Background: Autonomic activity is possibly influenced by physical activity (PA). However, it remains unclear whether this association is modified by insulin resistance.
Methods: This population-based study between 2009 and 2012 included 2016 men and women aged 30-79 years. The PA was assessed using a validated questionnaire based on sleep, occupation, transportation, household characteristics, and leisure-time PA. Heart rate (HR) and heart rate variability (HRV) in the sitting position were determined from 5-minute recordings of pulse waves detected by a fingertip sensor. The HRV was calculated as frequency (standard deviation of normal-to-normal [NN] intervals [SDNN]), root mean square of successive differences (RMSSD), and percentage differences between normal NN intervals >50 milliseconds [pNN50]) and time domains. Insulin resistance was evaluated using the homeostasis model assessment index (HOMA-IR).
Results: HR, RMSSD, and pNN50 were related to the total and moderate/vigorous PA tertiles in models that included HOMA-IR. The partial regression coefficient of total PA per 1-SD increase was .05 (P = .019) for log-transformed RMSSD and 1.86 (P = .001) for pNN50. No interactive associations were observed between PA and HOMA-IR.
Conclusions: Low total PA was associated with increased HR and low levels of RMSSD and pNN50, reflecting parasympathetic modulation that was not modified by insulin resistance.
Keywords: HOMA-IR; autonomic function; epidemiology; exercise.