Resting Energy Expenditure in Athletes: Accuracy of Cunningham Prediction Equation Using DXA-Derived Fat Free Mass

Abstract

Introduction: It is recommended to use the fat free mass (FFM) based Cunningham equation to accurately estimate resting energy expenditure (REE) in athletic populations. However, it is unknown if FFM derived from dual-energy x-ray absorptiometry (DXA) can accurately estimate REE in collegiate athletes. Therefore, our aim is to determine REE accuracy of Cunningham equation using DXA-derived FFM compared to measured REE in Division I collegiate athletes. Methods: 15 (14F, 1M) NCAA Division I athletes from 7 collegiate sports completed the test protocol. Indirect calorimetry (Parvo Medics’ TrueOne® 2400 metabolic cart with canopy system) was used to measure REE. A DXA (GE Lunar iDXA enCORE-based) measurement was completed within 14 days of measured REE to provide FFM. The DXA-derived FFM was inserted into Cunningham equation (REE [kcal/day] = 500 + 22 × FFM[kg]) to predict REE. A Wilcoxon signed-rank test and Bland-Altman analysis were used to determine respective difference and bias between measured and predicted REE. Results: Predicted REE by Cunningham equation using DXA-derived FFM was not statistically different from measured REE (Z = -1.306, p = 0.191). Average measured and predicted REE was 1448 kcal/day and 1587 kcal/day. Mean percent difference was ±20.09% with 47% of participants REE difference ±10%. Bland-Altman analysis suggested proportional bias, with overestimations among low measured REE values (~1200 kcal/day) and underestimations among high measured REE values (~1800 kcal/day). Conclusions: The Cunningham equation, using DXA-derived FFM to predict REE, was not significantly different from measured REE in Division 1 collegiate athletes. However, reported bias and possible clinical significance (i.e. % difference) warrant further investigation to validate our preliminary findings.