Reproducibility of peak oxygen consumption and the impact of test variability on classification of individual training responses in young recreationally active adults

Abstract

This study investigated whether VO₂ peak is reproducible across repeated tests before (PRE) and after (POST) training, and whether variability across tests impacts how individual responses are classified following 3 weeks of aerobic exercise training (cycle ergometry). Data from 45 young healthy adults (age: 20·1 ± 0·9 years; VO₂ peak, 42·0 ± 6·7 ml·min^-1 ) from two previously published studies were utilized in the current analysis. Non-responders were classified as individuals who failed to demonstrate an increase or decrease in VO₂ peak that was greater than 2·0 times the typical error of measurement (107 ml·min^-1 ) away from zero, while responders and adverse responders were above and below this cut-off, respectively. VO₂ peak tests at PRE (three total) and POST (three total) were highly reproducible (PRE and POST average and single measures ICCs: range 0·938-0·992), with low coefficients of variation (PRE:4·9 ± 3·1%, POST: 4·8 ± 2·7%). However, a potential learning effect was observed in the VO₂ peak tests prior to training, as the initial pretraining test was significantly lower than the third (p = 0·010, PRE 1: 2 946 ± 924 ml·min^-1 , PRE 3: 3 042 ± 919 ml·min^-1 ). This resulted in fewer individuals classified as adverse responders for Test 3 compared to any combination of tests that included Test 1, suggesting that a single ramp test at baseline may not be sufficient to accurately classify the VO₂ peak response in young recreationally active individuals. Thus, it is our recommendation that the initial VO₂ peak test be used as a familiarization visit and not included for analysis.

Keywords: cycle ergometer; individual response; intraclass correlation coefficient; learning effect; ramp test; stewart-maxwell test.