Author(s): AITKENHEAD, R., WALDRON, M., CONWAY, G., HORNER, K., HEFFERNAN, S., Institution: SWANSEA UNIVERSITY , Country: UNITED KINGDOM, Abstract-ID: 1143

INTRODUCTION:
Breakpoints in muscle oxygen saturation using near-infrared spectroscopy (NIRS) have been associated with ventilatory thresholds during exercise and are assumed to represent the same physiological constructs. However, whether both methods are able to track training-induced changes in these thresholds equivalently across time has not been established. We compared two methods of identifying the transition from the moderate-to-heavy exercise domain, during an incremental treadmill protocol: 1) NIRS-derived tissue saturation index (TSI%) and 2) pulmonary gas exchange (ventilatory threshold 1; VT1). Thresholds were expressed as a percentage of peak oxygen consumption (%V ̇O2peak) and agreement between methods was assessed before and after a period of training to evaluate their equivalence when monitoring changes in thresholds across time. An analytical goal of 3% (%V ̇O2peak) was set based upon the error of pulmonary gas measurement [1].
METHODS:
Eighteen males (27 ± 7 yrs) performed two (pre- and post-training) incremental running tests, interspersed by 12-weeks of training, wearing a portable NIRS system and pulmonary gas exchange measured. Across both tests, the NIRS-derived breakpoints in TSI% were identified with segmental regression and compared with pulmonary gas measures (VT1). Agreement was assessed using paired t-tests (mean bias) and 95% Limits of Agreement (95% LoA).
RESULTS:
For pre- and post-testing, there were no significant differences between methods (p = 0.17 and p = 0.81, respectively), with thresholds identified at 67.9 ± 3.7% and 66.9 ± 3.6% of %V ̇O2peak for all tests. There was no mean bias for %V ̇O2peak during pre-testing (95% LoA: -0.44 ± 5.59%) and post-testing (95% LoA: 0.06 ± 4.87%) but the LoA was beyond the analytical goal. Assessment of the change values between pre- and post-training was also not significantly different between the pulmonary (mean ± SD: 0.72 ± 2.27 %, range: 5.0 to 5.0) and NIRS (mean ± SD: 1.22 ± 2.10 %, range: -3.0 to 6.0) methods (p = 0.11), indicating uniform changes in thresholds across both methods.
CONCLUSION:
The NIRS-derived method did not uniformly disagree with pulmonary measures, across two time points. Therefore, irrespective of individual training effects, deflections in TSI% identified via segmental regression will not over- or underestimate VT1 but will randomly vary to an unacceptable level during incremental exercise tests in trained runners. These findings offer more insight into how NIRS can be applied as a method for continuous monitoring of ventilatory breakpoints throughout the training process but question the ability to identify equivalent thresholds in all cases.

Reference
[1] Gaskill, S. E., Ruby, B. C., Walker, A. J., Sanchez, O. A., Serfass, R. C., & Leon, A. S. (2001). Validity and reliability of combining three methods to determine ventilatory threshold. Medicine & Science in Sports & Exercise, 33(11), 1841-1848. 10.1097/00005768-200111000-00007