Fatigue is an exercise-induced impairment in the ability to produce muscular force (19) in the presence of an increased perception of effort (16). The neuromuscular mechanisms contributing to fatigue can be broadly attributed to processes along the motor pathway that are deemed central or peripheral in origin. For high-intensity locomotor cycling exercise, a consistent magnitude of peripheral fatigue (~35% reduction in potentiated quadriceps twitch force) has been documented at exercise termination under various experimental conditions (2–6). This observation has been proposed to represent a centrally mediated “individual critical threshold” regulated by inhibitory afferent feedback to protect against excessive disruption to homeostasis (1). Although the consistency of end-exercise peripheral fatigue in these studies is striking, whether this plays a determining role in exercise tolerance is controversial, and recent evidence suggests any individual critical threshold is likely to be task dependent. For example, subsequent work from the same group has demonstrated greater end-exercise peripheral fatigue after dynamic single compared with double-limb exercise (32,33), and Johnson et al. (25) recently reported end-exercise peripheral fatigue after high-intensity cycling is attenuated, central fatigue is exacerbated, and exercise tolerance is reduced if exercise is preceded by high-intensity arm cranking.
For self-paced cycling time-trial exercise, the magnitude of end-exercise peripheral fatigue also varies with the intensity and duration of the task. We recently demonstrated greater peripheral fatigue after short-duration, high-intensity 4-km self-paced time trials in comparison with longer, lower-intensity 20- and 40-km time trials, where central fatigue was exacerbated (38). It was hypothesized that the task-dependent nature of the fatigue observed after self-paced cycling time trials could be explained by differences in the exercise intensity domain in which the trials were predominantly completed. The exercise intensity domain concept describes the distinct physiological responses that occur during exercise above and below an individual’s maximum sustainable intensity or critical power (10,26,31). The shorter, high-intensity 4-km trial, where peripheral fatigue was higher, was characterized by nonlinear physiological responses consistent with exercise in the severe-intensity domain (38). By contrast, the longer time trials, where peripheral fatigue was lower but central fatigue was exacerbated, were characterized by steady-state physiological responses for the majority of the trial, consistent with sustainable exercise below critical power in the heavy exercise intensity domain (38). The distinct physiological responses associated with non–steady-state, severe-intensity exercise compared with steady-state, heavy-intensity exercise might therefore be associated with a distinct profile of neuromuscular fatigue. However, the varying nature of self-selected power output in these trials meant participants were traversing between exercise intensity domains, particularly at the start and end of the trial, limiting the validity of this conclusion.
These limitations notwithstanding, a similar intensity- and duration-dependent pattern of central and peripheral processes to fatigue has been observed for exhaustive, intermittent, single-limb exercise above and below the critical torque (10). Interestingly, this study reported that at varying intensities above the critical torque, central fatigue was exacerbated in a duration-dependent manner, but the decline in potentiated twitch force was consistent between trials (10). This similar decline in potentiated twitch force suggests that peripheral fatigue is a primary determinant of exercise tolerance during single-limb contractions at intensities above the critical torque. Whether this posit can be extended to locomotor exercise above critical power in the severe-intensity domain is not known. Recently, de Souza et al. (13) demonstrated no difference in the magnitude of end-exercise fatigue after exercise in the severe-intensity domain of 3 min versus 10 min duration, but the contribution of central and peripheral processes was not assessed. Given 1) the accelerated development of fatigue and associated metabolic perturbations at severe exercise intensities (10,26,31), 2) the consistency of end-exercise peripheral fatigue after high-intensity cycling exercise (2–6), and 3) the consistency of exercise-induced fatigue after severe-intensity cycling of varying durations (13), it makes the expectation tenable that exercise termination during short-duration, high-intensity exercise above critical power in the severe-intensity domain would be concurrent with a consistent magnitude of peripheral fatigue. Conversely, longer-duration, steady-state exercise in the heavy domain likely terminates with a smaller magnitude of peripheral fatigue, but a greater contribution from central fatigue mechanisms (10,38). Accordingly, the aims of this study were to test whether central and peripheral fatigue would differ after short-duration, severe-intensity constant-load locomotor exercise compared with longer-duration, heavy-intensity exercise, and to assess whether exercise at intensities in the severe domain would terminate with a similar degree of peripheral fatigue. We hypothesized that 1) central fatigue would be exacerbated in a duration-dependent manner; 2) that peripheral fatigue would be lower after long-duration constant-load cycling exercise compared with short-duration, severe-intensity cycling exercise; and 3) that exercise in the severe-intensity domain would terminate with a similar degree of peripheral fatigue.
METHODS
Participants
Twelve well-trained male cyclists (mean ± SD age, 28 ± 8 yr; stature, 1.80 ± 0.05 m; body mass, 76 ± 8 kg; maximum oxygen uptake (V˙O2max), 4.49 ± 0.35 L·min−1; power at V˙O2max, 399 ± 31 W) gave written informed consent to volunteer. Ethical approval was obtained from the Northumbria University Faculty of Health and Life Sciences Ethics Committee. All participants were in regular cycling training and competition in road and/or time-trial disciplines.
Design
Participants visited the laboratory on four separate occasions to complete a preliminary ramp test, followed by three experimental constant-load cycling trials to the limit of tolerance in a randomized crossover design. Exercise intensities were set relative to the cardiorespiratory responses measured during the preliminary ramp test. Two of the trials were set at intensities predicted to elicit times to the limit of tolerance of 2–4 min, and 8–15 min, and physiological responses consistent with exercise above critical power in the severe-intensity domain. One trial was set at an intensity predicted to elicit a time to task failure of >30 min and steady-state physiological responses consistent with sustainable exercise in the heavy-intensity domain. Trials were conducted at the same time of day (±1 h), separated by a minimum of two and a maximum of 7 d. Before each experimental trial, participants were instructed to refrain from caffeine (for at least 12 h) and strenuous exercise (for at least 24 h) and to arrive at the laboratory 2 h postprandial in a fully rested, hydrated state. Participants completed a 48-h food and activity diary before their first experimental trial and were required to replicate their exercise and nutrition as closely as possible before each subsequent trial. Cardiorespiratory, blood lactate concentration, and RPE were recorded during each trial and measures of neuromuscular function were assessed pretrial and within 2.5 min posttrial.
