The influence of contraction amplitude and firing history on spike-triggered averaged trapezius motor unit potentials
1 Department of Industrial Economics and Technology Management, Norwegian University of Science and Technology, N-7491 Trondheim, Norway
Abstract
The spike-triggered averaged (STA) technique was used to examine trapezius motor unit potentials and their dependence on contraction amplitude and firing history. Individual motor unit firings were identified by a fine-wire intramuscular electrode, while STA-derived potentials were extracted from the simultaneously recorded surface electromyographic (SEMG) signal. Amplitude-controlled contractions and contractions with typing tasks and mental stress were carried out. STA potentials were mostly derived from 20 s intervals of firing. Motor unit synchrony was estimated by peristimulus time histograms (PSTHs). An association between SEMG amplitude and STA-derived motor unit potentials was found: motor unit area showed a four-fold increase when SEMG amplitude increased from 1.5 to 10.5% of the root mean square-detected SEMG signal at maximal voluntary contraction (%EMGmax). Low- and higher threshold motor unit potentials, all with recruitment thresholds <10% EMGmax, had similar area at the same contraction amplitude. A significant increase in the STA-derived potentials was observed after 3 min of constant-amplitude contractions; however, this difference was reduced after 10 min and no longer present after 30 min of contraction. Motor unit synchrony accounted for, on average, 2.8% additional firings within 2 ms of the triggering motor unit. We conclude that the increase in STA-derived potentials with contraction amplitude is, to a major extent, due to motor unit synchrony, limiting the applicability of this method in postural muscles presenting wide motor unit potentials. The similar area of motor units at same SEMG amplitude may indicate that trapezius motor units recruited below 10% EMGmax are of similar size and thus not organized according to the Henneman size principle.
, 百拇医药
Introduction
Most studies of motor activity in postural muscles use surface (SEMG) or multiunit intramuscular electromyographic recordings, while it is tacitly assumed that motor units are organized and controlled in a similar way to extremity muscles (e.g. De Luca et al. 1982). One of the governing principles of motor control is the Henneman size principle (Henneman et al. 1965); motor units become progressively larger and develop more force with increasing recruitment threshold. However, the data to support this principle all derive from experimentation on extremity muscles. The force requirement of postural muscles, such as sustained force development over long time periods, is quite different from that of limb muscles. It is therefore pertinent to query whether the size principle is applicable to postural muscles.
, 百拇医药
Trapezius motor units can be continuously active for up to one hour (Thorn et al. 2002). Whole-day SEMG recordings have shown that trapezius has a more sustained activity pattern than upper and lower limb muscles (Mork and Westgaard, unpublished observations; Kern et al. 2001). Trapezius motor unit firing pattern deviates from motor units of a distal hand muscle (first dorsal interosseus; FDI) in slow ramp contractions, by sudden onset and higher firing rates of new relative to previously recruited motor units (Westgaard & De Luca, 2001). Abrupt changes in recruitment threshold during sustained contractions have been observed, causing ‘silent’ periods without motor unit firing, and ‘substitution’ by other motor units with initially higher threshold (Westgaard & De Luca, 1999; Westad et al. 2003). Altogether, this shows that the control of low-threshold trapezius motor units has characteristics that are different from those of limb muscles.
, http://www.100md.com
As it is not possible to determine motor unit force in a postural muscle like trapezius with the available technology, we wanted to examine whether motor unit size could be deduced from SEMG recordings. The representation of single motor units in the SEMG signal was determined by spike triggered averaging (STA). This was anticipated to generate relatively consistent motor unit potentials, as trapezius has a sheet-like appearance with mean thickness 5 mm at the preferred SEMG electrode location (range 3–9 mm; Jensen et al. 1994). A muscle thickness of 5 mm corresponds well to the typical extent of motor unit territory (Buchthal et al. 1957; Roeleveld et al. 1997).
