Training status differences in torque and neuromuscular system variability and complexity during maximal and submaximal fatiguing exercise

Abstract

Nonlinear methods are useful in studying systems where outputs are not proportional or predictable given the inputs and are becoming increasing popular to analyze EMG signal. Complexity and entropy are two of the most used nonlinear methods to examine physiological systems. Physiological complexity has been shown to allow a wider range of adaptable states for the system to deal with stressors. Previous research has shown that complexity is reduced in response to fatiguing exercise. To date no studies have examined the effect of training status on neuromuscular system complexity. Purpose: Therefore, the purposes of this study are to investigate whether training status (aerobic or resistance) affects neuromuscular system complexity at baseline compared to untrained individuals and to investigate whether training status changes the loss of complexity due to fatigue compared to untrained individuals. Methods: The study was split into two experimental visits: a maximal fatiguing test and a submaximal fatiguing test. Thirty-three individuals participated in the maximal test, while 30 individuals participated in the submaximal test. The participants were split into three groups: untrained (UT), aerobic trained (AT), and resistance trained (RT). During the maximal test, the participants performed 30 maximal intermittent isometric knee extensions (6-second contraction/4 second rest) over a 5-minute test. For the submaximal test, the participants maximal voluntary contraction was obtained. The participants then performed a time to task failure protocol by completing as many intermittent isometric knee extensions (6-second contraction/4 second rest) until failure. Surface EMG (sEMG) of the rectus femoris (RF), vastus lateralis (VL), and vastus medialis (VM) was recorded continuously during the protocols. The torque signals and sEMG recordings from the contractions were analyzed for complexity using Detrended Fluctuation Analysis (DFA), Sample Entropy (SampEn), Recurrence Rate (REC), and Determinism (DET). The first three contractions (Fresh) and last three contractions (End-Task) of each test were averaged together. Results: During the maximal test, complexity of the torque and EMG signals was reduced at End-Task compared to Fresh (p<0.05). The AT group had lower torque complexity (SampEn) than the UT and RT groups during at Fresh (p<0.05). In addition, the AT group had higher sEMG complexity (DET and SampEn) in the RF and VL at Fresh (p<0.05). During the submaximal test, complexity of the torque signals and sEMG was reduced at task end (p<0.05). The AT group had higher torque complexity (SampEn) and higher sEMG complexity (DET and SampEn) than the UT and RT groups at Fresh (p<0.05). Conclusion: Decreases in complexity and entropy have been shown when there is high motor unit synchronization and conduction velocity. The lower complexity and entropy due to fatigue may be due to changes in central and peripheral mechanisms to continue to achieve task output. Aerobic training may provide increased neuromuscular system complexity.