Sir,

We thank Professor Mocellin for his letter concerning our article entitled “A practical and transferable new protocol for treadmill testing of children and adults”.Reference Dubowy, Baden, Bernitzki and Peters¹ In this letter, he questions whether the population we chose when making our comparison meets the demands to be expected for normal controls. He wonders whether the part of the study relating to the children is representative.

We had recognised the problems existing in choosing an adequate control population. Our 849 healthy volunteers younger then 18 years of age had already been examined. The distribution of their ages is now shown in Tables 1 and 2 below. We exclude anyone expressing apprehension of the test, which proved the case for some of those attending kindergarten. Despite these exclusions, 68 children younger than 7 years of age, with the youngest being 35 months of age, successfully performed lung function testing, exercise tests and at least 3 gas checks, including measurements of lactate. It is hardly surprising that some young children attending kindergarten refused to participate, were not able to attend, or else found difficulty in understand the instructions offered by the nurses and physicians. We also excluded trained athletes. Those participating in sports at school or in their leisure time, including adults, nonetheless, were included in the study. We recognize, therefore, that there were limitations with our chosen sample. It remains the case, nonetheless, that any study providing standard values requires volunteers. In our opinion, the pre-existing standard values were unsuitable for making comparisons with either children or the elderly. Those existing standards are impractical, unreproducible, and based on very small samples. Before starting our study, therefore, we made a maximum likelihood estimate, calculating percentiles by means of the LMS-MethodReference Cole and Green². Bodily features, such as weight and height, had already been compared to available nomograms.Reference Hesse, Jaeger and Vogel^3–Reference Komlos and Kriwy⁵

Table 1

Uptake of oxygen at ventilatory threshold related to the square of the measured height.

Table 2

Peak uptake of oxygen related to the square of the measured height.

Professor Mocellin suggests that we should have used values for peak consumption of oxygen related to standard values for height. This is a good suggestion, and such values have already been calculated by our working group for peak uptake, and uptake of oxygen at the ventilatory threshold (see supporting files on the website – Figs 1–4). We also agree that the square of the measurement of height is an appropriate reference parameter for standard values of peak uptake of oxygen. This is true for children, but would be wrong when considering the elderly (see Tables 1 and 2). Our data are in keeping with that suggested by Mocellin when considering the paediatric age groupReference Mocellin⁶. But we investigated exercise performance in a healthy Caucasian population ranging in age from 4 to 75 years. In common practice, consumption of oxygen is related to body mass. Our intention was to create a practical exercise protocol suitable for each age group, thus providing physicians with a helpful tool in the form of percentiles with which they could compare performance during the entire growth and aging of the individual.

Last, but not least, exercise testing, especially ergospirometry, is very difficult to practice and interpret, especially in children or the elderly. In this respect, it remains my own opinion that we have been successful in providing a practical solution to this difficult topic. For those who require them, we also have available many more centile curves relating to ventilatory parameters and gas exchange. I will be more than happy to supply them to interested parties.