Breath‐by‐breath pulmonary O2 uptake kinetics: effect of data processing on confidence in estimating model parameters

Abstract

What is the central question of this study? In groups of young and older adults, we investigated whether techniques used as common practice for processing breath‐by‐breath pulmonary O₂ uptake data from repeated step transitions in work rate into the moderate‐intensity exercise domain influence the model parameter estimations and confidence of describing the phase II pulmonary O₂ uptake response. What is the main finding and its importance? Results demonstrate that regardless of age group, during transitions into the moderate‐intensity exercise domain, techniques for processing individual transitions did not affect parameter estimates describing the phase II pulmonary O₂ uptake response; however, the confidence in the parameter estimation could be improved by the technique used to process individual trials. To improve the signal‐to‐noise ratio of breath‐by‐breath pulmonary O₂ uptake () data, it is common practice to perform multiple step transitions, which are subsequently processed to yield an ensemble‐averaged profile. The effect of different data‐processing techniques on phase II kinetic parameter estimates ( amplitude, time delay and phase II time constant (τ)] and model confidence [95% confidence interval (CI₉₅)] was examined. Young (n = 9) and older men (n = 9) performed four step transitions from a 20 W baseline to a work rate corresponding to 90% of their estimated lactate threshold on a cycle ergometer. Breath‐by‐breath was measured using mass spectrometry and volume turbine. Mono‐exponential kinetic modelling of phase II data was performed on data processed using the following techniques: (A) raw data (trials time aligned, breaths of all trials combined and sorted in time); (B) raw data plus interpolation (trials time aligned, combined, sorted and linearly interpolated to second by second); (C) raw data plus interpolation plus 5 s bin averaged; (D) individual trial interpolation plus ensemble averaged [trials time aligned, linearly interpolated to second by second (technique 1; points joined by straight‐line segments), ensemble averaged]; (E) ‘D’ plus 5 s bin averaged; (F) individual trial interpolation plus ensemble averaged [trials time aligned, linearly interpolated to second by second (technique 2; points copied until subsequent point appears), ensemble averaged]; and (G) ‘F’ plus 5 s bin averaged. All of the model parameters were unaffected by data‐processing technique; however, the CI₉₅ for τ in condition ‘D’ (4 s) was lower (P < 0.05) than the CI₉₅ reported for all other conditions (5–10 s). Data‐processing technique had no effect on parameter estimates of the phase II response. However, the narrowest interval for CI₉₅ occurred when individual trials were linearly interpolated and ensemble averaged.