Procedures
Preliminary visit
Participants attended the laboratory to complete an incremental ramp test to measure V˙O2max and respiratory thresholds and habituated to the measurement tools of the study, specifically electrical stimulation of the femoral nerve and magnetic stimulation of the motor cortex pre- and postexhaustive cycling exercise. The incremental ramp test was performed on an electromagnetically braked cycle ergometer (Excalibur Sport, Lode, Groningen, Netherlands) and consisted of cycling at 100 W followed by a continuous ramped increase in power of 30 W·min−1 to the limit of tolerance. The test was terminated when participant cadence reduced by >20 rpm from their self-selected cadence for the test. Expired air was analyzed breath by breath using an online system (Oxycon Pro, Care Fusion, Hoechberg, Germany). Oxygen (O2) and carbon dioxide (CO2) concentrations were analyzed via a paramagnetic chemical fuel cell and nondispersive infrared cell, respectively. Before each test, the analyzers were calibrated using ambient air, and a gas of known O2 (14.00%) and CO2 (6.00%) concentrations. Ventilatory volumes were inferred from the measurement of gas flow using a digital turbine transducer (volume, 0–10 L; resolution, 3 mL; flow, 0–15 L·s−1), calibrated before each test using a 3-L syringe (Hans Rudolph Inc., Shawnee, KS) and attached to a mask via a transducer holder (dead space of 30 mL). The gas exchange threshold was determined using multiple parallel methods as follows: 1) the first disproportionate increase in carbon dioxide output (V˙CO2) relative to V˙O2, 2) an increase in the ventilatory equivalent for oxygen (V˙E/V˙O2) with no increase in the ventilatory equivalent for carbon dioxide (V˙E/V˙CO2), and 3) an increase in end-tidal O2 tension with no fall in end-tidal CO2 tension (8,40). Respiratory compensation point (RCP) was determined from plots of minute ventilation (V˙E) versus V˙CO2 (8). Maximum oxygen uptake was calculated as the highest 30 s mean value attained before test termination, and the end test power was recorded as power at V˙O2max (Pmax).
Experimental trials
Participants completed three constant-load cycling trials to the limit of tolerance. The exercise intensities were determined from the respiratory indices and associated intensities measured during the preliminary ramp test. Two trials were set to elicit non–steady-state physiological responses consistent with exercise in the severe-intensity domain with predicted exercise durations of 2–4 min (hereafter called S+) and 8–15 min (hereafter called S−). The third trial was set at a sustainable intensity (hereafter called RCP), with a predicted exercise duration of >30 min. The S+ trial was set at Pmax, which approximates the upper limit of the severe-intensity domain. The S− trial was set at an exercise intensity equivalent to 60% of the difference between gas exchange threshold and V˙O2max, which would also elicit responses consistent with exercise above critical power in the severe domain, but for a greater duration than S+ (20). The RCP trial was set at the power output associated with RCP, as this intensity approximates an upper limit for sustainable exercise (15). When determining constant-load exercise intensities from a ramp protocol, it is recommended the intensity be adjusted to accommodate the mean rise time of V˙O2 during ramp exercise, which approximates two-thirds of the ramp rate (i.e., 20 W [41]). For all trials, this recommendation was adhered to, and 20 W was subtracted from the calculated exercise intensity. Each experimental trial was preceded by a 5-min warm-up at 150 W before a “step” increase in exercise intensity. Participants remained seated throughout exercise, and the test was terminated when participants could not maintain a cadence within 10 rpm of their self-selected cadence for the test. Expired air was measured breath by breath throughout. Fingertip capillary blood was sampled after the warm-up, at 5-min intervals during constant-load exercise and 3 min posttrial, and immediately analyzed for [lactate] using an automated instrument (Biosen C_Line, EFK Diagnostic, Germany). At the same time points during exercise, participants were asked to provide an RPE taking into account all sensations of physical stress, effort, and fatigue (9).
Neuromuscular function
Measures of neuromuscular function were assessed pre- and postexercise using electrical stimulation of the femoral nerve and transcranial magnetic stimulation (TMS) of the motor cortex, with evoked force and EMG responses recorded from the knee extensors of the dominant leg at rest and during voluntary contraction. After two practice attempts to ensure adequate potentiation, participants completed three isometric maximum voluntary contractions (MVC) of the knee extensors separated by a 60-s rest, with femoral nerve stimulation delivered during and ~2 s after to measure voluntary activation (VA) and potentiated quadriceps twitch force (Qtw,pot). Subsequently, TMS was delivered during brief (3–5 s) contractions at 100%, 75%, and 50% MVC, separated by a 5-s rest, for the determination of corticospinal VA. This procedure was repeated three times with a 30-s rest between each set. Finally, TMS (×8 stimulations) and electrical nerve stimulation (×5 stimulations) were delivered during submaximal contraction at 20% MVC for determination of corticospinal excitability. Immediately postexercise, these measures were repeated. In accordance with other similar investigations of exercise-induced fatigue of the knee extensors, the assessment of VA measured with motor nerve and motor cortical stimulation was completed within 2.5 min of exercise cessation (20,32,34,38). The rapid nature of this procedure is necessary to capture the extent of fatigue elicited by the exercise before it dissipates (18), and the duration (2–2.5 min) was consistent between trials. Further details on these procedures are presented in subsequent sections.