, 百拇医药
The first aim of this study was to examine the hypothesis that low-threshold trapezius motor units show high consistency in their contribution to the SEMG signal, but with an increase in amplitude (i.e. size) of motor unit potentials with higher recruitment threshold as formulated in the Henneman size principle. However, it appeared the STA-derived potentials of the same motor unit were highly dependent on SEMG amplitude. A second aim was therefore to examine the cause of the increase in the area of STA-derived motor unit potentials with contraction amplitude. A possible explanation is short-term synchrony in motor unit firing, which causes an increase in the apparent SEMG amplitude of STA-derived motor unit potentials (Kirkwood, 1979; Yao et al. 2000; Taylor et al. 2002). Motor unit synchrony was examined by constructing peristimulus time histograms (PSTHs) of firing events for pairs of motor units.
, 百拇医药
Methods
Material primarily collected for other study purposes was analysed (Westad et al. 2003; Westad et al. 2004). This included trapezius electromyographic recordings from 16 healthy subjects, six males and ten females, age ranging from 20 to 57 years. Experimental procedures and methods are described in detail in the previous papers; here the description of methods and procedures is limited to those aspects relevant to the present study. The experiments were carried out according to the Declaration of Helsinki. Each subject read and signed an informed consent form approved by the local Institutional Review Board prior to participating in the study.
, http://www.100md.com
Experimental procedures
Simultaneous electromyographic (EMG) recordings by surface and intramuscular electrodes were performed. The indwelling EMG signal was used to detect firing events of identified trapezius motor units while STA-derived potentials were extracted from the surface EMG (SEMG) signal. The trapezius SEMG signal was also root-mean-square (RMS) detected and used to indicate and, in some procedures, to control trapezius contraction amplitude by displaying signal amplitude on a visual display screen in front of the subject. The SEMG signal was calibrated in technical units (μV) and as a percentage of the RMS-detected EMG activity at maximal voluntary contraction (%EMGmax).
, 百拇医药
The feedback-controlled contractions consisted of slow ramp contractions (nominally from 0 to 10% EMGmax in 3 min, rise time 0.055% EMGmax s–1; some contractions reached lower amplitudes) and constant amplitude contractions of one to 30 min duration, amplitudes ranging from 1 to 8% EMGmax. All feedback-controlled contractions were carried out with the subject seated and straps placed over the shoulders to provide resistance to the attempted movement of elevating the shoulders. Shoulder elevation was performed bilaterally, with EMG data collected from the left trapezius. Three 30-min and four 10-min contractions with imposed brief periods of transient increase in SEMG activity (Westad et al. 2003) and 10 contractions of 10 min duration without transient increase in amplitude (Westad et al. 2004) provided long-duration, constant-amplitude recordings. These contractions were used to examine the stability of STA-derived potentials over time.
, 百拇医药
Motor tasks mimicking muscle activation in daily living, such as motor activity in typing and trapezius motor response to mental stress were carried out (Westad et al. 2004). These tasks were of 10 min duration and were performed without shoulder straps or any other device (except EMG electrodes) that could influence shoulder geometry or shoulder movement. Contraction amplitude varied within a range of 2–4% EMGmax during these contractions, which had mean amplitudes from <1–9% EMGmax. The mental strain task consisted of a complex, attention-demanding two-choice reaction test presented on the computer screen. An open (‘frame’) and a solid (‘brick’) quadrangle were placed in a square pattern, and an alphanumeric suggestion on how to move the brick to superimpose on the frame was given (Westgaard & Bjrklund, 1987). The subject responded by pressing one of two keys, ‘correct’ or ‘wrong’, with the right or left index finger. A new position of the ‘brick’ and ‘frame’ in the square pattern and a new suggestion then appeared. The execution of the test was self-paced, but the subject first carried out the test for 2 min at a steady pace while attempting to maintain a low failure rate. On the basis of this performance the subject was offered a small monetary reward to perform 10% faster with the same or lower failure rate for 10 min. Feedback provided on the computer screen informed the subjects about the response speed (very slow, slow, OK, fast, or very fast) and whether the answer was correct (Wrsted et al. 1994).
