Key pointsr Healthy physiological systems generate time series possessing complex structures, as seen for example in heart rate variability, respiratory rate and gait.r A loss of complexity in physiological time series has been associated with system dysfunction, and this loss is a characteristic feature of torque output from the ageing neuromuscular system. r We sought to determine the effect of neuromuscular fatigue on the complexity of knee extensor torque output in healthy young humans performing repeated maximal and submaximal contractions.r Fatigue resulted in a substantial loss of knee extensor torque complexity, with the noise in the torque signal becoming increasingly Brownian in character.r Complexity has been associated with system adaptability, and the fatigue-induced loss of complexity, the physiological origin of which is obscure, may contribute to the inability to sustain physical exercise.Abstract Neuromuscular fatigue increases the amplitude of fluctuations in torque output during isometric contractions, but the effect of fatigue on the temporal structure, or complexity, of these fluctuations is not known. We hypothesised that fatigue would result in a loss of temporal complexity and a change in fractal scaling of the torque signal during isometric knee extensor exercise. Eleven healthy participants performed a maximal test (5 min of intermittent maximal voluntary contractions, MVCs), and a submaximal test (contractions at a target of 40% MVC performed until task failure), each with a 60% duty factor (6 s contraction, 4 s rest). Torque and surface EMG signals were sampled continuously. Complexity and fractal scaling of torque were quantified by calculating approximate entropy (ApEn), sample entropy (SampEn) and the detrended fluctuation analysis (DFA) scaling exponent α. Fresh submaximal contractions were more complex than maximal contractions (mean ± SEM, submaximal vs. maximal: ApEn 0.65 ± 0.09 vs. 0.15 ± 0.02; SampEn 0.62 ± 0.09 vs. 0.14 ± 0.02; DFA α 1.35 ± 0.04 vs. 1.55 ± 0.03; all P < 0.005). Fatigue reduced the complexity of submaximal contractions (ApEn to 0.24 ± 0.05; SampEn to 0.22 ± 0.04; DFA α to 1.55 ± 0.03; all P < 0.005) and maximal contractions (ApEn to 0.10 ± 0.02; SampEn to 0.10 ± 0.02; DFA α to 1.63 ± 0.02; all P < 0.01). This loss of complexity and shift towards Brownian-like noise suggests that as well as reducing the capacity to produce torque, fatigue reduces the neuromuscular system's adaptability to external perturbations.

The complexity of knee extensor torque time series decreases during fatiguing isometric muscle contractions. We hypothesized that because of peripheral fatigue, this loss of torque complexity would occur exclusively during contractions above the critical torque (CT). Nine healthy participants performed isometric knee extension exercise (6 s of contraction, 4 s of rest) on six occasions for 30 min or to task failure, whichever occurred sooner. Four trials were performed above CT (trials S1-S4, S1 being the lowest intensity), and two were performed below CT (at 50% and 90% of CT). Global, central, and peripheral fatigue were quantified using maximal voluntary contractions (MVCs) with femoral nerve stimulation. The complexity of torque output was determined using approximate entropy (ApEn) and the detrended fluctuation analysis-α scaling exponent (DFA-α). The MVC torque was reduced in trials below CT [by 19 ± 4% (means ± SE) in 90%CT], but complexity did not decrease [ApEn for 90%CT: from 0.82 ± 0.03 to 0.75 ± 0.06, 95% paired-samples confidence intervals (CIs), 95% CI = -0.23, 0.10; DFA-α from 1.36 ± 0.01 to 1.32 ± 0.03, 95% CI -0.12, 0.04]. Above CT, substantial reductions in MVC torque occurred (of 49 ± 8% in S1), and torque complexity was reduced (ApEn for S1: from 0.67 ± 0.06 to 0.14 ± 0.01, 95% CI = -0.72, -0.33; DFA-α from 1.38 ± 0.03 to 1.58 ± 0.01, 95% CI 0.12, 0.29). Thus, in these experiments, the fatigue-induced loss of torque complexity occurred exclusively during contractions performed above the CT.

Caffeine ingestion slowed the fatigue-induced loss of torque complexity and increased the time to task failure during intermittent isometric contractions, most likely through central mechanisms.

No abstract

Neuromuscular fatigue reduces the temporal structure, or complexity, of torque output during muscular contractions. To determine whether the fatigue-induced loss of torque complexity could be accounted for by central or peripheral factors, nine healthy participants performed four experimental trials involving intermittent isometric contractions of the knee extensors at 50% of the maximal voluntary contraction torque. These trials involved: (i) two bouts of contractions to failure using the right leg separated by 3 min recovery (IPS); (ii) the same protocol but with cuff occlusion during the 3 min recovery (IPS-OCC); (iii) contractions of the left leg to failure, followed 1 min later by contractions of the right leg to failure (CONT); and (iv) the same protocol but with cuff occlusion applied to the left leg throughout both the recovery period and right leg contractions (CONT-OCC). Supramaximal electrical stimulation during maximal voluntary contractions was used to determine the degree of central and peripheral fatigue, whilst complexity was determined using approximate entropy (ApEn) and detrended fluctuation analysis α exponent (DFA α). Neuromuscular fatigue was consistently associated with a loss of torque complexity in all conditions [e.g. IPS bout 1, ApEn from (mean ± SD) 0.46 ± 0.14 to 0.12 ± 0.06 (P < 0.001)]. In IPS-OCC, occlusion abolished the recovery from fatigue, and torque complexity remained at the values observed at task failure in the preceding bout (IPS-OCC bout 2, first minute 0.14 ± 0.03, P < 0.001). Prior contralateral contractions, with or without blood flow occlusion, had no effect on torque complexity.

