Assessing sleeping energy expenditure in children using heart-rate monitoring calibrated against open-circuit indirect calorimetry: a pilot study

Total energy expenditure (EE) can be assessed in children by the heart-rate (HR) monitoring technique calibrated against open-circuit indirect calorimetry (IC). In this technique, sleeping EE is usually estimated as the lowest value of a 30 min resting EE measurementx0.90 to give an average for the total sleeping period. However, sleeping is a dynamic process in which sleeping EE is modulated by the effect of factors such as body movement and different sleep stages. The aim of the present study was to determine a new method to improve the sleeping EE measurement by taking into account body movements during sleep. Twenty-four non-obese children participated in the present study. All subjects passed through a calibration period. HR and EE measured by IC were simultaneously collected during resting, the postprandial period, and during different levels of activity. Different methods for computing sleeping EE (resting EEx0.90 with different breakpoints ('flex point' HR with linear regression or 'inflection point' (IP) HR with the third order polynomial regression equation (P3)) were compared with EE measured for least 2.0 h in eight sleeping children. The best method of calculation was then tested in sixteen children undergoing HR monitoring and with a body movement detector. In a subset of eight children undergoing simultaneous sleeping EE measurement by IC and HR, the use of the equation resting EEx0.8 when HR<IP and P3 when HR>IP during the sleeping period gave the lowest difference (1 (sd 5.4) %) compared with other methods (linear or polynomial regressions). The new formula was tested in an independent subset of sixteen other children. The difference between sleeping EE computed with the formula resting EEx0.90 and sleeping EE computed with resting EEx0.8 when HR<IP and the P3 equation when HR>IP during sleeping periods was significant (13 (sd 5.9) %) only for active sleeping subjects (n 6 of 16 subjects). The correlation between the difference in the results from the two methods of calculation and body movements was close (r 0.63, P<0.005, Spearman test) as well as computed sleeping EE (Spearman test, r 0.679, P<0.001), indicating that this new method is reliable for computing sleeping EE with HR monitoring if children are moving during sleep and improves the total EE assessment.