Disparate kinetics of change in responses to electrical stimulation at the thoracic and lumbar level during fatiguing isometric knee extension
Abstract
The present study compared the fatigue-induced change of matched-amplitude thoracic evoked potential (TMEP) and lumbar evoked potential (LEP) following electrical stimulation. Ten participants performed a 3 × 3 min isometric knee extension contraction separated by 4 min of recovery at the level of EMG required to produce 50% maximal voluntary contraction (MVC) force at baseline. The TMEP and LEP were evoked during the ongoing contraction at baseline and every minute into the fatiguing protocol and during recovery. Both responses were also assessed during a transcranial magnetic stimulation (TMS) evoked silent period to elicit a TMS-TMEP and TMS-LEP to assess responses without the confounding influence of descending drive. The results displayed disparate kinetics of the TMS-TMEP and TMS-LEP throughout the fatiguing protocol. The TMS-TMEP was reduced at all time points during exercise (P < 0.001), whereas the TMS-LEP was reduced at 2 min into set 1 and 1 min into sets 2 and 3 (P ≤ 0.04). TMS-LEPs were higher than the TMS-TMEPs at most time points (P ≤ 0.04). No change was observed in the TMEP or LEP. When evoked during the silent period, the reduction in TMEP is greater than the LEP during fatiguing isometric exercise. The disparate kinetics of change suggest that differential mechanisms are responsible for evoked responses to thoracic and lumbar stimulation. More research is required to identify the mechanisms responsible for the TMEP and LEP before precise inferences can be made on what fatigue-induced changes in these variables reflect.
NEW & NOTEWORTHY Assessing spinal excitability using lumbar stimulation when measuring responses in lower limbs has been suggested as an alternative method that could circumvent the issues associated with thoracic stimulation. The present study compared responses to the two types of stimuli throughout a fatiguing protocol and demonstrated that lumbar evoked responses differ substantially from thoracic responses when measured in the absence of voluntary drive. These findings suggest that different mechanisms are responsible for evoked responses to thoracic and lumbar stimuli.
INTRODUCTION
The corticospinal pathway represents the primary conduit for voluntary movement in humans. This pathway is composed of the corticospinal tract, including the motor cortex and descending axons, and spinal motoneurons (2, 30). During fatiguing exercise, i.e., that which elicits impairments in the force generating capacity of muscle, the corticospinal pathway can be subject to perturbations (26). Classically, alterations within the corticospinal pathway are investigated using transcranial magnetic stimulation (TMS). At sufficient intensity, this noninvasive technique permits the quantitative assessment of corticospinal excitability through the motor evoked potential (MEP). However, as the MEP alone is unable to discern between cortical and spinal changes, there is a growing recognition that methodologies which permit the assessment of corticospinal excitability at a segmented level should be applied to more precisely delineate fatigue-induced effects on the corticospinal pathway (10, 13, 30).
Accordingly, stimulations of the descending spinal tracts (i.e., below the motor cortex) at the mastoid [cervicomedullary motor evoked potential (CMEP); Ref. 25] or thoracic [thoracic motor evoked potential (TMEP); Ref. 10] levels are deemed appropriate methods to allow the segmented assessment of corticospinal excitability. These measures are thought to provide the most direct assessment of spinal motoneuron excitability in response to descending input for a number of reasons (for details, see Ref.13). In particular, it has been shown that TMEP has a significant monosynaptic component (10). Unlike the H reflex, which has been used previously to discern the contribution of spinal motoneurons to the overall MEP response (16), electrical stimulation of the descending spinal tract has been shown to be devoid of presynaptic inhibition (7). Finally, studies have demonstrated that cervicomedullary (29) and thoracic stimuli (10) activate many of the same input axons and synapses as TMS over the motor cortex. With the use of these methods, studies have detected fatigue-induced alterations in spinal motoneuron excitability in response to isometric fatiguing in the upper limbs (14, 26) and lower limbs (4) and in response to locomotor exercise (21, 22, 31).
The assessment of CMEP and TMEP has permitted greater insight into the etiology of fatigued-induced alterations in neuromuscular function. However, mastoid stimulation (i.e., CMEP) is limited in its ability to provide information on spinal motoneuron behavior when responses are measured in the lower limbs due to difficulties in evoking discernible responses (27, 28). Thoracic stimulation (i.e., TMEP) is considered a more appropriate tool to examine fatigue-induced changes in lower limbs (10), yet there are various methodological challenges when using this method. Namely, it is difficult to evoke responses in all participants in the relaxed muscle (10, 27). Martin et al. (10) reported that even when responses were evoked at rest, they tended to be small (≤10% of the maximum muscle response). Furthermore, high stimulus intensities are required to evoke a discernible response during a muscle contraction, causing a high degree of participant discomfort. Moreover, stimulation of the descending tracts at the thoracic level also stimulates spinal motoneurons associated with control of upper limb and trunk musculature, with the current applied likely shared between the spinal motoneuron pool of multiple muscle groups (8, 23). Such methodological issues limit the applicability of the TMEP when experiments assessing responses in the lower limbs are performed.
A potential solution to mitigate the methodological challenges associated with TMEP stimulations is to deliver stimuli at the lumbar level, where there is a greater relative density of descending tracts projecting to lower limb motoneurons (19). In a recent study, Škarabot et al. (23) delivered electrical stimuli over the first lumbar spinous process to assess responses in the rectus femoris and tibialis anterior muscles. Using a relatively low intensity of stimulation, they reported responses of ~15% of the maximum compound muscle action potential (Mmax) when evoked at rest and ~50% Mmax when recorded during a 50% maximal voluntary contraction (MVC). Furthermore, when electrical stimulation was paired at the lumbar region with TMS over the motor cortex at appropriately timed intervals, occlusive or facilitatory interactions between the stimuli were demonstrated (23). These results suggest that stimulation at the lumbar level activates some of the same axons as TMS. Moreover, the study demonstrated an increase in the LEP with increasing contraction strength with no concurrent change in onset latency, suggesting that responses to lumbar stimulation (LS) are evoked transsynaptically and have a monosynaptic component. Accordingly, LS could provide an alternative paradigm for the assessment of the spinal excitability when assessing responses in the lower limb muscles.
