Validity of predictive equations for resting energy expenditure in Belgian normal weight to morbid obese women

Abstract

Individual energy requirements of overweight and obese adults can often not be measured by indirect calorimetry, mainly due to the time-consuming procedure and the high costs. To analyze which resting energy expenditure (REE) predictive equation is the best alternative for indirect calorimetry in Belgian normal weight to morbid obese women. Predictive equations were included when based on weight, height, gender, age, fat free mass and fat mass. REE was measured with indirect calorimetry. Accuracy of equations was evaluated by the percentage of subjects predicted within 10% of REE measured, the root mean squared prediction error (RMSE) and the mean percentage difference (bias) between predicted and measured REE. Twenty-seven predictive equations (of which 9 based on FFM) were included. Validation was based on 536 F (18-71 year). Most accurate and precise for the Belgian women were the Huang, Siervo, Muller (FFM), Harris-Benedict (HB), and the Mifflin equation with 71%, 71%, 70%, 69%, and 68% accurate predictions, respectively; bias -1.7, -0.5, +1.1, +2.2, and -1.8%, RMSE 168, 170, 163, 167, and 173kcal/d. The equations of HB and Mifflin are most widely used in clinical practice and both provide accurate predictions across a wide range of BMI groups. In an already overweight group the underpredicting Mifflin equation might be preferred. Above BMI 45kg/m(2), the Siervo equation performed best, while the FAO/WHO/UNU or Schofield equation should not be used in this extremely obese group. In Belgian women, the original Harris-Benedict or the Mifflin equation is a reliable tool to predict REE across a wide variety of body weight (BMI 18.5-50). Estimations for the BMI range between 30 and 40kg/m(2), however, should be improved.