Force and EMG recordings
During stimulations, participants sat upright in a custom-built chair with hips and knees at 90° of flexion. A calibrated load cell (MuscleLab force sensor 300, Ergotest Technology, Norway) was attached via a noncompliant cuff positioned on the participant’s right leg, superior to the malleoli, to measure knee extensor force (N). Surface Ag/AgCl electrodes (Kendall H87PG/F; Covidien, Mansfield, MA) were placed 2 cm apart over the rectus femoris (RF), vastus lateralis (VL), and biceps femoris to record the compound muscle action potential (M-wave) elicited by the electrical stimulation of the femoral nerve, the motor evoked potential (MEP) elicited by TMS, and the root-mean-square amplitude during isometric and dynamic (cycling) exercise. The position of the electrodes was consistent with the guidelines of the Surface EMG for the Non-invasive Assessment of Muscles, and a reference electrode was placed over the patella. Electrode placement was marked with permanent ink to ensure a consistent placement between trials. Signals were amplified: gain ×1000 for EMG and ×300 for force (CED 1902; Cambridge Electronic Design, Cambridge, UK), band-pass filtered (EMG only: 20–2000 Hz), digitized (4 kHz; CED 1401, Cambridge Electronic Design), and analyzed offline (Spike2 v7.12, Cambridge Electronic Design).
Motor nerve stimulation
Single electrical stimuli (200 μs duration) were applied to the right femoral nerve using a constant-current stimulator (DS7AH Digitimer Ltd., Welwyn Garden City, UK) via adhesive surface electrodes (CF3200; Nidd Valley Medical Ltd., Harrogate, UK) at rest and during voluntary contraction. The cathode was positioned over the nerve in the femoral triangle in the location that elicited the maximum quadriceps twitch amplitude (Qtw) and the M-wave (Mmax) at rest. The anode was positioned midway between the greater trochanter and the iliac crest. The optimal stimulation intensity was determined as the minimum current that elicited maximum values of Qtw and Mmax at rest. To ensure a supramaximal stimulus and to account for fatigue-dependent changes in axonal excitability, the intensity was increased by 30% and was not different between trials (211 ± 71, 215 ± 76, and 219 ± 74 mA, P = 0.13).
TMS
Single pulse magnetic stimuli (1 ms duration) were delivered to the left motor cortex via a concave double cone coil (110 mm diameter; maximum output, 1.4 T) powered by a monopulse magnetic stimulator (Magstim 200; The Magstim Company Ltd., Whitland, UK). The coil position was tilted and held lateral to the vertex over the area relating to Brodmann area 4 of the primary motor cortex, generating a posterior–anterior intracranial current flow within the left hemisphere. The exact “hot spot” (marked in semipermanent ink for reproducible placement between trials) was determined as the coil position that elicited the largest MEP in the RF during a submaximal (20% MVC) contraction and a concurrent small MEP in the antagonist muscle. For measures of corticospinal excitability, active motor threshold was determined as the stimulator output that elicited a consistent MEP of approximately 0.2 mV in at least three of five stimulations during 20% MVC (37). The stimulator output was increased by 20% and was not different between trials (48% ± 7%, 48% ± 7%, and 47% ± 6% of mean stimulator output, P = 0.58). Setting the stimulator intensity relative to active motor threshold in this manner ensures that the resulting MEP is not close to saturation on the stimulus–response curve (39) permitting a suitable window for exercise-induced changes. The use of a submaximal contraction allows for the monitoring of the status of smaller motoneurons (29). For the measurement of corticospinal VA, the stimulator output was set to produce the largest superimposed twitch force during a contraction at 50% MVC (37), which averaged 67 ± 17 N, or 12% ± 5% of MVC. The stimulation intensity, which was not different between trials (63% ± 5%, 63% ± 4%, and 62% ± 4% of mean stimulator output, P = 0.68), elicited a large MEP in the RF (approximately 70% of Mmax during contractions ≥50% MVC), indicating that the TMS stimulus activated a high proportion of knee extensor motor units while causing only a small MEP in the antagonist (0.40 mV on average, or <10% of RF Mmax during knee-extensor contractions). Participants were given specific instructions to achieve a plateau in the target force when contracting at varying force levels while receiving TMS (22).
Data analysis
VA measured via motor nerve stimulation was quantified using the twitch interpolation method: VA (%) = (1 − [SIT / Qtw,pot] × 100), where SIT is the amplitude of the superimposed twitch force measured during MVC, and Qtw,pot is the amplitude of the resting potentiated twitch force assessed 2 s post-MVC. For motor cortical stimulation, VA was assessed by the measurement of the superimposed twitch responses to TMS at 100%, 75%, and 50% MVC. As corticospinal excitability increases during voluntary contraction, it is necessary to estimate the amplitude of the resting twitch in response to motor cortex stimulation (21,39). The amplitude of the estimated resting twitch (ERT) was calculated as the y-intercept of the linear regression between the mean amplitude of the superimposed twitches evoked by TMS at 100%, 75%, and 50% MVC and voluntary force; regression analyses confirmed the existence of a linear relationship both pre- and postexercise (r2 = 0.90 ± 0.05 and 0.86 ± 0.11, respectively). VA (%) measured with TMS was subsequently calculated as (1 − [SIT / ERT] × 100). The reproducibility and validity of this procedure for the knee extensors has been previously established (21). The peak-to-peak amplitude and the area of the evoked Mmax and MEP responses were quantified offline. Corticospinal excitability was determined as the ratio between the MEP and the M-wave responses measured during a 20% MVC from eight cortical and five motor nerve stimulations. Contraction strength was adjusted posttrial to equate to 20% of the fatigued MVC force. The EMG of the VL and RF during the cycling bout was band-pass filtered in both directions between 20 and 450 Hz using a fourth-order zero lag Butterworth filter. The root mean square of the EMG amplitude was measured for 10 consecutive pedal revolutions from the middle of minute 4 of the warm-up, and the first and final minutes of cycling exercise, to quantify changes in neuromuscular activation. Onsets and offsets of EMG bursts were determined visually by the same investigator (11,25). The warm-up and the first minute EMG data were normalized to the Mmax measured pretrial, and the final minute EMG data were normalized to the posttrial Mmax (average of three electrical stimulations at rest for both).