The interaction between TMS and LS and the lack of change in latency during voluntary contraction displayed by Škarabot et al. (23) was similar to the findings of Martin et al. (10). These authors used electrical stimulation at the thoracic level when investigating responses in the rectus femoris and tibialis anterior and similarly displayed an occlusive or facilitative interaction between thoracic stimulation and TMS at appropriate interstimulus intervals. Thus, given that both of these studies suggest that the LEP and TMEP activate some of the same axons as TMS and are evoked with a monosynaptic component (10, 23), it seems important to investigate whether the behavior of the two responses is similar during interventions known to modulate motoneuronal excitability, such as fatiguing submaximal isometric contractions (4, 14). If the amplitude and temporal pattern of change in LEP and TMEP during fatigue were similar, it would provide further support for the suggestion that both measures represent monosynaptic responses that can be used to determine the spinal contribution to corticospinal excitability and that the LEP can be used as an alternative method to the TMEP to assess spinal motoneuron excitability in lower limbs. Accordingly, the aim of this study was to compare the TMEP and LEP during fatiguing submaximal isometric contractions of the knee extensors.
METHODS
Participants
Ten healthy male volunteers (age: 25 ± 3 yr; stature: 178 ± 7 cm; and mass: 76 ± 11 kg) participated in the study. All participants were free from neurological illness or musculoskeletal injury and had no contraindications to TMS. The study was approved by the institutional ethics committee and conformed to the Declaration of Helsinki. Written informed consent was obtained from each participant before the study taking place.
Experimental Design
Participants visited the laboratory on two separate occasions, including a familiarization visit followed by the experimental trial. The experimental protocol is depicted in Fig. 1. Participants performed a fatiguing isometric protocol consisting of 3 × 3 min isometric contractions at a constant level of EMG. Each set was separated by 4 min of recovery. At baseline, every minute during the fatiguing protocol, and after 3 min of the between-set recovery periods (i.e., R1 between sets 1 and 2, and R2 between sets 2 and 3), responses to both thoracic and lumbar electrical stimuli were recorded. Postexercise recovery of evoked responses was assessed at 2 (R3), 4 (R4), and 6 min (R5) following the last set. The thoracic and lumbar stimuli were delivered both during the ongoing constant EMG contraction (to evoke the TMEP and LEP, respectively) and during the TMS evoked silent period (TMS-TMEP and TMS-LEP). At baseline, the size of the TMEP and LEP were matched at ~50% of the maximum muscle compound action potential (Mmax), and the TMS-TMEP and TMS-LEP were matched at ~15% Mmax. Stimuli were delivered during the silent period to assess the responses to thoracic and lumbar stimulation without the confounding influence of ongoing descending drive, the level of which is known to alter motoneuron excitability (14). The matching in size of the TMEP and LEP and TMS-TMEP and TMS-LEP permitted the comparison of the behavior of these two responses during fatiguing submaximal contractions, since both sets of measurements would reflect the excitability of a similar population of motoneurons. Further details on the experimental procedures are provided below.
Fig. 1.Experimental protocol. At baseline, 5 sets of contractions were performed to a level of root mean square electromyography (EMG) required to generate 50% of maximum voluntary contraction (MVC). Each set of baseline measurements composed of thoracic and lumbar stimulation delivered during the ongoing contraction to evoke the thoracic (TMEP) and lumbar evoked potential (LEP), respectively, with the same stimuli delivered during the transcranial magnetic stimulation (TMS) evoked silent period to elicit the TMS-TMEP and TMS-LEP. The 5 measurements for each stimulus were averaged to provide the baseline measure. After every minute of the 3 × 3 min fatiguing protocol, 3 min into the between-set recovery periods (R1 and R2), and at 2, 4, and 6 min following set 3 for postexercise recovery measures (R3, R4, and R5, respectively), a single measurement of the TMEP, LEP, TMS-TMEP, and TMS-LEP was taken, along with the rate of perceived exertion (RPE). R1–R5, recovery periods 1–5.
Experimental Procedures
The experiment began with two maximal voluntary contractions (MVCs) to determine maximal force, with the peak value used for calculation of submaximal forces. Subsequently, participants performed a 5-s contraction at 50% MVC using visual feedback displayed on a computer monitor. The averaged vastus lateralis (VL) root mean square EMG (RMSEMG) from the 5-s contraction was then calculated. Afterwards, VL Mmax was measured using femoral nerve stimulation (described below). The appropriate intensities for the TMEP, LEP, TMS-TMEP, and TMS-LEP were then determined (described below). After the appropriate TMEP and LEP intensities were determined, participants completed a visual analogue scale to assess the level of discomfort associated with the TMEP and the LEP, in which they were asked to draw a vertical line on a 100-mm horizontal line in response to the question “how painful did the stimulation feel,” anchored with the descriptors “not at all” to “extremely.” Once all appropriate intensities had been determined, baseline measurements for the fatiguing protocol were performed. At baseline, five sets of measurements were taken, with the four stimuli (TMEP, LEP, TMS-TMEP, and TMS-LEP) delivered during each set. The order of the stimuli was randomized and counterbalanced. Each set lasted ~20 s, with 5 s between stimuli and 2 min between each set. The five measurements for each stimulus were averaged to provide the baseline measure. Two minutes following the last set, participants began the fatiguing protocol, which required them to sustain a level of EMG corresponding to 50% MVC (50% MVC-EMG) for 3 × 3 min, with 4 min of recovery between each set. A constant EMG was utilized to maintain a consistent level of motoneuron output and was thus deemed suitable to allow the comparison of the TMEP and LEP at the same level of motoneuron output. The intensity of the protocol was selected after pilot testing revealed that the TMS-TMEP evoked during lower intensity contractions (i.e., 25% MVC-EMG) was too small (< 10% Mmax) in most participants. Following every minute of the 3 × 3 min fatiguing protocol, 3 min into the between-set recovery periods (R1 and R2), and at 2, 4, and 6 min following set 3 for postexercise recovery measures (R3, R4, and R5, respectively), a single measurement of the TMEP, LEP, TMS-TMEP, and TMS-LEP was taken. Before each set of stimuli throughout the protocol, participants were asked to verbally report their rating of perceived effort (RPE) on a scale from 0 to 10 (1).
