Peripheral muscle fatigue revealed by near-infrared spectroscopy and surface electromyography

Abstract

Muscle fatigue is generally defined as an exercise-induced decline in the muscular capacity to maintain the maximal force or power generation. Muscle fatigue is commonly divided into central fatigue and peripheral fatigue. This study mainly concentrates on the peripheral muscle fatigue. Studies have confirmed that the integral of near-infrared spectroscopy (NIRS) and surface electromyography (EMG) methods can provide a comprehensive understanding, as well as evaluation, of muscle fatigue development. In this study, three experiments were conducted. The first experiment aims to explore the muscle fatigue development and the recovery process revealed by tissue oxygenation and hemodynamics. Near-infrared spectroscopy (NIRS) was used to measure the hemodynamic responses of gastrocnemius lateralis muscle (GL) during a pre-exercise resting phase, a heel-lift exercise, and a recovery phase. Wavelet transform was applied to decompose the resting-phase total hemoglobin signals (ΔtHb) and to reveal the relative contributions of the six characteristic frequency intervals to the total energy of the blood volume fluctuations. Inverse wavelet transform was utilized to extract the exercise-induced oscillatory components in both tissue oxygenation index (TOI) and ΔtHb signals. In exercise, the contraction-induced fluctuations in the ΔtHb signal presented a descending trend while the oscillations in the TOI showed an ascending trend. The TOI maintained a significantly higher level after the exercise. The normalized wavelet energy of the ΔtHb signal showed significant increases in frequency intervals I to IV in the recovery phase compared to the rest phase, while the interval VI displayed significant decrease. These results demonstrated that the NIRS method can provide useful information regarding the muscle fatigue development and recovery. The second study is designed to investigate the relationship between exercise quantity (or duration) and muscle hemodynamic responses during recovery. Fifteen healthy subjects were recruited from the University. The subject was instructed to perform two plantarflexion exercise sessions with 70% maximum voluntary contraction (MVC) force with each session consisting of 10-cycle isometric contractions. The ΔtHb signal of the gastrocnemius lateralis muscle was detected in three phases (before and after each exercise session) by the NIRS method. Similar to the first experiment, wavelet transform was applied to obtain the total wavelet energy (tWE) in the frequency range of 0.005-2 Hz. The wavelet amplitude (WA) and wavelet energy (WE) in the six characteristic frequency bands were also calculated. Results displayed that the tWE showed an increasing trend after each exercise session with a significant increase from rest phase 1 to rest phase 3. The WE values presented statistically significant increases in intervals I, III, IV and V from rest phase 1 to rest phase 3 and in frequency intervals III and IV from rest phase 2 to rest phase 3. The WE value had similar trend with the WA parameter. These findings indicate that regional microvascular regulators contribute markedly to the blood volume fluctuations. The increase in the hemodynamic response level is affected by the exercise quantity (or duration). The results also manifest that although the influence of skin blood flow on the tissue NIRS signal cannot be removed, the above-mentioned parameters revealed by wavelet transform mainly reflect the hemodynamic responses of the muscular microcirculation. The purpose of the third experiment is to explore the development of peripheral muscle fatigue using both NIRS and EMG measurements, as well as exploring the relationship between the fatigue-evaluation parameters derived from NIRS and EMG signals. Fifteen young and healthy subjects were recruited. During the experiment, the subject was asked to perform three trials of plantarflexion exercise with 70% MVC. Same to Experiment 2, each trial is composed of 10-cycle isometric contraction, with the durations of contraction and relaxation both being 10 s. NIRS and EMG signals of gastrocnemius medialis (GM) and lateralis (GL) were simultaneously recorded during all three trials. The MVC force after each exercise session was also measured to reflect the fatigue level. The oxygen consumption rate (RO2) and peak oxygen extraction (PKV) were obtained from the NIRS-derived deoxygenated signals. The root mean square (RMS), integrated EMG (iEMG) and median frequency (MDF) of EMG signals were also calculated for analysis. Results showed that both RO2 and PKV values presented an increasing trend as contraction continued in all three exercise trials of GM and GL. The MDF value showed a decreasing trend during each exercise in both GM and GL muscles, while RMS and iEMG only had a raising trend in GL muscle. Correlation analysis indicated that on one hand, the amplitude-related parameters of EMG signal (RMS and iEMG) displayed significantly high positive correlations with the NIRS parameters (RO2 and PKV). On the other hand, the frequency-relevant variable of EMG (MDF) presented significantly high negative correlations with the NIRS-derived PO2 and PKV. These results demonstrate that NIRS method can directly detect the transition from the fast-twitch fibers to the slow-twitch fibers during the peripheral fatigue process. The combination of NIRS and EMG measurements can be used to comprehensively evaluate the development of peripheral muscle fatigue.