Reproducibility coefficients
Typical error as a coefficient of variation (CV, %) and intraclass correlation coefficients (ICC [23]) between the pretrial scores were calculated to quantify the reproducibility of neuromuscular function measures. Reproducibility was high for MVC (ICC = 0.96, CV = 4.4%), Qtw,pot (ICC = 0.90, CV = 4.8%), motor nerve VA (ICC = 0.89, CV = 3.1%), corticospinal VA (ICC = 0.90, CV = 2.6%), and moderate for ERT (ICC = 0.85, CV = 10.0%), Mmax (RF: ICC = 0.70, CV = 24.7%; VL: ICC = 0.79, CV = 25.2%) and MEP/Mmax ratio (RF: ICC = 0.74, CV = 13.6%; VL: ICC = 0.79, CV = 26.3%). The reproducibility of the cardiorespiratory responses (V˙E, V˙O2, V˙CO2, respiratory exchange ratio, RER, and HR) was determined from responses to the standardized 5-min warm-up and was considered high for all measures (ICC range = 0.87–0.93, CV range = 4.0%–5.3%).
Statistical analysis
All analyses were planned a priori. For all neuromuscular measures, the expected effects of exhaustive exercise on measures of fatigue were assessed using Student’s paired-sample t-tests. One-way repeat-measures ANOVA analyses of the pre- to postexercise change scores were used to assess the effect of exercise intensity (S+ vs S− vs RCP) on measures of fatigue and neuromuscular function. Differences between trials for exercise time, power output, and cardiorespiratory and metabolic responses were assessed using the same procedure. Trial (S+ vs S− vs RCP) by time (warm-up vs first min vs final minute) factorial repeat-measures ANOVA values were used to assess EMG responses; significant main effects were followed up using one-way repeat-measures ANOVA. For all ANOVA analyses, pairwise comparisons were made using Tukey’s test. Friedman’s ANOVA with post hoc Wilcoxon signed-ranks test was used for nonparametric data (RPE). The assumptions underpinning these statistical procedures were verified, and all data were considered normal. Descriptive data are presented as means ± SD in text, tables, and figures. Statistical analysis was conducted using Graphpad (Version 5.03, GraphPad Prism 5, La Jolla, CA). Statistical significance was assumed at P < 0.05.
RESULTS
Exercise responses
Power output for S+ (379 ± 31 W), S− (305 ± 23 W), and RCP (254 ± 26 W) equated to relative intensities of 100% Pmax, 76% ± 3% Pmax, and 64% ± 5% of Pmax. The resulting exercise times to the limit of tolerance for each trial were 3.14 ± 0.59, 11.11 ± 1.86, and 42.14 ± 9.09 min for S+, S−, and RCP, respectively.
The mean cardiorespiratory (Fig. 1) and blood [lactate] (Fig. 2A) responses to S+ and S− were consistent with exercise above critical power in the severe-intensity domain, and for RCP consistent with exercise in the heavy domain. For S+ exercise, oxygen uptake rose rapidly (Fig. 1C), reaching peak values of 98% ± 8% of maximum, and peak RER and V˙CO2 were higher compared with S− and RCP exercise (P < 0.05). For S− exercise, oxygen uptake also rose rapidly, followed by a gradual increase to task failure where peak values averaged 98% ± 6% of maximum oxygen uptake. For RCP, an oxygen uptake slow component was observed, with peak values averaging 91% ± 6% of V˙O2max (Fig. 1B). Blood [lactate] postexercise was highest after S+ (12.3 ± 1.8 mmol·L−1) compared with both S− (10.6 ± 2.1 mmol·L−1, P < 0.05) and RCP (5.9 ± 1.8 mmol·L−1, P < 0.001) exercise, and higher after S− compared with RCP (P < 0.001) (Fig. 2A).
;)
FIGURE 1: Minute ventilation (V˙E, A), respiratory exchange ratio (B), oxygen uptake (V˙O2, C), and carbon dioxide output (V˙CO2, D) during severe-intensity (S+ ◯ and S− ▴) and heavy-intensity (RCP ▪) constant-load cycling exercise. Values are presented as mean ± SD (n = 12). The vertical dashed line on the x axis marks the end of the 5-min warm-up and the start of the constant-load trial. The horizontal error bars on the final data point show the variability in time to the limit of tolerance for each trial. Symbols indicate a statistically significant difference (P < 0.05) in the peak scores compared with the following: *different from S−; †different from RCP.
FIGURE 2: Blood lactate (A), RPE (B), and RPE relative to trial duration (C) responses to severe-intensity (S+ ◯ and S− ▴) and heavy-intensity (RCP ▪) constant-load cycling exercise. The vertical dashed line on the x axis marks the end of the 5-min warm-up and the start of the constant-load trial. The horizontal error bars on the final data point in panels A and B show the variability in time to the limit of tolerance for each trial. Values are presented as mean ± SD (n = 12). Symbols indicate a statistically significant difference (P < 0.05) in the peak trial scores compared with the following: *different from S−; †different from RCP.
Surface EMG data were successfully sampled for all three trials in nine participants. Estimated muscle activation (RMS/Mmax) was different across time (P < 0.001), between trials (P = 0.001), and responded differently to S+, S−, and RCP (time × trial P = 0.001, Fig. 3). There were no differences observed in muscle activation during the warm-up between trials. During the first minute of exercise, EMG RMS/Mmax was higher in S+ compared with RCP (P = 0.01), and in the final minute, it was higher in S+ compared with both S− (P = 0.04) and RCP (P = 0.001). The increase in EMG RMS/Mmax during S+ exercise was higher than RCP between warm-up and the first minute of exercise (P = 0.001), and higher than both S− (P = 0.04) and RCP (P = 0.001) between the first and the final minute of exercise (Fig. 3).
