Validation of a Non-Invasive Approach to Assess Muscular Oxygen Uptake Kinetics

Introduction
In the exercise steady-state, pulmonary oxygen uptake (V’O₂pulm) reflects the rate of oxygen consumed by the tissues (V’O₂musc). However, during non-steady-state the dynamics of V’O₂pulm are dissociated from V’O₂musc. It is frequently assumed that the V’O₂pulm kinetics are simply a time-shifted version of V’O₂musc, with an early distortion due to the dynamics of cardiac output (CO) and venous return (cardio-dynamic phase). Therefore, we analyzed a range of V’O₂musc, CO, and V’O₂pulm kinetics to determine the ability to estimate V’O₂musc kinetics from V’O₂pulm.
Methods
Twenty simulations for V’O₂musc- and CO-kinetics (time constants: 10-70s; 10-55s) were generated as mono-exponential responses to dynamic work rate changes between 30 and 80 W. Two models (Barstow et al. (1990; 5 data sets); Benson et al. (under review; 15 data sets)) were applied to generate computed V'O₂pulm kinetics from known V'O₂musc and CO dynamics, which account for the dynamics of venous transport and flow-weighted mixing from different vascular beds. Sample-to-sample noise was added to computed V’O₂pulm responses, which was sampled at a typical breathing frequency for the exercise transient. Blinded estimation of the time constant for V’O₂musc kinetics from V’O₂pulm was based on the approach of Hoffmann et al. (2013). SPSS 21 was applied for statistical analysis (Spearman’s rho, Wilcoxon signed-rank test).Results
We found significant correlations between the time constants of actual (known) and predicted V’O2musc (r=0.913; p<0.01; N=20), and between actual V’O₂musc and V’O2pulm (r=0.879; p<0.01; N=20). However, a significant difference was observed between the time constants of V’O2pulm (34.2 ± 14.3s) and actual V’O2musc (30.5 ± 15.2s; p<0.05; N=20), but not between actual and predicted V’O2musc (30.6 ± 14.6 s; p=0.709; N=20).DiscussionThe results show that significant differences can occur between muscular and pulmonary V’O₂ kinetics during dynamic work rate changes. This is likely due to the influence of CO and venous return dynamics causing a transient non-linear distortion in O₂ exchange between muscle and lung. Accounting for these distortions (Hoffmann et al., 2013) allows the time constants for V’O₂musc to be recovered from pulmonary measurements with high validity. The study was funded by the DLR (Deutsches Zentrum für Luft- und Raumfahrt), Germany (FKZ 50WB0726); and the Biotechnology and Biological Sciences Research Council, UK (BB/I00162/X).

References
Barstow TJ, Lamarra N, Whipp BJ (1990) Modulation of muscle and pulmonary O2 uptakes by circulatory dynamics during exercise. J Appl Physiol 68:979–989

Benson, AP, Grassi, B, Rossiter, HB. (under review) A validated model of oxygen uptake and circulatory dynamic interactions at exercise onset in humans.

Hoffmann, U, Drescher, U, Benson, AP, Rossiter, HB, Essfeld, D (2013) Skeletal muscle V’O2 kinetics from cardio-pulmonary measurements: assessing distortions through O2 transport by means of stochastic work-rate signals and circulatory modelling. Eur J Appl Physiol DOI10.1007/s00421-013-2598-7