, http://www.100md.com
Physiological recordings and analyses
The SEMG signal was detected by an active differential electrode with two circular recording surfaces (6 mm in diameter, 20 mm interelectrode distance). The electrode was positioned with the medial recording surface 20 mm lateral to the midpoint of a line between the C7 spinous process and the acromion (Jensen et al. 1993). The SEMG signal was band-pass filtered at 10–1000 Hz. The RMS-detected SEMG signal was averaged at a time resolution of 0.2 s. The intramuscular EMG signal was recorded with quadrifilar wire electrodes, constructed by bonding together four 50-μm nylon-coated nickel–chrome alloy wires (‘Stablohm 800A’, California Fine Wire Co, Grover Beach, CA, USA). The wire bundle was cut transversely, exposing only the cross-section of the wires. The wire bundle was placed in a 27-gauge needle and a hook was formed at approximately one millimetre from the exposed end of the wire. The needle was inserted to a depth of approximately 10 mm at a location approximately 10 mm medial to the midpoint of a line between the C7 spinous process and the acromion, along the direction of the muscle fibres recorded by the SEMG electrode. The insertion aimed for the midpoint of the transverse extent of the SEMG electrode pick-up area. The needle was removed and the wire bundle remained lodged in the muscle. Three pairs were chosen as the differential input to the amplifiers. The signals were band-pass filtered from 1 to 10 kHz. All EMG signals were stored on a digital recorder (DATaRec-A160, Racal-Heim Systems GmbH, Bergisch Gladbach, Germany). The signals were subsequently reconverted to an analog form and digitized at a sampling rate of 50 kHz on a PC.
, 百拇医药
The intramuscular EMG signals were resolved into individual motor unit firing trains using the Precision Decomposition technique (LeFever & De Luca, 1982; De Luca & Adam, 1999). This technique uses template matching, template updating, firing probabilities and superposition resolution to identify the individual firing times of the motor units (Mambrito & De Luca, 1984). The firing rates of the motor units were obtained by inverting the time series of the interpulse intervals. Firing rates were low-pass filtered at 0.5 Hz for display purposes, but STA motor unit potentials and estimates of motor unit synchrony were derived from the unfiltered firing rates.
, http://www.100md.com
Data analysis
Firing events of identified motor units were used as triggers in the STA analyses, which were based on recording intervals of 20 s, corresponding to 180–250 firings, unless otherwise stated. The intramuscular electrode was located medial and the SEMG electrode lateral to the endplate region (Jensen et al. 1993), making the STA potential sometimes appear to start before the firing event. The size of the STA potential was quantified as the area underneath the curve traced by the rectified STA potential. All derivations of STA area for a continuous recorded motor unit used the same start and end points. In long-duration, constant-amplitude contractions, STA potentials were extracted at the start of the experiment and 3, 10 and 30 min into the recording. Additional epochs were used to examine STA potentials at different SEMG amplitudes or at marked changes in firing patterns, e.g. STA potential before and after motor unit silent periods.
, 百拇医药
Motor unit synchronization was determined by constructing peristimulus time histograms between simultaneously firing motor units (PSTHs; e.g. Kirkwood et al. 1982). The CUSUM method (Ellaway, 1978) was used to determine the number of extra triggering events synchronous with the reference motor unit and the width of the central peak in the PSTH. CUSUM values were normalized by number of firings of the motor unit with the lowest firing rate.
Synthesized SEMG responses from STA-derived motor unit potentials
, 百拇医药
Ten STA-derived motor unit potentials were extracted at low SEMG amplitude (range 0.6–1.6% EMGmax). The rectified area of the motor unit potentials differed by a factor of 2 (from 0.017 to 0.037% EMGmax s). Trains of firings at 10 and 12.5 pulses per second (pps), corresponding to mean firing rates in constant-amplitude contractions of approximately 1 and 5% EMGmax (Westad et al. 2004), were constructed. The start of the individual trains of firings was decided by random draw within the intervals 0–100 and 0–80 ms for motor unit trains at 10 and 12.5 pps, respectively. The SEMG response for the set of motor unit firings was determined by algebraic summation of the motor unit trains (Day & Hulliger, 2001) and average rectified value (ARV) over 0.2 s intervals determined. The simulation of the SEMG signal was repeated, using a STA-extracted motor unit potential from the first dorsal interosseus (FDI), recorded by an electrode with bar-shaped recording surfaces that had a separation of 10 mm (Westgaard & De Luca, 2001). Ten trains of firing were constructed, with the motor unit potentials scaled to mimic the variation in area of the trapezius motor unit potentials.