FIGURE 3: Muscle activation estimated from the EMG of the VL and RF during a standardized warm-up and the first and final minute of severe-intensity (S+ ◯ and S− ▴) and heavy-intensity (RCP ▪) constant-load cycling exercise. Values are presented as mean ± SD (n = 9). Symbols above error bars indicate a statistically significant difference between trials for the measured time point. Symbols on horizontal lines indicate a statistically significant interaction between time points (all P < 0.05). *Different from S−. †Different from RCP.
For RPE, participants reported higher peak scores for S+ compared with RCP exercise (P < 0.05), with no difference between S+ and S− or between S− and RCP exercise (Fig. 2B). When expressed relative to exercise time, the RPE response was similar between trials (Fig. 2C).
Neuromuscular responses pre- and postexercise
One participant exhibited small responses to TMS (MEP:Mmax ratio in RF < 60%, superimposed twitch force during 50% MVC < 5% MVC force). Low MEP:Mmax ratios and small SIT are indicative of an incomplete activation of the available motoneuron pool by the magnetic stimulus, which could invalidate the measurement of VA. This participant was subsequently excluded from analysis of data elicited by TMS.
Exercise resulted in a similar level of global fatigue after all trials, as evidenced by substantial reductions in MVC force (P < 0.05) that were not different between trials (−111 ± 54 N, −93 ± 48 N, and −98 ± 59 N for S+, S−, and RCP exercise, respectively, P = 0.61) (Fig. 4A). The loss in voluntary force was accompanied by significant peripheral and central fatigue in all trials (all P < 0.05, Fig. 4); however, there were differences in the magnitude of central and peripheral fatigue between trials. Peripheral fatigue (ΔQtw,pot amplitude) was greater after S+ exercise (−33% ± 9%) compared with both S− (−16% ± 9%, P < 0.001) and RCP exercise (−11% ± 9%, P < 0.001), and greater after S− compared with RCP exercise (P < 0.05) (Fig. 4B). For central fatigue, this pattern was reversed. For motor nerve estimates of VA, central fatigue was less after S+ (−2.7% ± 2.2%) compared with RCP (−9.0% ± 4.7%, P < 0.01) but not after S+ compared with S− (−6.3% ± 5.5%, P = 0.056) or after S− compared with RCP (P = 0.084) (Fig. 4C). For motor cortical estimates of VA, no statistical differences were observed after S+ (−2.7% ± 3.5%) compared with RCP (−9.2% ± 8.0%, P = 0.06) or S− (−7.9% ± 7.4%, P = 0.07), or after S− compared with RCP (P = 0.46) (Fig. 4D). Changes in VA were not accompanied by any changes in MEP or M-wave properties measured during a 20% MVC (Table 1) or during higher force contractions (Table, Supplementary Digital Content 1, Neuromuscular function and surface EMG responses to motor nerve and motor cortical stimulation, https://links.lww.com/MSS/A683).
TABLE 1: MEP, muscle compound action potentials (M-waves), and corticospinal excitability (MEP:M-wave) measured in VL and RF during a submaximal contraction (20% MVC) before and after severe-intensity (S+ and S−) and heavy-intensity (RCP) constant-load cycling exercise.
FIGURE 4: Pre- to posttrial percentage change in maximum voluntary contraction (A), potentiated twitch force (B), VA measured with motor nerve stimulation (C) (n = 12), and VA measured with cortical stimulation (VATMS) (n = 11) (D) after severe-intensity (S+ and S−) and heavy-intensity (RCP) constant-load cycling exercise. Values are presented as mean ± SD. Symbols indicate a statistically significant difference (P < 0.05) compared with the following: *different from S−; †different from RCP.
DISCUSSION
This study shows that the magnitude of central and peripheral fatigue observed after constant-load locomotor cycling varies with exercise intensity and duration. Specifically, the data suggest a task-dependent contribution of peripheral and central fatigue; peripheral fatigue is exacerbated with increases in exercise intensity, whereas central fatigue tends to increase as exercise intensity decreases and duration increases. For non–steady-state exercise in the severe domain, the degree of end-exercise peripheral fatigue was exacerbated at higher intensities, a finding incongruent with the proposal of an individual critical threshold for muscle fatigue after high-intensity locomotor exercise above critical power.
Contrary to our hypothesis, the degree of end-exercise peripheral fatigue was not consistent between exercise intensities in the severe domain, a finding that does not support the proposal that muscle fatigue after high-intensity locomotor exercise is regulated to an individual critical threshold. The concept of an individual critical threshold, or sensory tolerance limit, was originally proposed by Amann et al. (1–7), who completed a series of studies demonstrating that the magnitude of exercise-induced peripheral fatigue was remarkably similar after locomotor exercise under conditions of prefatiguing exercise (3), altered arterial oxygen saturation (4), and, perhaps most convincingly, was exacerbated only under conditions of impaired afferent feedback (2,5). Similar support has also been provided for all-out repeat-sprint exercise with and without prefatiguing exercise (24). These studies proposed a feedback loop whereby inhibitory input from group III/IV afferents in the active musculature acts on the central nervous system to compel a reduction in exercise intensity to prevent excessive homeostatic disruption and to restrain the development of muscle fatigue to a consistent limit that would not be exceeded in normal conditions. Although conceptually attractive, our previous work comparing different durations of cycling time-trial exercise (38) and evidence from the work of others comparing single versus double leg exercise (32) have demonstrated that any peripheral limit is task dependent, a point which the authors of this proposal were careful to emphasize (7). Nonetheless, in the current study, we hypothesized that in an individual, consistent limit for peripheral fatigue might exist for non–steady-state exercise in the severe-intensity domain. This posit was based on the distinct, nonlinear responses to severe-intensity exercise (26), the observation of a consistent reduction in potentiated twitch force of ~35% after high-intensity cycling exercise (2,3,5,6), and the finding that peripheral fatigue is similar after single-limb, intermittent, isometric knee extensor contractions at varying intensities above the critical torque (10). Contrary to our original hypothesis, the greater reduction in potentiated twitch force observed after the highest intensity trial suggests that the degree of peripheral fatigue incurred after exhaustive locomotor exercise above critical power is intensity and duration dependent and not regulated to a critical limit.