To test the hypothesis that a system’s metabolic rate and the complexity of fluctuations in the output of that system are related, thirteen healthy participants performed intermittent isometric knee extensor contractions at intensities where a rise in metabolic rate would (40% maximal voluntary contraction, MVC) and would not (20% MVC) be expected. The contractions had a 60% duty factor (6 sec contraction, 4 sec rest) and were performed until task failure or for 30 min, whichever occurred sooner. Torque and surface EMG signals were sampled continuously. Complexity and fractal scaling of torque were quantified using approximate entropy (ApEn) and the detrended fluctuation analysis (DFA) α scaling exponent. Muscle metabolic rate was determined using near‐infrared spectroscopy. At 40% MVC, task failure occurred after (mean ± SD) 11.5 ± 5.2 min, whereas all participants completed 30 min of contractions at 20% MVC. Muscle metabolic rate increased significantly after 2 min at 40% MVC (2.70 ± 1.48 to 4.04 ± 1.23 %·s‐1, P < 0.001), but not at 20% MVC. Similarly, complexity decreased significantly at 40% MVC (ApEn, 0.53 ± 0.19 to 0.15 ± 0.09; DFA α, 1.37 ± 0.08 to 1.60 ± 0.09; both P < 0.001), but not at 20% MVC. The rates of change of torque complexity and muscle metabolic rate at 40% MVC were significantly correlated (ApEn, ρ = −0.63, P = 0.022; DFA, ρ = 0.58, P = 0.037). This study demonstrated that an inverse relationship exists between muscle torque complexity and metabolic rate during high‐intensity contractions.

Introduction Distinct physiological responses to exercise occur in the heavy- and severe-intensity domains, which are separated by the critical power or critical torque (CT). However, how the transition between these intensity domains actually occurs is not known. We tested the hypothesis that CT is a sudden threshold, with no gradual transition from heavy- to severe-intensity behavior within the confidence limits associated with the CT. Methods Twelve healthy participants performed four exhaustive severe-intensity trials for the determination of CT, and four 30-min trials in close proximity to CT (one or two SE above or below each participant’s CT estimate; CT − 2, CT − 1, CT + 1, CT + 2). Muscle O 2 uptake, rectified electromyogram, and torque variability and complexity were monitored throughout each trial, and maximal voluntary contractions (MVC) with femoral nerve stimulation were performed before and after each trial to determine central and peripheral fatigue responses. Results The rates of change in fatigue-related variables, muscle O 2 uptake, electromyogram amplitude, and torque complexity were significantly faster in the severe trials compared with CT − 2. For example, the fall in MVC torque was −1.5 ± 0.8 N·m·min −1 in CT − 2 versus –7.9 ± 2.5 N·m·min −1 in the lowest severe-intensity trial ( P < 0.05). Individual analyses showed a low frequency of severe responses even in the circa-CT trials ostensibly above the CT, but also the rare appearance of severe-intensity responses in all circa-CT trials. Conclusions These data demonstrate that the transition between heavy- and severe-intensity exercise occurs gradually rather than suddenly.

New Findings What is the topic of this review?Physiological complexity in muscle force and torque fluctuations, specifically the quantification of complexity, how neuromuscular complexityis altered by perturbations and the potential mechanism underlying changes in neuromuscular complexity. What advances does it highlight?The necessity to calculate both magnitude‐ and complexity‐based measures for the thorough evaluation of force/torque fluctuations. Also the need for further research on neuromuscular complexity, particularly how it relates to the performance of functional activities (e.g. manual dexterity, balance, locomotion). Abstract Physiological time series produce inherently complex fluctuations. In the last 30 years, methods have been developed to characterise these fluctuations, and have revealed that they contain information about the function of the system producing them. Two broad classes of metrics are used: (1) those which quantify the regularity of the signal (e.g. entropy metrics); and (2) those which quantify the fractal properties of the signal (e.g. detrended fluctuation analysis). Using these techniques, it has been demonstrated that ageing results in a loss of complexity in the time series of a multitude of signals, including heart rate, respiration, gait and, crucially, muscle force or torque output. This suggests that as the body ages, physiological systems become less adaptable (i.e. the systems’ ability to respond rapidly to a changing external environment is diminished). More recently, it has been shown that neuromuscular fatigue causes a substantial loss of muscle torque complexity, a process that can be observed in a few minutes, rather than the decades it requires for the same system to degrade with ageing. The loss of torque complexity with neuromuscular fatigue appears to occur exclusively above the critical torque (at least for tasks lasting up to 30 min). The loss of torque complexity can be exacerbated with previous exercise of the same limb, and reduced by the administration of caffeine, suggesting both peripheral and central mechanisms contribute to this loss. The mechanisms underpinning the loss of complexity are not known but may be related to altered motor unit behaviour as the muscle fatigues.