Force and EMG recordings.
Participants were seated in an isometric knee extension dynamometer (ARS dynamometry, SP2, Ltd., Ljubljana, Slovenia) with the hips and the right knee at 70° flexion (with 0° referring to the extended neutral position). Electrical activity from the VL was recorded using self-adhesive surface electrodes (Meditrace 100, Covidien, Mansfield, MA) using a bipolar electrode configuration, with electrodes placed 30 mm apart. A reference electrode was placed on the patella. The placement of the EMG electrodes was based on Surface ElectroMyoGraphy for the Non-Invasive Assessment of Muscles (SENIAM) guidelines (5). Before electrode placement, the skin was shaved, abraded and cleaned using isopropyl alcohol to limit impedance. The electrodes recorded electrical activity in the VL, with the signal processed to permit the analysis of RMS amplitude for submaximal contractions, Mmax from the electrical stimulation of the femoral nerve, and the responses to thoracic and lumbar spine stimulation and TMS. Signals were amplified with an octal bio-amplifier (ML138, ADInstruments, Bella Vista, Australia), bandpass filtered (5–500 Hz), and analogue-to-digitally converted at a sampling rate of 2 kHz by PowerLab System (16/30, ADInstruments). To assist participants in visualizing and maintaining the ongoing EMG contraction at the 50% MVC-EMG from baseline, the RMSEMG was smoothed using a 400-ms time constant.
Femoral nerve stimulation.
Motor nerve stimulation was used for the measurement of the Mmax amplitude and area. Single rectangular electrical pulses with 0.2-ms duration and 400-V maximal output voltage were delivered via a constant-current stimulator (DS7AH, Digitimer, Hertfordshire, UK) to the right femoral nerve via a 30-mm diameter surface cathode (Meditrace 100) taped to the skin onto the femoral triangle and a 50 × 90 mm anode (Durastick Plus, DJO Global, Vista, CA) in the gluteal fold. Electrical stimuli were first administered at 50 mA and then were increased in 50-mA increments until the maximum quadriceps twitch amplitude and Mmax were elicited. The resulting stimulation intensity was then increased by 30% to account for activity-induced changes in axonal excitability (570 ± 157 mA).
Transcranial magnetic stimulation.
Single-pulse TMS of a 1-ms duration was delivered to the left motor cortex via a concave double cone coil using a magnetic stimulator (Magstim 2002, The Magstim Co., Whitland, UK). The junction of the double cone coil was aligned tangentially to the sagittal plane, with its center 1–2 cm to the left of the vertex and was orientated to induce current in the posterior-to-anterior direction. Optimal TMS location was established as the position that elicited the largest MEP in the VL during a light voluntary contraction (20% MVC). The optimal position was marked on a swim cap secured to the scalp and used throughout the experiment. Subsequently, the TMS intensity was adjusted to evoke a 200-ms silent period at 50% MVC-EMG (63 ± 8% maximum stimulator output).
Thoracic stimulation.
A constant-current stimulator (DS7AH, Digitimer) was used to evoke TMEP amplitude and area by passing a single rectangular electrical pulse with a 0.2-ms duration and 400-V maximal output voltage between surface electrodes (Meditrace 100) positioned between the spinous processes of T1 and T2 (cathode) and ~5 cm below the center of the cathode between T5 and T6 (anode) (Fig. 2) (28). The thoracic stimulation intensity was adjusted to elicit a TMEP amplitude of 50% Mmax during the ongoing contraction at 50% MVC-EMG. This stimulation intensity was then used to measure TMS-TMEP, during which the thoracic stimulation was delivered 100 ms after the TMS triggered during the ongoing contraction. With the use of this stimulation intensity, the TMS-TMEP was 12 ± 15% Mmax amplitude.
Fig. 2.Illustration of electrode placement. Thoracic electrical stimulation was performed with the cathode placed over the spinous process between T1 and T2 and the anode between T5 and T6. Lumbar electrical stimulation was performed with the cathode placed over L1 and anode placed over the T8 spinous process. TMEP, thoracic evoked potential; LEP, lumbar evoked potential. [Adapted from Škarabot et al. (23) with permission from John Wiley and Sons.].
Lumbar stimulation.
The same constant current stimulator and pulse width as with the femoral nerve and thoracic stimuli was used to deliver lumbar spine stimulation (23). Figure 2 displays the electrode placement for lumbar stimulation. The cathode electrode (5 × 9 cm; Durastick Plus) was centered over the first lumbar spinous process (L1), with the long axis of the electrode aligned to the center of the vertebral column. The surface of the cathode covered two spinous processes above and below the center point (T11-L3). The bottom of the anode (Meditrace 100) was placed in the midline of the vertebral column, 5 cm above the upper edge of the cathode, corresponding to the level of the eight thoracic spinous process (T8). As with the TMEP, the stimulation intensity was adjusted to evoke a LEP amplitude of 50% Mmax during the 50% MVC-EMG. Response latency of LEP was continuously monitored, and paired LS at a 50-ms interstimulus interval was used to ensure ventral and dorsal roots, respectively, were not stimulated, as described elsewhere (23). For the TMS-LEP, the intensity (326 ± 109 mA) was adjusted to match the amplitude of the conditioned response with that of the TMS-TMEP, producing a response of 13 ± 10% Mmax. All participants tolerated the nonconditioned and conditioned lumbar stimulations well.
Data Analysis
The data processing was performed using standardized Matlab scripts. During the prebaseline measurements, the peak-to-peak amplitudes of the Mmax, TMEP, LEP, TMS-TMEP, and TMS-LEP were measured immediately by placing two cursors on the initial deflection from baseline to the second crossing of the horizontal axis. All of the evoked responses to spinal stimulation were subsequently normalized to Mmax amplitude to ensure they met the target amplitude. The RMSEMG and average force were calculated in the 500 ms before the last stimulation during each of the baseline sets, with values averaged across each set. The same epoch was used to assess RMSEMG and force throughout the fatiguing protocol, between-set and postexercise recovery periods. The RMSEMG was subsequently normalized to the maximum RMSEMG, obtained over a 500-ms epoch during the plateau in force during the MVC, which derived the peak value. For baseline measurements, each variable was averaged across the five sets. The areas of the evoked responses to femoral nerve and lumbar and thoracic stimulation were measured by marking the initial deflection from baseline to the second crossing of the horizontal axis (12), with areas of spinal responses normalized to the baseline Mmax area.
