Errors in Measurement Using a Portable Metabolic System (AE-100i): Compared with a Stationary Metabolic System (AE-310S)

Abstract

[Purpose] To compare measurement errors when using portable (AE-100i) and stationary (AE-310S) metabolic systems. [Participants and Methods] We conducted an exercise load test for 23 health young persons to measure their VO₂, VCO₂, and VE levels using AE-100i and AE-310S. [Results] The intraclass correlation coefficient (ICC) between AE-100i and AE-310S was 0.98 in all VO₂, VCO₂, and VE measurements. In Bland‐Altman analysis, fixed bias was only observed in VCO₂ measurement. The minimal important difference in each measurement was as follows: VO₂: ± 4.2 ml/kg/min, VCO₂: –4.9 to 3.3 ml/kg/min, and VE: ± 8.2 l/min. [Conclusion] The ICC between AE-100i and AE-310S was markedly high, and errors in measurement using the former were within a clinically acceptable range. On the other hand, the results suggest the importance of considering the minimal important difference when using this system in clinical settings.