, 百拇医药
Statistics
Pearson correlation coefficients were determined for regression of STA-derived motor unit area versus SEMG amplitude. A Student t test was used to compare STA potentials obtained with and without the use of shoulder straps. Wilcoxon signed ranks test was used to compare STA areas extracted at different times during sustained, constant-amplitude contractions to initial STA area. A P-value of 0.05, two-tailed, was considered to indicate significant differences.
, 百拇医药
Results
Figure 1 shows trapezius SEMG activity and firing rates of three motor units (1–3) recruited at different amplitudes (i.e. recruitment thresholds) of a slow ramp contraction increasing from 0 to 5% EMGmax in 180 s. Surface representations of the motor unit potentials were extracted by STA at intervals marked by horizontal bars on top of the time plot of SEMG activity in the upper left panel. The STA-derived potentials for each motor unit are shown as superimposed waveforms to the right of the plots of motor unit firing rates. The top right scatter plot shows rectified STA area as a function of SEMG amplitude. There was a marked increase in amplitude and area of the STA potentials with increasing SEMG amplitude. A near linear (r = 0.99), three-fold increase in STA area (from 0.22 to 0.66 μV s) was observed for the lowest threshold motor unit for SEMG amplitude increasing from 1.4 to 5.0% EMGmax. Although the higher-threshold motor units presented larger STA potentials than the low-threshold motor unit at recruitment of the respective motor units, the STA areas were closely similar at same contraction amplitude. A widening of the STA potential of the lowest threshold motor unit, in the order of 4 ms, from low to high SEMG amplitude was observed.
, 百拇医药
Three motor units with different recruitment thresholds were detected during the contraction (numbered 1–3); time plots of firing rates are shown below the SEMG recording. Spike-triggered averaged (STA) motor unit potentials were derived at intervals indicated by horizontal bars above the SEMG recording (20 s duration, >200 firings). Superimposed plots of the STA-derived potentials are shown to the right of each firing rate plot. STA area is determined as the area underneath the curve between vertical broken lines. The top right scatter plot shows STA area versus SEMG amplitude (1: , 2: , 3: ).
, http://www.100md.com
STA-derived motor unit potentials in other slow ramp contractions behaved as illustrated in Fig. 1. STA area always increased with contraction amplitude and in most instances there was a widening of the STA potential similar to that shown in Fig. 1. The width of STA-derived potentials at low contraction amplitude was relatively constant: mean width of the biphasic wave for 10 motor units was 22 ms (range 17–29 ms), with SEMG amplitude ranging from 0.6 to 1.6% EMGmax. Figure 2A shows regression lines of STA area versus SEMG amplitude for 15 low-threshold motor units in slow ramp contractions. The regression lines are based on four or five derivations of the STA potential. The extent of the lines delineates the range of STA derivations. Regression coefficients were near unity (mean value 0.96, range 0.84–1.0). The regression lines are re-plotted in Fig. 2B with STA amplitude calibrated in% EMGmax; intersubject variability is presumably reduced because the re-calibration compensates for intersubject variation in signal transmission to the surface electrode. Finally, regression of STA area (% EMGmax s) versus SEMG amplitude was established for all STA areas derived in these experiments (470 values from 163 motor unit recordings; Fig. 2C). Separate regressions were determined for STA potentials obtained in shoulder elevation (with straps providing resistance to the movement) and in trials without contact pressure to the shoulders. STA potentials were sorted by SEMG amplitude in intervals spanning 2% EMGmax. Mean and S.D. of STA area were determined. The increase in STA area with SEMG amplitude was nearly linear, with low S.D. especially at low SEMG amplitude. STA area was reduced in contractions with shoulder straps, relative to contractions without straps for SEMG amplitude >8% EMGmax (P < 0.01; STA areas normalized to 10% EMGmax by regression). There was no difference between the two conditions at lower SEMG amplitudes.