Statistical Analysis
SPSS version 20.0 (SPSS, Chicago, IL) was used of all statistical analyses. All data are presented as means ± SD. Statistical significance was set at an α level of 0.05. Normality of the data was assessed using the Shapiro-Wilks test. Assumptions of sphericity were explored and controlled for all variables using the Greenhouse-Geisser adjustment, where necessary. Analyses were performed to assess the kinetics of change of the TMEP compared with the LEP and the TMS-TMEP compared with the TMS-LEP during the fatiguing protocol and during the recovery periods. A two-way repeated measures ANOVA was performed, with the independent variables of stimulation type (i.e., TMEP vs. LEP and TMS-TMEP vs. TMS-LEP) and time (15 time points, including baseline, 9 throughout the fatiguing contractions, and 5 during between-set and postexercise recovery). Over the same time points, a one-way ANOVA assessed the change in force and EMG. In the event of a statistically significant interaction effect (stimulation × time), Bonferroni post hoc analysis was performed to locate where the differences lie. A paired sample t test was used to compare stimulation intensities required to elicit the TMEP and LEP. The Friedman test was used to assess changes in RPE throughout the protocol. The discomfort scale was analyzed using the Wilcoxon signed-rank test. Partial eta squared () was calculated to estimate effect sizes, with values representing small ( = 0.1), medium ( = 0.25), and large ( = 0.40) effects.
RESULTS
Force, EMG, and RPE
The force and EMG measured at baseline and throughout the fatiguing protocol and recovery periods are displayed in Fig. 3. One-way ANOVA showed a significant effect of time on force (F14,126 = 43.5, P < 0.001, = 0.82), which was reduced at all time points throughout the fatiguing protocol and remained below baseline during all recovery periods (all P < 0.001). Although a significant main effect of time was found for VL RMSEMG (F14,126 = 2.09, P = 0.02, = 0.19), post hoc tests displayed no differences in RMSEMG relative to baseline. The RPE increased throughout each set and was higher at minutes 2 and 3 compared with minute 1 for sets 1, 2, and 3 (all P < 0.001).
Fig. 3.Changes in force [%maximum voluntary contraction (%MVC), left y-axis] and root mean square (RMS) electromyography (EMG) (%maximum RMSEMG, right y-axis) of the vastus lateralis during the 3 × 3 min fatiguing protocol and recovery contractions. Gray area represents the fatiguing task. Data were analyzed via a one-way ANOVA to assess the change in force and EMG over time. *P < 0.05, significant differences from baseline for force. Bl, baseline; R1–R5, recovery periods 1–5.
TMEP Versus LEP
The intensity required to elicit the TMEP was higher than for the LEP (697 ± 148 vs. 413 ± 187 mA, respectively; P < 0.001). In addition, the level of perceived discomfort was reported to be significantly higher in response to thoracic (i.e., TMEP) versus lumbar (i.e., LEP) stimulation (4.7 ± 1.5 vs. 3.3 ± 2.0, respectively; P = 0.04). The Mmax measured at baseline was 15.7 ± 5.1 mV. While the average size of the evoked TMEP was close to the target response of 50% Mmax, there were considerable between subject variability (47 ± 19% Mmax; range 18–83% Mmax) due to difficulties in evoking large TMEPs in some participants and fluctuations from prebaseline to baseline measurements. However, we matched the amplitude of the LEP with the TMEP at baseline within reasonable limits, with differences in two responses at baseline of 7 ± 4% (range 1.4–12.5%). Figure 4A displays the results obtained for TMEP and LEP throughout the fatiguing protocol. No changes were reported for both TMEP and LEP amplitude during the protocol, with no effect of time (F14,126 = 1.36, P = 0.18, = 0.13) and no stimulation × time interaction during the fatiguing protocol (F14,126 = 1.52, P = 0.15, = 0.14). Similarly, there was no effect of time (F14,126 = 1.29, P = 0.22, = 0.13) and no stimulation × time interaction (F14,126 = 1.56, P = 0.1, = 0.15) for TMEP or LEP area. Table 1 displays raw values from TMEP and LEP amplitudes and areas. Individual responses for the TMEP and LEP are displayed in Fig. 4B, while raw traces from a representative participant are displayed in Fig. 5.
Fig. 4.Thoracic (TMEP) and lumbar evoked potential (LEP) amplitude measured in the vastus lateralis during ongoing contraction (A) and during the transcranial magnetic stimulation (TMS) evoked silent period (C) normalized to maximum compound muscle action potential (Mmax) (n = 10). Gray area represents the fatiguing task. For A and C, data were analyzed via a two-way repeated measures ANOVA (stimulation type × time). *P < 0.05, significant difference between TMS-TMEP and TMS-LEP; +P < 0.05, significant difference in TMS-TMEP relative to baseline; #P < 0.05, difference in TMS-LEP relative to baseline. Individual responses for the TMEP and LEP are displayed in B, and for the TMS-TMEP and TMS-LEP in D. Bl, baseline; R1–R5, recovery periods 1–5.