An alternative explanation for the differing magnitude of end-exercise peripheral fatigue observed in this and other studies (2,5,24,32,33) could simply be related to the force and motoneuron recruitment strategy required to complete the task. In the current study, the force demands of the task and the subsequent activation of muscle were highest in the S+ trial, where peripheral fatigue was also highest; the higher peripheral fatigue could simply be explained by the recruitment and subsequent fatigue of a greater portion of the available motoneuron pool. This proposal would also explain several previous observations relating to the critical threshold concept. For example, observations of a greater end-exercise peripheral fatigue after single versus double-limb exercise (32,33), and after locomotor cycling exercise with afferent blockade (2,5), could both be explained by a greater activation of muscle, as evidenced by the higher EMG that was concurrent with greater end-exercise peripheral fatigue in these experimental trials. Additionally, the highest absolute reductions in potentiated twitch force (measured with single electrical stimuli) have been reported after repeat sprint cycling exercise: −51% (24) compared with a range of −33% to −40% after high-intensity time-trial (5,38) and constant-load cycling exercise (2). Repeat sprint exercise would theoretically require repeated maximum activation of the locomotor musculature, with compromised recovery, and as such, the high force demands of the task would elicit a greater recruitment and fatigue of larger, faster motor units and a consequent higher level of muscle fatigue. In support of this proposal, Decorte et al. (14) demonstrated that the increase in quadriceps EMG during exhaustive constant-load cycling was significantly correlated with the reduction in quadriceps potentiated twitch force. A limitation to this explanation is an acknowledgment that estimates of muscle activation via surface EMG are subject to several valid critiques (17,27), not least the insensitivity of measures of muscle activation to small differences in exercise intensities. These limitations notwithstanding, the higher EMG that accompanied the higher power output in the S+ trial gives credence to the interpretation that this is indicative of a higher degree of muscle activation. Thus, the proposal that peripheral fatigue after locomotor exercise can be explained by the force and/or motor unit recruitment strategy required for the task remains plausible and provides a better fit with the available data than a model that emphasizes regulation to an individual critical threshold.
Although these data question the concept of an individual critical threshold for peripheral fatigue, the proposal that group III/IV muscle afferent feedback acts in an inhibitory manner to contribute to the regulation of exercise intensity remains a logical proposition. Without such feedback, regulation is almost certainly compromised, at least for high-intensity locomotor exercise lasting <10 min (5). Indeed, Amann et al. (2,5) clearly demonstrate that when such feedback is blocked with the administration of an intrathecal opioid analgesic, participants self-select exercise intensities and/or adopt inappropriate recruitment strategies that result in significant additional peripheral fatigue in comparison with a control, with no improvement in performance. These data clearly support the idea that group III/IV afferent feedback is important for the regulation of high-intensity locomotor exercise. However, the findings of the present study, the previous work comparing different durations of self-paced cycling exercise (38), and the observation that prior fatiguing arm exercise can modulate end-exercise peripheral fatigue in high-intensity cycling (25) indicate that peripheral muscle fatigue contributes to, rather than determines, the decision to terminate or modulate locomotor exercise intensity.
In accordance with our hypothesis, longer-duration, heavy-intensity exercise terminated with more central fatigue and less peripheral fatigue compared with severe-intensity exercise. This duration-dependent contribution of central fatigue is similar to that previously observed in self-paced (38) and constant-load cycling (14) and single-limb intermittent contractions (10), and it indicates that exercise tolerance in longer-duration tasks is mediated to a greater extent by mechanisms of central fatigue. Despite differences in the etiology of neuromuscular fatigue, the sensory perception of exertion rose similarly over time independent of exercise duration/intensity (Fig. 2C). This is a consistent response to various exercise tasks under varying conditions, which has prompted the development of holistic models that emphasize a mediating role for sensory perception in the etiology of fatigue (28,30). Interestingly, the highest RPE at exercise termination was reported in the highest intensity trial, where peripheral fatigue was also highest, a finding concurrent with our previous work in self-paced exercise (38). This could support the idea that afferent feedback contributes to the sensory perception of fatigue and the decision to terminate exercise (30), or it could be argued that the shorter duration of the trial and the lower degree of central fatigue permitted a higher potential motivation for the task and a consequent higher tolerance to a high perception of effort (28). Although a definitive explanation is elusive, it is clear that the decision to terminate exercise was influenced differently depending on the intensity of the trial, and understanding how the mediating inputs to this decision vary across exercise tasks remains an important endeavor to advance our understanding of the factors that limit or modulate exercise performance.
The task-dependent, intensity-domain specificity of peripheral fatigue observed in the present study after constant-load cycling exercise supports our previous work in different durations of self-paced cycling exercise (38), with one caveat. In this previous study, we observed a similar degree of peripheral fatigue after 20- and 40-km time-trial exercise (−31% and −29% on average, respectively) despite significant differences in the duration (32 vs 66 min) and average intensity (279 vs 255 W) of the bouts (38). We hypothesized that this lack of difference could reflect a similarity in the exercise domain in which the trials were completed, predominantly in the heavy domain and below critical power. However, in the present study using the same femoral nerve stimulation methods, and a similar population of well-trained cyclists, the absolute decline in potentiated twitch force after 45 min of constant-load cycling in the heavy exercise intensity domain was only 11%. On the basis of this new information and the proposal that peripheral fatigue is determined in a task-dependent manner based on the force and/or motoneuron recruitment strategies required for the task, we now consider it likely that the higher and similar degree of peripheral fatigue observed after 20- and 40-km self-paced exercise was determined primarily by the finishing sprint, which was similar between trials (38). Such a sprint at the end of the trial is common in self-paced exercise and requires higher force and therefore recruitment of higher threshold motor units. Collectively, this reasoning could explain the greater degree of peripheral fatigue observed after self-paced exercise compared with the relatively modest degree of peripheral fatigue observed in the present study after a similar duration and intensity constant-load bout. These data therefore suggest that the magnitude of peripheral fatigue observed at the end of a self-paced bout of exercise likely reflects the recent contractile history of the muscle and emphasizes the importance and challenge of developing methods to ascertain the time-course development of fatigue during as well as after exercise.
