Response

To the Editor-in-Chief, In their letter to the editor, Telford et al. (1) continued a very important discussion on the best practice to scale cardiorespiratory fitness (CRF) by body size and composition. There has been a critical debate on the most appropriate and meaningful method to control variability in body size and composition as well as maturity status when assessing CRF (2,3). In line with our view, Welsman and Armstrong (2), Blais et al. (3), and Shaibi et al. (4) have suggested that scaling CRF by whole-body mass confounds the associations between CRF and health outcomes by adiposity. Furthermore, we recently showed that despite the level of CRF, fat mass was directly associated with insulin resistance. Therefore, to understand the relationship between CRF and health outcomes in youth, it is important to use a scaling approach for CRF that is not confounded by adiposity. Although Telford et al. (1) suggested that cardiopulmonary exercise tests performed during seated cycling, running on a treadmill, or in the field require different scaling approaches, it has been suggested that the increase in fat-free mass (FFM) during growth and maturation has the greatest influence on the development of CRF regardless of the testing methods (5). This is a reasonable assumption because FFM is also the strongest determinant of the growth of the heart and the left ventricular mass (6), both of which are strong correlates of maximal cardiac output, the main determinant of the variability in peak oxygen uptake between individuals. FFM is also mainly responsible for increased oxygen requirements of the body during exercise; FFM augments venous return, and it improves left ventricular end-diastolic volume increasing left ventricular stroke volume during exercise. Therefore, FFM would provide physiologically the most appropriate factor for controlling the effect of body size and composition on CRF (7). Although the evidence suggests that FFM should be used in scaling of CRF (7), there are several different approaches for using FFM or other morphological factors for this purpose. Although log-linear allometric scaling used in our study is effective in removing the effect of body size and composition on CRF particularly in cross-sectional studies, it is not the only approach that can be used. For example, Shaibi et al. (4) used statistical adjustment for FFM and fat mass in their analyses to obtain information on the relationship between CRF and the metabolic syndrome independent of body composition. Furthermore, Armstrong and Welsman (8) recommended multilevel regression modeling for the analysis of longitudinal CRF data. Therefore, the methodology suggested by Telford et al. is highly appreciated and adds an important aspect to the discussion dealing with best practices in analyzing and interpreting data on CRF. In conclusion, scaling CRF by body mass without allometric or other modeling confounds the associations of CRF with health outcomes. Therefore, it is important that pediatric exercise scientists continue the discussion on the preferred method for expressing and interpreting CRF in pediatric populations. Eero A. Haapala Faculty of Sport and Health Sciences University of Jyväskylä Jyväskylä, FINLAND Alan R. Barker Children’s Health and Exercise Research Centre University of Exeter Exeter, UNITED KINGDOM Timo A.