| BL | Set 1 | R1 | Set 2 | R2 | Set 3 | R3 | R4 | R5 | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 min | 2 min | 3 min | 1 min | 2 min | 3 min | 1 min | 2 min | 3 min | |||||||
| Evoked amplitudes, mV | |||||||||||||||
| TMEP | 7.3 ± 3.7 | 6.7 ± 4.8 | 6.2 ± 3.6 | 5.8 ± 3.8 | 6.9 ± 3.3 | 6.9 ± 4.5 | 4.7 ± 3.1 | 6.2 ± 3.5 | 5.9 ± 2.6 | 6.0 ± 2.8 | 4.6 ± 3.7 | 5.7 ± 3.4 | 6.3 ± 2.8 | 6.7 ± 3.1 | 6.7 ± 3.5 |
| LEP | 8.3 ± 4.5 | 7.4 ± 4.0 | 7.8 ± 3.8 | 8.5 ± 5.0 | 7.8 ± 4.2 | 8.9 ± 4.8 | 8.3 ± 4.9 | 7.8 ± 5.3 | 6.9 ± 4.6 | 7.9 ± 4.8 | 7.9 ± 6.1 | 8.5 ± 4.8 | 8.0 ± 5.4 | 7.4 ± 4.6 | 7.1 ± 4.6 |
| TMS-TMEP | 1.7 ± 1.8 | 0.3 ± 0.2 | 0.2 ± 0.0 | 0.1 ± 0.1 | 0.9 ± 1.6 | 0.2 ± 0.2 | 0.2 ± 0.1 | 0.2 ± 0.1 | 1.0 ± 1.3 | 0.3 ± 0.2 | 0.2 ± 0.3 | 0.2 ± 0.1 | 0.7 ± 0.7 | 1.2 ± 1.2 | 1.2 ± 1.1 |
| TMS-LEP | 1.9 ± 1.3 | 1.3 ± 1.1 | 1.1 ± 1.1 | 1.5 ± 1.4 | 1.5 ± 1.3 | 1.1 ± 0.9 | 1.3 ± 1.2 | 1.8 ± 1.7 | 1.5 ± 1.6 | 1.1 ± 0.8 | 1.5 ± 1.4 | 1.9 ± 2.0 | 1.7 ± 1.8 | 1.8 ± 1.8 | 1.6 ± 1.3 |
| Evoked areas, μV/s | |||||||||||||||
| TMEP | 44.8 ± 20.5 | 39.9 ± 21.2 | 40.9 ± 18.2 | 31.9 ± 17.7 | 41.1 ± 15.4 | 41.4 ± 19.5 | 32.4 ± 16.8 | 42.8 ± 21.1 | 38.8 ± 19.3 | 33.9 ± 11.4 | 26.2 ± 17.0 | 36.2 ± 20.6 | 41.0 ± 20.1 | 42.3 ± 22.0 | 42.0 ± 22.6 |
| LEP | 44.0 ± 22.6 | 41.5 ± 22.5 | 47.5 ± 22.6 | 49.8 ± 22.7 | 45.7 ± 22.7 | 47.3 ± 25.4 | 47.8 ± 28.9 | 42.9 ± 29.2 | 39.6 ± 24.2 | 40.9 ± 25.3 | 44.8 ± 33.1 | 49.9 ± 25.9 | 46.6 ± 31.3 | 41.9 ± 23.7 | 39.9 ± 24.8 |
| TMS-TMEP | 9.6 ± 8.4 | 1.5 ± 1.7 | 0.8 ± 0.4 | 0.8 ± 0.4 | 4.9 ± 7.8 | 1.3 ± 1.0 | 1.0 ± 0.8 | 1.0 ± 0.8 | 5.5 ± 7.0 | 1.4 ± 1.2 | 1.6 ± 1.5 | 1.1 ± 1.1 | 3.7 ± 3.7 | 7.0 ± 7.2 | 7.5 ± 6.4 |
| TMS-LEP | 10.9 ± 6.5 | 8.7 ± 6.6 | 8.0 ± 7.1 | 8.9 ± 7.9 | 8.3 ± 6.6 | 7.7 ± 6.7 | 7.6 ± 6.9 | 8.9 ± 7.3 | 7.7 ± 6.9 | 7.1 ± 6.3 | 8.6 ± 7.9 | 10.1 ± 10.0 | 8.4 ± 6.5 | 9.2 ± 6.3 | 9.8 ± 7.3 |
Fig. 5.Raw traces from a single participant across the experiment. The thoracic (TMEP) and lumbar evoked potential (LEP) (top 2 rows) were recorded during an ongoing contraction, while the transcranial magnetic stimulation (TMS)-TMEP and TMS-LEP (bottom 2 rows) were recorded during a TMS evoked silent period. The beginning of each trace began immediately following the stimulus artifact. Responses were measures in the vastus lateralis. R1–R5, recovery periods 1–5.
TMS-TMEP Versus TMS-LEP
Similar to the TMEP and LEP evoked during the ongoing contraction, there was considerable between subject variability in the TMS-TMEP at baseline (12 ± 15% Mmax; range 3.5–42% Mmax). However, the TMS-LEP was again matched within reasonable limits with the TMS-TMEP at baseline, with differences in the two responses at baseline of 5 ± 2% (range 2.5–7.1%). There were differences in the kinetics of change in the TMS-TMEP and TMS-LEP throughout the fatiguing protocol (Fig. 4C), with a significant stimulation × time interaction (F14,126 = 1.68, P = 0.04, = 0.16). The TMS-TMEP was reduced at all time points compared with baseline throughout the 3 × 3 min contraction (all P ≤ 0.02). During R1 and R3, the TMS-TMEP remained below baseline (P ≤ 0.05) and returned to baseline by R4 (P = 0.14). The TMS-LEP was reduced at 2 min into set 1 (P = 0.02) and 1 min into set 2 (P = 0.04) and set 3 (P = 0.01). No differences in TMS-LEP relative to baseline were found during any of the recovery periods (all P ≤ 0.87). A significant main effect of stimulation was found (F1,9 = 8.73, P = 0.02, = 0.28), with the TMS-TMEP being smaller than the TMS-LEP at all time points during the 3 × 3 min contraction (all P ≤ 0.04) apart from 1 min into set 3 (P = 0.06). During recovery, the TMS-TMEP was lower than the TMS-LEP at R3 (P = 0.04).
