Scheadler CM, Garver MJ, Hanson NJ.

Med Sci Sports Exerc. 2017 Sep;49(9):1911-1916

INTRODUCTION: There is a plethora of gas sampling intervals available during
cardiopulmonary exercise testing to measure peak oxygen consumption (V˙O2peak).
Different intervals can lead to altered V˙O2peak. Whether differences are
affected by the exercise protocol or subject sample is not clear. The purpose of
this investigation was to determine whether V˙O2peak differed because of the
manipulation of sampling intervals and whether differences were independent of
the protocol and subject sample.
METHODS: The first subject sample (24 ± 3 yr; V˙O2peak via 15-breath moving
averages: 56.2 ± 6.8 mL·kg·min) completed the Bruce and the self-paced V˙O2max
protocols. The second subject sample (21.9 ± 2.7 yr; V˙O2peak via 15-breath
moving averages: 54.2 ± 8.0 mL·kg·min) completed the Bruce and the modified
Astrand protocols. V˙O2peak was identified using five sampling intervals: 15-s
block averages, 30-s block averages, 15-breath block averages, 15-breath moving
averages, and 30-s block averages aligned to the end of exercise. Differences in
V˙O2peak between intervals were determined using repeated-measures ANOVAs. The
influence of subject sample on the sampling effect was determined using
independent t-tests.
RESULTS: There was a significant main effect of sampling interval on V˙O2peak
(first sample Bruce and self-paced V˙O2max P < 0.001; second sample Bruce and
modified Astrand P < 0.05). The difference in V˙O2peak between sampling intervals
followed a similar pattern for each protocol and subject sample, with 15-breath
moving average presenting the highest V˙O2peak.
CONCLUSIONS: The effect of manipulating gas sampling intervals on V˙O2peak
appears to be protocol and sample independent. These findings highlight our
recommendation that the clinical and scientific community request and report the
sampling interval whenever metabolic data are presented. The standardization of
reporting would assist in the comparison of V˙O2peak.