Corticospinal excitability was unchanged in response to locomotor exercise, when measured >2 min postexercise. This response is common in studies using locomotor exercise where corticospinal excitability is measured pre- and postexercise (34,38). However, the lack of change from pre- to postexercise might not fully reflect modulations in corticospinal excitability that could occur during exercise. Indeed, when responses are elicited during locomotor exercise using a bespoke experimental set-up, there is evidence to suggest that corticospinal excitability is reduced (35) and intracortical inhibition is increased (36), which might therefore contribute to the etiology of fatigue. The lack of change in measures of excitability commonly observed postexercise might be explained by a rapid recovery and the lack of specificity of the task in which it is measured. A limitation of the present study is that the measurement of corticospinal excitability took place >2 min postexercise during isometric contraction, and as such any change in response to exercise might not be elucidated.
The estimate of exercise intensities in the present study was based on the ventilatory and performance responses to a ramp exercise protocol. Two of these intensities (S+ and S−) were set to elicit physiological responses consistent with exercise above critical power in the severe-intensity domain, for exercise durations between 2 and 4 min and between 8 and 15 min, respectively. In both of these trials, oxygen uptake progressively rose to values that were within 2% of maximum oxygen uptake at exercise termination (Fig. 1), and blood [lactate] was not stabilized (Fig. 2A), indicative of exercise in the severe domain. The third trial (RCP) was set at the RCP, which approximates the critical power (15), and was designed to elicit exercise durations >30 min and physiological responses consistent with sustainable exercise. Although it is acknowledged that the weight of evidence suggests the RCP and critical power are distinct thresholds (12,31), the exercise duration in RCP exceeded 30 min for all participants, and cardiorespiratory (Fig. 1) and blood [lactate] (Fig. 2A) responses were submaximal and stable for the duration of the RCP trial on a group level, indicative of exercise below critical power in the heavy-intensity domain. However, the range in exercise durations (31–59 min) suggests a less precise estimate of this intensity, which was likely in the heavy domain for most, but not all, participants. This limitation notwithstanding, the contribution of central and peripheral processes to fatigue after this longer-duration bout was distinctive from the shorter, higher-intensity bouts. A more precise estimate of the critical power would allow future research to better assess any differences in the etiology of neuromuscular fatigue at exercise intensities below critical power.
To conclude, the contribution of central and peripheral processes to fatigue after constant-load cycling exercise differs in a task-dependent manner. Central fatigue is exacerbated as exercise duration increases and intensity decreases, whereas peripheral fatigue is greater at higher intensities and shorter durations of exercise. These data suggest that the extent of end-exercise peripheral fatigue is not regulated to an individual critical threshold but is determined by the force and/or motor unit recruitment strategies required for the task. These data have implications for the concept of a critical limit for peripheral fatigue and our understanding of the etiology of fatigue under different locomotor exercise conditions.
The authors have no competing interests to declare. No external funding was provided for the study. The results of the study do not constitute endorsement by the American College of Sports Medicine.
REFERENCES
1. Amann M. Central and peripheral fatigue: interaction during cycling exercise in humans.
Med Sci Sports Exerc. 2011;43(11):2039–45.
2. Amann M, Blain GM, Proctor LT, Sebranek JJ, Pegelow DF, Dempsey JA. Implications of group III and IV muscle afferents for high-intensity endurance exercise performance in humans.
J Physiol. 2011;589(Pt 21):5299–309.
3. Amann M, Dempsey JA. Locomotor muscle fatigue modifies central motor drive in healthy humans and imposes a limitation to exercise performance.
J Physiol. 2008;586(1):161–73.
4. Amann M, Eldridge MW, Lovering AT, Stickland MK, Pegelow DF, Dempsey JA. Arterial oxygenation influences central motor output and exercise performance via effects on peripheral locomotor muscle fatigue in humans.
J Physiol. 2006;575(Pt 3):937–52.
5. Amann M, Proctor LT, Sebranek JJ, Pegelow DF, Dempsey JA. Opioid-mediated muscle afferents inhibit central motor drive and limit peripheral muscle fatigue development in humans.
J Physiol. 2009;587(Pt 1):271–83.
6. Amann M, Romer LM, Pegelow DF, Jacques AJ, Hess CJ, Dempsey JA. Effects of arterial oxygen content on peripheral locomotor muscle fatigue.
J Appl Physiol. 2006;101(1):119–27.
7. Amann M, Secher NH. Point: afferent feedback from fatigued locomotor muscles is an important determinant of endurance exercise performance.
J Appl Physiol (1985). 2010;108(2):452–4.
8. Beaver WL, Wasserman K, Whipp BJ. A new method for detecting anaerobic threshold by gas exchange.
J Appl Physiol (1985). 1986;60(6):2020–7.
9. Borg GA. Psychophysical bases of perceived exertion.
Med Sci Sports Exerc. 1982;14(5):377–81.
10. Burnley M, Vanhatalo A, Jones AM. Distinct profiles of neuromuscular fatigue during muscle contractions below and above the critical torque in humans.
J Appl Physiol (1985). 2012;113(2):215–23.
11. Chapman AR, Vicenzino B, Blanch P, Hodges PW. Patterns of leg muscle recruitment vary between novice and highly trained cyclists.
J Electromyogr Kinesiol. 2008;18(3):359–71.
12. Cross TJ, Sabapathy S. The respiratory compensation “point” as a determinant of O
2 uptake kinetics?
Int J Sports Med. 2012;33(10):854 author reply 855.