The results for TMS-TMEP and TMS-LEP area followed a similar trend as with amplitudes, with a significant stimulation × time interaction (F14,126 = 2.66, P < 0.01, = 0.23). The TMS-TMEP area was reduced at all time points during the 3 × 3 min contraction (all P ≤ 0.01) and R1, R2, and R3 (all P ≤ 0.01) before recovering by R4 (P = 0.10). The TMS-LEP area was reduced at 2 min into set 2 (P = 0.04) and 1 min into set 2 (P = 0.02) and was below baseline during R2 (P = 0.04). A significant main effect of stimulation was found (F1,9 = 8.56, P = 0.02, = 0.48), with the TMS-TMEP smaller than the TMS-LEP at all time points throughout the 3 × 3 min contraction (all P ≤ 0.03) and no differences during any recovery period (all P ≥ 0.06). Table 1 displays raw data for TMS-TMEP and TMS-LEP raw amplitudes and areas. Individual responses for the TMS-TMEP and TMS-LEP are displayed in Fig. 4D, with raw traces from a representative participant displayed in Fig. 5.
DISCUSSION
The aim of the present study was to compare the change in evoked responses measured in the knee extensors following thoracic and lumbar stimulation throughout a fatiguing isometric exercise protocol performed at a constant level of EMG. A key and novel finding from the study is that when the TMEP and LEP are evoked during the TMS-induced silent period, at which time they are devoid of the influence of descending drive on motoneuronal excitability, there is considerable discrepancy in the kinetics of change between the TMS-TMEP and TMS-LEP. Specifically, while the TMS-TMEP was virtually abolished throughout the entire fatiguing protocol, the TMS-LEP was only reduced at three time points, with the decrease in the TMS-LEP substantially lower than the TMS-TMEP. Consistent with previous results (4), no change in TMEP or LEP measured during ongoing descending drive was found. Finally, both the stimulus intensity required to elicit a response of 50% Mmax and the level of discomfort associated with the stimulus were significantly less for the LEP than the TMEP. The mechanisms behind the disparate kinetics of change of the TMS-TMEP and TMS-LEP are unclear but suggest that the two types of stimuli activate different neural structures within the spinal cord. The results highlight that more research is required to elucidate the precise mechanisms responsible for the LEP and TMEP when responses are measured in the lower limbs.
Disparate Kinetics in TMS-TMEP and TMS-LEP with Fatigue
During the 3 × 3 min contraction at 50% MVC-EMG, there was a rapid decline in the TMS-TMEP, which was virtually abolished following just 1 min of exercise and remained below baseline throughout the exercise protocol. The approach used in the present study, in which the contraction intensity was set at a percentage maximal EMG and spinal stimulation was delivered during the TMS evoked silent period, has been similarly utilized in previous studies assessing fatigue-induced changes in spinal motoneuron excitability in the upper (14, 15) and lower limbs (4). These studies similarly displayed that the TMS-TMEP (4) and TMS-CMEP (14, 15) are rapidly diminished with fatigue. For example, using a 10-min sustained contraction at 25% MVC-EMG, Finn et al. (4) found that the TMS-TMEP set to evoke a response of ~15% Mmax at baseline was reduced as soon as 1 min after exercise onset, and remained depressed throughout the entire contraction. Using a similar fatiguing protocol with TMS-CMEP recorded in the biceps brachii, McNeil et al. (14) displayed that the TMS-CMEP was reduced following 1.5 min of exercise and remained below baseline throughout. Thus the time course of the decline of the TMS-TMEP in the present studies corroborates previous findings.
In contrast to the TMS-TMEP, the TMS-LEP was only reduced at three time points throughout the protocol (i.e., at 2 min into set 1, and at 1 min into sets 2 and 3). The TMS-LEP responses were consequently higher than the TMS-TMEP during most of the protocol. Thus, despite the intensities of thoracic and lumbar stimuli being set to evoke responses of a similar size (~15% Mmax) at baseline, with the aim of activating a similar portion of the motoneuron pool, the two types of stimuli displayed disparate changes with fatigue. Based on the present results, a logical question that arises is which response (i.e., TMS-TMEP vs. TMS-LEP) provides the more valid measure spinal excitability with fatigue. Previous studies have displayed that electrical stimulation over both the thoracic spine (i.e., for TMEP) and lumbar spine (i.e., for LEP) activates corticospinal axons projecting to the knee extensors. Specifically, both Martin et al. (10) and Škarabot et al. (23) showed that when thoracic and lumbar stimulation, respectively, are paired with TMS at appropriately timed intervals, there is either occlusion or facilitation of responses compared with the MEP alone. Thus these studies provided evidence that thoracic and lumbar stimulation activate some of the same corticospinal axons projecting to the knee extensors as with TMS of the motor cortex. In addition, both studies showed an increase in the size of TMEP and LEP during increasing contraction intensities without a concurrent change in onset latency, suggesting a monosynaptic component of both responses. Given these similar results, the disparate kinetics of change of the TMS-TMEP and TMS-LEP in the present study is surprising, and we cannot conclusively deduce which of the TMS-TMEP or TMS-LEP is more accurate in investigating spinal motoneuron excitability during fatigue. One factor lending indirect support for the TMS-TMEP providing a more valid measure of spinal motoneuron excitability is its virtual abolishment with fatigue, which is consistent with findings from studies using the TMS-CMEP during fatiguing isometric contraction, although these studies were conducted in upper rather than lower limb muscles (14, 15). In turn, there is strong evidence that the CMEP is primarily the result of motoneuron activation by a single descending volley elicited by excitation of corticospinal axons (13, 18, 28), confirmed through epidural recordings (18). Nevertheless, given that such assessments have not taken place for the TMEP or LEP, we cannot conclude that either measure is more valid than the other.
