The Development of a Resting Metabolic Rate Prediction Equation for Professional Male Rugby Union Players

Abstract

Determining resting metabolic rate (RMR) is an important aspect when calculating energy requirements for professional rugby union players. Prediction equations are often used for convenience to estimate RMR. However, the accuracy of current prediction equations for professional rugby union players remains unclear. The aims of this study were to examine the RMR of professional male rugby union players compared to nine commonly used prediction equations and develop and validate RMR prediction equations specific to professional male rugby union players. One hundred and eight players (body mass (BM) = 102.9 ± 13.3 kg; fat-free mass (FFM) = 84.8 ± 10.2 kg) undertook Dual-energy X-ray Absorptiometry scans to assess body composition and indirect calorimetry to determine RMR. Mean RMR values of 2585 ± 176 kcal∙day^-1 were observed among the group with forwards (2706 ± 94 kcal·day^-1), demonstrating significantly (p < 0.01; d = 1.93) higher RMR compared to backs (2465 ± 156 kcal·day^-1), which appeared to be due to their higher BM and FFM measures. Compared to the measured RMR for the group, seven of the nine commonly used prediction equations significantly (p < 0.05) under-estimated RMR (-104-346 kcal·day^-1), and one equation significantly (p < 0.01) over-estimated RMR (192 kcal·day^-1). This led to the development of a new prediction equation using stepwise linear regression, which determined that the strongest predictor of RMR for this group was FFM alone (R² = 0.70; SEE = 96.65), followed by BM alone (R² = 0.65; SEE = 104.97). Measuring RMR within a group of professional male rugby union players is important, as current prediction equations may under- or over-estimate RMR. If direct measures of RMR cannot be obtained, we propose the newly developed prediction equations be used to estimate RMR within professional male rugby union players. Otherwise, developing team- and/or group-specific prediction equations is encouraged.

Keywords: body composition; energy demands; indirect calorimetry; resting energy expenditure.

Figure 1

Bland–Altman plots for FFM equations: (a) Cunningham, (b) ten Haaf FFM, (c) Mackenzie-Shalders LBM, (d) Tinsley FFM, (e) Posthumus FFM. The middle, solid, horizontal line represents the constant error. The lower and upper dashed horizontal lines represent the 95% limits of agreement.

Figure 2

Bland–Altman plots for BM equations: (a) Harris–Benedict, (b) ten Haaf BM, (c) Jagim, (d) Mackenzie-Shalders BM, (e) Tinsley BM, (f) Posthumus BM. The middle, solid, horizontal line represents the constant error. The lower and upper dashed horizontal lines represent the 95% limits of agreement.

Figure 3

Relationship between (a) fat-free mass (FFM), (b) body mass (BM), and measured resting metabolic rate (RMR). The newly developed (c) FFM and (d) BM prediction equations plotted against measured RMR.