Use of oxygen uptake recovery curve to predict peak oxygen uptake in upper body exercise

A group of 18 well-trained white-water kayakers performed maximal upper body exercise in the laboratory and during a field test. Laboratory direct peak oxygen uptake (VO2) values were compared, firstly by a VO2 backward extrapolation estimation and secondly by an estimation calculated from VO2 measured during the first 20 s of exercise recovery. Direct peak VO2 correlated with VO2 backward extrapolation (r = 0.89), but the results of this study showed that the backward extrapolation method tended to overestimate significantly peak VO2 by [0.57 (SD 0.31) l.min-1 in the laboratory, and 0.66 (SD 0.33) l.min-1 in the field, P < 0.001]. The VO2 measured during the first 20 s of recovery, whether the exercise was performed in the laboratory or in the field, correlated well with the laboratory direct peak VO2 (r = 0.92 and r = 0.91, respectively). The use of the regression equation obtained from field data (VO2F20s), that is peak VO2 = 0.23 + 1.08 VO2F20s, gave an estimated peak VO2, the mean difference of which compared with direct peak VO2 was 0.22 (SD 0.13) l.min-1. In conclusion, we propose the use of a regression equation to estimate peak VO2 from a single sample of the gas expired during the first 20 s of recovery after maximal exercise involving the upper part of the body.