13. de Souza KM, Dekerle J, Salvador PC, et al. Rate of utilization of a given fraction of W’ (the curvature constant of the power-duration relationship) does not affect fatigue during severe-intensity exercise.
Exp Physiol. 2016;101(4):540–8.
14. Decorte N, Lafaix PA, Millet GY, Wuyam B, Verges S. Central and peripheral fatigue kinetics during exhaustive constant-load cycling.
Scand J Med Sci Sports. 2012;22(3):381–91.
15. Dekerle J, Baron B, Dupont L, Vanvelcenaher J, Pelayo P. Maximal lactate steady state, respiratory compensation threshold and critical power.
Eur J Appl Physiol. 2003;89(3–4):281–8.
16. Enoka RM, Stuart DG. Neurobiology of muscle fatigue.
J Appl Physiol (1985). 1992;72(5):1631–48.
17. Farina D, Merletti R, Enoka RM. The extraction of neural strategies from the surface EMG.
J Appl Physiol (1985). 2004;96(4):1486–95.
18. Froyd C, Millet GY, Noakes TD. The development of peripheral fatigue and short-term recovery during self-paced high-intensity exercise.
J Physiol. 2013;591(5):1339–46.
19. Gandevia SC. Spinal and supraspinal factors in human muscle fatigue.
Physiol Rev. 2001;81(4):1725–89.
20. Goodall S, Gonzalez-Alonso J, Ali L, Ross EZ, Romer LM. Supraspinal fatigue after normoxic and hypoxic exercise in humans.
J Physiol. 2012;590(Pt 11):2767–82.
21. Goodall S, Romer LM, Ross EZ. Voluntary activation of human knee extensors measured using transcranial magnetic stimulation.
Exp Physiol. 2009;94(9):995–1004.
22. Gruet M, Temesi J, Rupp T, Levy P, Millet GY, Verges S. Stimulation of the motor cortex and corticospinal tract to assess human muscle fatigue.
Neuroscience. 2013;231:384–99.
23. Hopkins WG. Spreadsheets for analysis of validity and reliability.
Sportscience. 2015;19:36–42.
24. Hureau TJ, Olivier N, Millet GY, Meste O, Blain GM. Exercise performance is regulated during repeated sprints to limit the development of peripheral fatigue beyond a critical threshold.
Exp Physiol. 2014;99(7):951–63.
25. Johnson MA, Sharpe GR, Williams NC, Hannah R. Locomotor muscle fatigue is not critically regulated after prior upper body exercise.
J Appl Physiol (1985). 2015;119(7):840–50.
26. Jones AM, Wilkerson DP, DiMenna F, Fulford J, Poole DC. Muscle metabolic responses to exercise above and below the “critical power” assessed using 31
P-MRS.
Am J Physiol Regul Integr Comp Physiol. 2008;294(2):R585–93.
27. Keenan KG, Farina D, Merletti R, Enoka RM. Amplitude cancellation reduces the size of motor unit potentials averaged from the surface EMG.
J Appl Physiol (1985). 2006;100(6):1928–37.
28. Marcora SM, Staiano W. The limit to exercise tolerance in humans: mind over muscle?
Eur J Appl Physiol. 2010;109(4):763–70.
29. McNeil CJ, Giesebrecht S, Gandevia SC, Taylor JL. Behaviour of the motoneurone pool in a fatiguing submaximal contraction.
J Physiol. 2011;589(Pt 14):3533–44.
30. Millet GY. Can neuromuscular fatigue explain running strategies and performance in ultra-marathons? The flush model.
Sports Med. 2011;41(6):489–506.
31. Poole DC, Ward SA, Gardner GW, Whipp BJ. Metabolic and respiratory profile of the upper limit for prolonged exercise in man.
Ergonomics. 1988;31(9):1265–79.
32. Rossman MJ, Garten RS, Venturelli M, Amann M, Richardson RS. The role of active muscle mass in determining the magnitude of peripheral fatigue during dynamic exercise.
Am J Physiol. 2014;306(12):R394–340.
33. Rossman MJ, Venturelli M, McDaniel J, Amann M, Richardson RS. Muscle mass and peripheral fatigue: a potential role for afferent feedback?
Acta Physiol (Oxf). 2012;206(4):242–50.
34. Sidhu SK, Bentley DJ, Carroll TJ. Locomotor exercise induces long-lasting impairments in the capacity of the human motor cortex to voluntarily activate knee extensor muscles.
J Appl Physiol (1985). 2009;106(2):556–65.
35. Sidhu SK, Cresswell AG, Carroll TJ. Motor cortex excitability does not increase during sustained cycling exercise to volitional exhaustion.
J Appl Physiol (1985). 2012;113(3):401–9.
36. Sidhu SK, Lauber B, Cresswell AG, Carroll TJ. Sustained cycling exercise increases intracortical inhibition.
Med Sci Sports Exerc. 2013;45(4):654–62.
37. Sidhu SK, Weavil JC, Venturelli M, et al. Spinal μ-opioid receptor-sensitive lower limb muscle afferents determine corticospinal responsiveness and promote central fatigue in upper limb muscle.
J Physiol. 2014;592(22):5011–24.
38. Thomas K, Goodall S, Stone M, Howatson G, St Clair Gibson A, Ansley L. Central and peripheral fatigue in male cyclists after 4-, 20-, and 40-km time trials.
Med Sci Sports Exerc. 2015;47(3):537–46.
39. Todd G, Taylor JL, Gandevia SC. Measurement of voluntary activation of fresh and fatigued human muscles using transcranial magnetic stimulation.
J Physiol. 2003;551(Pt 2):661–71.
40. Wasserman K, Whipp BJ, Koyl SN, Beaver WL. Anaerobic threshold and respiratory gas exchange during exercise.
J Appl Physiol. 1973;35(2):236–43.
41. Wasserman K, Whipp BJ, Davis JA. Respiratory physiology of exercise: metabolism, gas exchange, and ventilatory control.
Int Rev Physiol. 1981;23:149–211.
