Energy expenditure in children predicted from heart rate and activity calibrated against respiration calorimetry

The purpose of this study was to predict energy expenditure (EE) from heart rate (HR) and activity calibrated against 24-h respiration calorimetry in 20 children. HR, oxygen consumption ((V) over dot O-2), carbon dioxide production ((V) over dot CO2), and EE were measured during rest, sleep, exercise, and over 24 h by room respiration calorimetry on two separate occasions. Activity was monitored by a leg vibration sensor. The calibration day (day 1) consisted of specified behaviors categorized as inactive (lying, sitting, standing) or active (two bicycle sessions). On the validation day (day 2), the child selected activities. Separate regression equations for (V) over dot O-2, (V) over dot CO2, and EE for method 1 (combining awake and asleep using HR, HR2, and HR3), method 2 (separating awake and asleep), and method 3 (separating awake into active and inactive, and combining activity and HR) were developed using the calibration data. For day 1, the errors were similar for 24-h (V) over dot O-2, (V) over dot CO2, and EE among methods and also among HR, HR2, and HR3. The methods were validated using measured data from day 2. There were no significant differences in HR, (V) over dot O-2, (V) over dot CO2, respiratory quotient, and EE values during rest, sleep, or over the 24 h between days 1 and 2. Applying the linear KR equations to day 2 data, the errors were the lowest with the combined HR/activity.method (-2.6 +/- 5.2%, -4.1 +/- 5.9%, -2.9 +/- 5.1% for (V) over dot O-2, (V) over dot CO2, and EE, respectively). To demonstrate the utility of the HR/activity method, HR and activity were monitored for 24 h at home (day 3). Free-living EE was predicted as 7,410 +/- 1,326 kJ/day. In conclusion, the combination of HR and activity is an acceptable method for determining EE not only for groups of children, but for individuals.