Resting energy expenditure is an important physiological parameter that is useful in research involving weight loss because it is a measure of the energy required by the body at rest in a fasted state. Importantly resting energy expenditure can be useful when assessing the daily energy requirement of an individual and changes in resting energy expenditure can provide useful information on the effect of dietary changes and exercise on the utilisation of energy. Indirect calorimetry utilises oxygen and carbon dioxide exchange ratios from a subject in order to give measurements of substrate utilisation rates and are considered the best methods for estimating resting energy expenditure. Often these measuring devices are not available and so predictive equations are used to allow estimates of resting energy expenditure. However, research suggests that predictive equations do not produce good correlation with indirect calorimetry.
The accuracy of predictive equations was tested by one group of researchers1 in a sample of healthy young women (age 20 to 33years with a body mass index in the range of 18 to 24.9kg/m2). Comparison of indirect calorimetry with equations formulated by Nelson (1992), Mifflin (1990), Owen (1986), Schofieldweight (1985) Schofieldweight and height (1985) and Harris-Benedict (1919) resulted in overestimation of the resting energy expenditure by equations (by around 140 to 738 Kcal/d). The authors reported that the number of subjects with predicted resting energy expenditures 10% over the measured value ranged from 74% for the Nelson equation to 100% for the Harris-Benedict equation. Analysis of the equations revealed systematic errors in predicting resting energy expenditure in all equations except the Harris-Benedict equation. Therefore the predictive equations did not show good accuracy when compared to the accepted method of indirect calorimetry.
These results are important because they suggest that if the equations used to estimate resting energy expenditure are flawed, then re-evaluation should be given to research that has used values derived in this way. For example, research comparing estimates of resting energy expenditure from equations to data taken using indirect calorimetry should be treated with caution. However, if the inherent inaccuracy of the predictive equations is taken into account, changes in resting energy expenditure may be acceptable using these equations, as long as consistent use of the same formula is adhered to. Because resting energy expenditure is often used to calculate energy intakes of subjects, and the predictive equations can over estimate this value, care should be taken when estimating resting energy expenditure for this purpose. Because the accuracy of the equations is poor, effort should always be made to measure resting energy expenditure using indirect calorimetry.
