Reliability of two metabolic systems during spot-specific upper-body ergometry

In this study, we assessed the reliability of the portable Cosmed K4b^2 and the laboratory-based Powerlab 8M gas analysis systems for measuring peak physiological variables during sport-specific, upper-body exercise, to determine the error associated with repeated physiological performance tests. Ten trained female outrigger canoeists compelted two 1000-m ergometer time trials using each gas analysis system. Peak physiological responses were analysed using a two-way repeated measures analysis of variance and post hoc Tukey HSD tests. Reliability and within participant variability of peak responses were assessed using intraclass correlation coefficients (ICC) and measurement bias/ratios with 95% limits of agreement (LOA), respectively. Performance and peak physiological responses were similar across trials for each individual system, although the Powerlab 8M reported 13.6% significantly higher VCO2 and 7.4% non-significantly higher VO2 than the Cosmed K4b^2. Reliability of the Cosmed K4b^2 was high for VCO2 (0.98) and moderate for VE (0.85), whereas that for the Powerlab 8M was high for VCO2 (0.97) and tidal volume (0.94) and moderate for VO2 (0.87). Overall, the two systems report similar VO2 variability (8-9%) with the Powerlab 8M reporting higher VO2 and VCO2 values than the Cosmed K4b2. Therefore, while each system is reliable, results from each of the systems should not be used interchangeably