Measures of cardiorespiratory fitness in relation to measures of body size and composition among children

Background In the exercise testing measures of cardiorespiratory fitness need to be scaled by body size or composition to enable comparison between individuals. Traditionally used weight-proportional measures are potentially confounded by body adiposity that hampers their interpretation and applicability in the clinical assessment of cardiorespiratory fitness. Objective We aimed to find the most appropriate measure of body size or composition for scaling of measures of cardiorespiratory fitness among children. Methods We assessed body weight and height, maximal workload (W-MAX) and maximal oxygen uptake (VO2MAX) using cycle ergometer exercise test with respiratory gas analysis and body lean mass (LM) and fat mass (FM) by dual-energy X-ray absorptiometry and by bioimpedance analysis among 38 children. The data were analysed using Pearson's coefficients for correlation and stepwise linear regression models. Results Lean mass (r > 0.54) and height (r > 0.51) had stronger positive correlations with absolute W-MAX and VO2MAX than weight (r > 0.30) in girls and boys. None of the measures of body size or composition correlated with LM proportional W-MAX or VO2MAX in girls or boys. Only LM correlated positively with height-proportional W-MAX (r = 0.65) and VO2MAX (r = 0.71) in boys. FM correlated negatively with weight: proportional W-MAX (r < 0.58) and VO2MAX (r < -0.64) in girls and boys. FM was even stronger determinant of weight-proportional W-MAX (beta = -0.68) and VO2MAX (beta = -0.61) than exercise performance in multivariate linear regression models. Conclusions While assessing cardiorespiratory fitness, LM is the most appropriate measure of body size or composition for scaling of W-MAX and VO2MAX, because scaling by body weight introduces confounding by body adiposity.