, http://www.100md.com
A, regression lines of STA area (μV s) versus contraction amplitude (%EMGmax) in slow ramp contractions. The extent of the regression lines indicates the range of intervals used to extract STA potentials. B, regression lines in A are re-plotted with STA area calibrated as (%EMGmax s). C, scatter plot of STA area versus SEMG amplitude of all STA-derived motor unit potentials (470 potentials from 163 recordings). Two groups are distinguished: potentials derived in experiments with shoulder straps providing resistance to the attempted elevation of the shoulders (), and contractions without restrictions on shoulder movement (). Mean value and S.D. of STA area (%EMGmax s) are shown for STA potentials within SEMG amplitude intervals of 2% EMGmax. Horizontal position shows mean value of SEMG amplitude for the different samples (n = 14–99).
, 百拇医药
A four-fold increase in STA area was observed in trials without the use of straps, from 0.03 to 0.12% EMGmax s, for an increase in SEMG amplitude from 1.5 to 10.5% EMGmax. Extrapolation of the regression lines to zero SEMG amplitude indicates STA areas of 0.014 (no straps) and 0.016% EMGmax s (straps) for low-threshold motor units firing in isolation. Five motor units recorded at SEMG amplitude <1% EMGmax had a mean area of 0.023%EMGmax·s.
STA potentials of motor units with high and low recruitment threshold were compared by quantifying STA area of higher threshold motor units in percentage of the simultaneously determined STA area of the lowest threshold motor unit. Table 1 shows the results for lowest-threshold motor units recruited in the intervals 0–2 and 2–4% EMGmax. There was no systematic change in STA area with recruitment threshold for motor units with thresholds <10% EMGmax.
, 百拇医药
The stability of the STA potential with time was examined in constant-amplitude contractions of long duration (Fig. 3). The motor units were firing during most of the recordings, but silent periods of up to a few minutes occurred (Westad et al. 2003). A significant increase in
Mean values, statistical significance and numbers of motor units are shown in the diagrams.
STA area after 3 min (mean area 118% relative to start of contraction; P < 0.0001) was observed. However, there was also a small increase in mean SEMG amplitude (mean difference 0.3% EMGmax) and the proportional change in STA area at 3 min versus start of contraction was significantly correlated with the corresponding change in SEMG amplitude (r = 0.46, P = 0.0013). The increase in STA area was reduced but still higher than at the start of contraction when normalizing the STA area to initial SEMG amplitude (108%, P = 0.02). The STA area was lower than 3-min values at 10 min (110%), and showed no difference from the initial values at 30 min (99%). At the later times there was no net change in SEMG amplitude relative to the start of contraction.
, 百拇医药
STA potentials at start and after 3, 10 and 30 min of constant-amplitude contraction are shown in Fig. 4A for a continuously firing motor unit. The stability of these potentials is contrasted by the 2.5-fold increase in STA area during brief periods of elevated SEMG amplitude (Fig. 4B). STA area was also examined immediately before and after silent periods of more than 1 min duration. In the example of Fig. 4C there was a 10% reduction in STA area following the silent period; however, both a slight increase and decrease of the STA potential was observed following re-recruitment.
, 百拇医药
A, motor unit with sustained firing for 30 min in constant-amplitude contraction (motor unit 3 in Fig. 2 of Westad et al. 2003); motor unit potentials shown at start and after 3, 10, and 30 min. B, STA-derived potentials of the same motor unit before, during, and after transient elevation of SEMG amplitude from 6.5 to peak value 15.3% EMGmax (inset, horizontal bars show intervals used to extract STA potentials). Ten elevations from the first 10 min of contraction are averaged. Intervals with elevated SEMG amplitude, ranging from 2 to 4 s, were individually chosen from the contraction profiles (mean amplitude of included intervals 11.6% EMGmax). C, motor unit potentials derived immediately before and after a 2.5 min silent period in a motor unit train, from 20.5 to 23 min after the start of contraction (motor unit 1 in Fig. 2 of Westad et al. 2003).
