF modified the association between overall PA and MVPA with HOMA-IR independent of WC (P < 0.022). When stratified by CRF level (median split; high/low), MVPA predicted lower HOMA-IR [MVPA β -0.133 (95% CI: -0.223, -0.043); P = 0.004] and clustered cardiometabolic risk [MVPA β -0.094 (95% CI: -0.169, -0.019); P = 0.014] in children with low CRF, but not among their fitter peers (P > 0.232). There was neither direct association between sedentary time and cardiometabolic risk factors in any analyses, nor moderation by CRF.

CONCLUSION: CRF significantly moderated the prospective association between PA and the clustered cardiometabolic risk, but not for time spent sedentary. The magnitude of association between MVPA and clustered cardiometabolic risk was stronger in children with low CRF, and no associations appeared present in their high-fit peers.