The reliability of cycling efficiency.

PURPOSE: The aim of this experiment was to establish the reproducibility of gross efficiency (GE), delta efficiency (DE), and economy (EC) during a graded cycle ergometer test in seventeen male subjects.
METHODS: All subjects performed three identical exercise tests at a constant pedal cadence of 80 rpm on an electrically braked cycle ergometer. Energy expenditure was estimated from measures of oxygen uptake (VO(2)) and carbon dioxide production (VCO(2)) by using stoichiometric equations.
RESULTS: The subjects characteristics were age 24 +/- 6 yr, body mass 74.6 +/- 6.9 kg, body fat 13.9 +/- 2.2%, and VO(2max) 61.9 +/- 2.4 mL x kg(-1) x min(-1) (all means +/- SD). Average GE, DE, and EC for the three tests were 19.8 +/- 0.6%, 25.8 +/- 1.5%, and 5.0 +/- 0.1 kJ x L(-1), respectively. The coefficients of variation (confidence limits) were GE 4.2 (3.2-6.4)%, DE 6.7 (5.0-10.0)%, and EC 3.3 (2.4-4.9)%. GE was significantly lower at 95 W and 130 W when compared with 165 W, 200 W, 235 W, 270 W, and 305 W. GE at 165 W was significantly lower (P < 0.05) that GE at 235 W. A weak correlation (r = 0.491; P < 0.05) was found between peak oxygen uptake (VO(2peak)) and GE, whereas no correlations were found between VO(2max) and DE or EC.
CONCLUSION: We conclude that a graded exercise test with 3-min stages and 35-W increments is a method by which reproducible measurements of both GE and EC can be obtained, whereas measurements of DE seemed slightly more variable.