Purpose
We investigated the accuracy of the Moxus Modular Metabolic System (MOXUS) against the Douglas Bag Method (DBM) during high-intensity exercise, and whether the two methods agreed when detecting small changes in $$\dot{V}{\text{O}}_{2}$$ V ˙ O 2 between two consecutive workloads ( $$\Delta {\dot{{V}}\text{O}}_{ 2}$$ Δ V ˙ O 2 ).
Methods
Twelve trained male runners performed two maximal incremental running tests while gas exchange was analyzed simultaneously by the two systems using a serial setup for four consecutive intervals of 30 s on each test. Comparisons between methods were performed for $$\dot{V}{\text{O}}_{2}$$ V ˙ O 2 , $${\dot{{V}}}_{\text{E}}$$ V ˙ E , fractions of expired O2 (FeO2) and CO2 (FeCO2) and $$\Delta {\dot{{V}}\text{O}}_{ 2}$$ Δ V ˙ O 2 .
Results
The MOXUS produced significant higher (mean ± SD, n = 54) readings for $$\dot{V}{\text{O}}_{2}$$ V ˙ O 2 (80 ± 200 mL min−1, p = 0.005) and $${\dot{{V}}}_{\text{E}}$$ V ˙ E (2.9 ± 4.2 L min−1, p < 0.0001), but not FeO2 (−0.01 ± 0.09). Log-transformed 95 % limits of agreement for readings between methods were 94–110 % for $$\dot{V}{\text{O}}_{2}$$ V ˙ O 2 , 97–108 % for $$\dot{V}_{\text{E}}$$ V ˙ E and 99–101 % for FeO2. $$\Delta \dot{V}{\text{O}}_{2}$$ Δ V ˙ O 2 for two consecutive measurements was not different between systems (120 ± 110 vs. 90 ± 190 mL min−1 for MOXUS and DBM, respectively, p = 0.26), but agreement between methods was very low (r = 0.25, p = 0.12).
Discussion
Although it was tested during high-intensity exercise and short sampling intervals, the MOXUS performed within the acceptable range of accuracy reported for automated analyzers. Most of the differences between equipments were due to differences in $$\dot{V}_{\text{E}}$$ V ˙ E . Detecting small changes in $$\dot{V}{\text{O}}_{2}$$ V ˙ O 2 during an incremental test with small changes in workload, however, might be beyond the equipment’s accuracy.