PURPOSE: Although obesity has been associated with imbalances in cardiac autonomic nervous system, it is unclear whether there are differential relationships between adiposity measures and heart rate variability (HRV) measures. We aimed to examine differences in the relationship between adiposity measures and HRV indices in a healthy Korean population. MATERIALS AND METHODS: In all, 1409 non-smokers (811 males, 598 females) without known histories of cardiovascular (CV), endocrine, or neurological diseases underwent adiposity measurements [(body mass index (BMI), percentage of body fat mass (PBF), and waist-to-hip ratio (WHR)], the HRV assessment (SDNN, RMSSD, LF, HF, LF/HF, and pNN50), and examination for CV risk factors (fasting glucose, LDL-cholesterol, HDL-cholesterol, triglycerides, hs-CRP, and blood pressure). RESULTS: Compared with BMI and PBF, WHR was more strongly correlated with each HRV index and more likely to predict decreased HRV (<15 percentile vs. > or =15 percentile of each HRV index) in ROC curves analysis. In linear regression analysis, all adiposity measures were inversely associated with each HRV measure before adjusting for age, gender, and CV risk factors (p<0.05). After adjusting for the covariates, WHR was inversely related to RMSSD, LF, and pNN50; PBF with RMSSD, HF, and pNN50; BMI with RMSSD (p<0.05). The inversed association between HRV indices and the gender-specific WHR tertile was significant for subjects with BMI > or =25 kg/m2, but not for those with BMI <25 kg/m2. CONCLUSION: WHR and PBF appear to be better indicators for low HRV than BMI, and the association between abdominal adiposity and HRV may be stronger in overweight subjects.
*Adiposity ; Adult ; Female ; Heart/physiopathology ; Heart Diseases/complications/diagnosis/ethnology ; Heart Rate/*physiology ; Humans ; Male ; Middle Aged ; Models, Statistical ; Obesity/*complications/*diagnosis/ethnology ; Overweight/complications/diagnosis/ethnology ; ROC Curve ; Regression Analysis ; Republic of Korea ; Risk Factors