The factors contributing to the attenuated decreases in TMS-LEP during fatigue are unclear. Although both thoracic and lumbar stimulations are likely to stimulate descending pathways other than the corticospinal tract (13), there is a possibility that lumbar stimulation could activate other neural elements at the lumbar region, thereby confounding the assessment of spinal motoneuron excitability when using the TMS-LEP. While the results from the study by Škarabot et al. (23) suggest that the LEP has a monosynaptic component, the authors acknowledged the possibility that stimulation at the lumbar level could activate other elements of the spinal cord with projections onto motoneurons. Indeed, the spinal cord consists of segmental excitatory interneurons with monosynaptic projections onto motoneurons, and activation of these projections through spinal stimulation would confound any measure of spinal motoneuron excitability. At the lumbar level, there exists a prominent system of short-axon propriospinal premotoneurons, which transmit part of the descending command to lower limb motoneurons (9, 17). While propriospinal interneurons also exist at the thoracic level, these are thought to primarily play a role in control of respiratory and trunk muscles, rather than lower limb muscle groups (20). It is plausible that lumbar stimulation could activate lumbar propriospinal premotoneurons, given that the site of greatest spinal cord activation with lumbar stimulation is thought to occur between L1 and L5 (23) and that dorsolateral and ventromedial lumbar propriospinal neurons with excitatory projections to motoneurons are located between L2 and L5 (3). These neurons receive descending input from the corticospinal tract and extra-pyramidal tracts (e.g., rubrospinal, tectospinal and reticulospinal tracts), as well as strong peripheral afferent input (17). These multiple inputs would increase the potential for “contamination” of the LEP if propriospinal neurons were to be activated by lumbar stimulation. Moreover, this could partially explain why a LEP of a similar size to the TMEP can be obtained at a considerably lower stimulus intensity and thereby cause a lower degree of discomfort. While the behavior of lumbar propriospinal neurons during fatiguing exercise is unknown, it is interesting to note that one study examining the cervical propriospinal system during fatigue suggested that the proportion of descending drive relayed through propriospinal neurons is increased with fatigue, possibly to maintain force production by compensating for modifications elsewhere in the nervous systems (11). The increase in drive through propriospinal neurons concurrently increases the gain of this pathway (11). Although speculative, if the same holds true in the lumbar propriospinal system, and this system was indeed activated by lumbar stimulation, this could explain the attenuated decrease in TMS-LEP and the kinetics of change in the TMS-LEP with fatigue, in which the TMS-LEP was reduced during the early/middle stages of each set but was not different from baseline during the later stages. Nevertheless, whether the lumbar propriospinal system is activated, and the behavior of this system during fatigue, are unclear. It is feasible that the attenuated reduction in the TMS-LEP was a result of activation of other neural elements with projections onto motoneurons, and further work is required to elucidate on the precise neural elements which contribute to the LEP.
No Change in TMEP or LEP Measured During Ongoing Voluntary Contraction
In previous studies in which force has been held constant, there is a considerable increase in the level of EMG and a concomitant increase in CMEP amplitude due to increases in motoneuron recruitment and discharge rate (24) to compensate for peripheral impairments in contractile function. In the present study, a contraction intensity set at a percentage of maximal EMG was preferred to a constant force contraction to maintain a consistent level of motoneuron output, providing a more suitable approach for the comparison of the TMEP and LEP. During the 50% MVC-EMG contraction, there were no changes in RMSEMG relative to baseline and force was significantly reduced as expected. When thoracic and lumbar stimulation were delivered during the ongoing contraction throughout the fatiguing protocol and recovery periods, there was no change in either the unconditioned TMEP or LEP. This finding corroborates that of Finn et al. (4), who showed no change in the TMEP despite a substantial reduction in the TMS-TMEP during fatiguing isometric knee extension. The lack of change in TMEP or LEP is not wholly unexpected. Indeed, an increase in descending drive is necessary to compensate for the decline in motoneuron excitability when attempting to maintain a given level of EMG, and higher levels of descending drive are known to increase motoneuron excitability (14). The increased level of descending drive is further supported by the increase in the RPE during each set, which was likely due in part to the increase efferent commands giving rise to corollary discharge and subsequent activation of sensory areas within the brain (6). This finding further highlights the importance of measuring motoneuron excitability without ongoing descending drive to improve interpretation of segmental changes within the neuromuscular pathway during fatigue.
Conclusion
The results from the present study display different kinetics of change of the TMEP and LEP when evoked during the TMS-induced silent period throughout fatiguing isometric knee extension exercise. The TMS-TMEP was virtually abolished following 1 min of exercise and remained depressed throughout all exercise, whereas the TMS-LEP was sporadically reduced throughout the protocol and to a lesser extent than the TMS-TMEP. The mechanisms behind the disparate kinetics of change of the two responses are unclear but suggest that thoracic and lumbar stimuli activate different neural structures to evoke responses in the knee extensors. The results of the study highlight that further research is required to elucidate on the precise mechanisms responsible for the TMEP and LEP.
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by the authors.
AUTHOR CONTRIBUTIONS
C.G.B., R.S., T.L., and G.Y.M. conceived and designed research; C.G.B., R.S., N.R., and B.S. performed experiments; C.G.B., R.S., and N.R. analyzed data; C.G.B., R.S., T.L., and G.Y.M. interpreted results of experiments; C.G.B. prepared figures; C.G.B. and R.S. drafted manuscript; R.S., N.R., B.S., T.L., and G.Y.M. edited and revised manuscript; C.G.B., R.S., N.R., B.S., T.L., and G.Y.M. approved final version of manuscript.
REFERENCES
- 1. . Psychophysical bases of perceived exertion. Med Sci Sports Exerc 14: 377–381, 1982. doi:10.1249/00005768-198205000-00012.
Crossref | PubMed | Web of Science | Google Scholar - 2. . Corticospinal projections to lower limb motoneurons in man. Exp Brain Res 89: 649–654, 1992. doi:10.1007/BF00229889.
Crossref | PubMed | Web of Science | Google Scholar - 3. . Spinal control of locomotion: individual neurons, their circuits and functions. Front Physiol 9: 784, 2018. doi:10.3389/fphys.2018.00784.
Crossref | PubMed | Web of Science | Google Scholar - 4. . Motoneuron excitability of the quadriceps decreases during a fatiguing submaximal isometric contraction. J Appl Physiol (1985) 124: 970–979, 2018. doi:10.1152/japplphysiol.00739.2017.
Link | Web of Science | Google Scholar - 5. . Development of recommendations for SEMG sensors and sensor placement procedures. J Electromyogr Kinesiol 10: 361–374, 2000. doi:10.1016/S1050-6411(00)00027-4.
