Influence of the base-line determination on work efficiency during submaximal cycling

The purpose of this study was to compare work efficiency values (WE = work accomplished/energy expenditure above exercising with 0 load) among different unloaded base-line correction techniques for different power outputs.

Twelve healthy men performed 6 5-min steady-state exercises of 0 (unloaded), 40, 80, 120, 160 and 200 W at a pedalling rate of 90 rpm on a cycle ergometer. Three different unloaded base-line corrections were used for WE calculation: an actual measurement of VO(2) corresponding to the unloaded pedalling exercise, the y-intercept value of the linear regression between VO(2) and power output and the y-intercept value of the curvilinear VO(2)-power regression.

The present study demonstrated that WE was significantly higher when determined using the actual measurement of the unloaded VO(2) than y-intercept values of the linear (p<0.001) and curvilinear (p<0.05) VO(2)-power regressions. WE based on theoretical determinations (linear vs curvilinear regressions) were not significantly different. The power output significantly affected all WE index, with higher WE being measured when the power output was low and lower WE when power output was high.

The high WE values determined using the actual VO(2) measurement could be explained by (i) the additional energy expended to stabilise the body in addition to the energy expenditure of moving the lower limbs without power production and by (ii) the difficulty to experimentally reproduce the unloaded condition. The large range of WE values measured in the present study is due to differences in the procedures used to determine the unloaded VO(2) (and thus differences in unloaded VO(2) values) as well as differences in the cycling intensities.