Crossref | PubMed | Web of Science | Google Scholar - 6. . The ‘sensory tolerance limit’: a hypothetical construct determining exercise performance? Eur J Sport Sci 18: 13–24, 2018. doi:10.1080/17461391.2016.1252428.
Crossref | PubMed | Web of Science | Google Scholar - 7. . Tests for presynaptic modulation of corticospinal terminals from peripheral afferents and pyramidal tract in the macaque. J Physiol 573: 107–120, 2006. doi:10.1113/jphysiol.2005.100537.
Crossref | PubMed | Web of Science | Google Scholar - 8. . Muscles: Testing and function with Posture and Pain. Philadelphia, PA: Lippincott Williams & Wilkins, 2005.
Google Scholar - 9. . Cortical control of spinal pathways mediating group II excitation to human thigh motoneurones. J Physiol 517: 301–313, 1999. doi:10.1111/j.1469-7793.1999.0301z.x.
Crossref | PubMed | Web of Science | Google Scholar - 10. . Noninvasive stimulation of human corticospinal axons innervating leg muscles. J Neurophysiol 100: 1080–1086, 2008. doi:10.1152/jn.90380.2008.
Link | Web of Science | Google Scholar - 11. . Muscle fatigue changes cutaneous suppression of propriospinal drive to human upper limb muscles. J Physiol 580: 211–223, 2007. doi:10.1113/jphysiol.2006.125997.
Crossref | PubMed | Web of Science | Google Scholar - 12. . Output of human motoneuron pools to corticospinal inputs during voluntary contractions. J Neurophysiol 95: 3512–3518, 2006. doi:10.1152/jn.01230.2005.
Link | Web of Science | Google Scholar - 13. . Testing the excitability of human motoneurons. Front Hum Neurosci 7: 152, 2013. doi:10.3389/fnhum.2013.00152.
Crossref | PubMed | Web of Science | Google Scholar - 14. . Behaviour of the motoneurone pool in a fatiguing submaximal contraction. J Physiol 589: 3533–3544, 2011. doi:10.1113/jphysiol.2011.207191.
Crossref | PubMed | Web of Science | Google Scholar - 15. . The response to paired motor cortical stimuli is abolished at a spinal level during human muscle fatigue. J Physiol 587: 5601–5612, 2009. doi:10.1113/jphysiol.2009.180968.
Crossref | PubMed | Web of Science | Google Scholar - 16. . On the comparability of H-reflexes and MEPs. Electroencephalogr Clin Neurophysiol Suppl 51: 93–101, 1999.
PubMed | Google Scholar - 17. . The Circuitry of the Human Spinal Cord: Its Role in Motor Control and Movement Disorders. Cambridge, UK: Cambridge University Press, 2005.
Crossref | Google Scholar - 18. . Transcranial electrical stimulation of the motor cortex in man: further evidence for the site of activation. J Physiol 481: 243–250, 1994. doi:10.1113/jphysiol.1994.sp020435.
Crossref | PubMed | Web of Science | Google Scholar - 19. . Spinal segment-specific transcutaneous stimulation differentially shapes activation pattern among motor pools in humans. J Appl Physiol (1985) 118: 1364–1374, 2015. doi:10.1152/japplphysiol.01128.2014.
Link | Web of Science | Google Scholar - 20. . Electrophysiological and morphological characterization of propriospinal interneurons in the thoracic spinal cord. J Neurophysiol 105: 806–826, 2011. doi:10.1152/jn.00738.2010.
Link | Web of Science | Google Scholar - 21. . Group III/IV locomotor muscle afferents alter motor cortical and corticospinal excitability and promote central fatigue during cycling exercise. Clin Neurophysiol 128: 44–55, 2017. doi:10.1016/j.clinph.2016.10.008.
Crossref | PubMed | Web of Science | Google Scholar - 22. . Fatigue-related group III/IV muscle afferent feedback facilitates intracortical inhibition during locomotor exercise. J Physiol 596: 4789–4801, 2018. doi:10.1113/JP276460.
Crossref | PubMed | Web of Science | Google Scholar - 23. . Electrical stimulation of human corticospinal axons at the level of the lumbar spinal segments. Eur J Neurosci 49: 1254–1267, 2019. doi:10.1111/ejn.14321.
Crossref | PubMed | Web of Science | Google Scholar - 24. . Sustained contraction at very low forces produces prominent supraspinal fatigue in human elbow flexor muscles. J Appl Physiol (1985) 103: 560–568, 2007. doi:10.1152/japplphysiol.00220.2007.
Link | Web of Science | Google Scholar - 25. . Stimulation at the cervicomedullary junction in human subjects. J Electromyogr Kinesiol 16: 215–223, 2006. doi:10.1016/j.jelekin.2005.07.001.
Crossref | PubMed | Web of Science | Google Scholar - 26. . Noninvasive stimulation of the human corticospinal tract. J Appl Physiol (1985) 96: 1496–1503, 2004. doi:10.1152/japplphysiol.01116.2003.
Link | Web of Science | Google Scholar - 27. . Electrical stimulation of the human descending motor tracts at several levels. Can J Neurol Sci 22: 36–42, 1995. doi:10.1017/S0317167100040476.
Crossref | PubMed | Web of Science | Google Scholar - 28. . Percutaneous electrical stimulation of corticospinal pathways at the level of the pyramidal decussation in humans. Ann Neurol 29: 418–427, 1991. doi:10.1002/ana.410290413.
Crossref | PubMed | Web of Science | Google Scholar - 29. . Magnetic stimulation of the descending and ascending tracts at the foramen magnum level. Electroencephalogr Clin Neurophysiol 105: 128–131, 1997. doi:10.1016/S0924-980X(97)96141-5.
Crossref | PubMed | Google Scholar - 30. . Corticospinal excitability during fatiguing whole body exercise. Prog Brain Res 240: 219–246, 2018. doi:10.1016/bs.pbr.2018.07.011.
Crossref | PubMed | Web of Science | Google Scholar - 31. . Fatigue diminishes motoneuronal excitability during cycling exercise. J Neurophysiol 116: 1743–1751, 2016. doi:10.1152/jn.00300.2016.
Link | Web of Science | Google Scholar