, 百拇医药
PSTHs of motor unit pairs active for 10 or 30 min were constructed to determine the strength of trapezius motor unit synchrony (Fig. 5A and B). A total of 125 motor unit pairs from 28 experiments (13 subjects) were included. Most PSTH histograms presented a narrow peak on top of a wider shoulder, detected as a steep rising phase in the CUSUM plot. In the example of Fig. 5A, CUSUM is 0.045 for the wide peak, i.e. an excess of 4.5% firings in synchrony with the reference unit. A narrow central peak within the wide peak, 5 ms wide with CUSUM = 0.025, is also defined. Some PSTHs presented a narrow peak only; in the example of Fig. 5B, peak width is 5 ms and CUSUM = 0.044. Histograms show CUSUM values and width of intervals with elevated probability of firing (Fig. 5C and E) and the corresponding values for the narrow central peak (Fig. 5D and F). Strength of synchronization was examined with respect to experimental condition (constant-amplitude, mental stress, typing), but no clear differences emerged. A comparison of the first versus the last 10 min of 30 min contractions did not show any indication of time-dependent change in strength of synchronization.
, 百拇医药
The size of the central peak is quantified by the cumulative sum (CUSUM) technique. A, histogram showing a combination of wide and narrow peak synchrony; the size and width of the peaks are determined by inflections in the CUSUM plot above the histogram (single arrows indicate narrow-peak values, feathered arrows wide-peak values). Axis on right shows CUSUM counts. B, motor units showing only narrow peak synchrony. C and E, normalized amplitude (C) and width (E) of central region with elevated probability of firing in the PSTH histograms, detected by CUSUM analysis. D and F, amplitude (D) and width (F) of narrow peak components in the PSTH histograms. If only a narrow-peak component is defined (e.g. example in B), this is included in all diagrams C–F. In 5 out of 125 histograms, no narrow-peak component could be defined within the relatively wide central peaks (11–27 ms). These are not included in D and F.
, 百拇医药
A simulation of the compound SEMG signal was performed by constructing trains of trapezius motor unit firings at 10 and 12.5 pps on the basis of 10 low-threshold, STA-derived potentials. Increasing from 1 to 40 motor units firing at 10 pps increased the SEMG signal six-fold. A similar simulation on the basis of an ensemble of low-threshold FDI motor units (width of the biphasic wave 7 ms versus mean width 22 ms for trapezius motor units) showed a 14-fold increase in ARV amplitude for 40 motor units firing at 10 pps. This demonstrates substantially more cancellation of opposite-phase components for the wider trapezius motor unit potentials (amplitude cancellation 85 versus 65% for FDI-type motor unit potentials). When increasing the number of trapezius motor units to 70 and the firing rate to 12.5 pps, a 17-fold increase in ARV was observed, relative to one motor unit firing at 10 pps. The mean area of trapezius motor unit potentials, extrapolated to zero SEMG amplitude, is close to 0.02% EMGmax s. A single motor unit, firing in isolation at 10 pps, thus contributes approximately 0.2% EMGmax to the SEMG amplitude. On this basis, the simulation indicates that 70 motor units firing at 12.5 pps generate a SEMG response of 3.4% EMGmax. If the actual area of the 10 low-threshold motor units (which can be spotted as singular events in the SEMG recordings) is used as basis for this calculation, the simulated SEMG amplitude is 4.4% EMGmax.
, 百拇医药
Discussion
The main findings of this study are, first, STA-derived trapezius motor unit potentials show strong dependence on SEMG amplitude in low-amplitude contractions. Second, STA-derived motor unit potentials recruited below 10% EMGmax are of similar size, regardless of recruitment threshold, when estimated from the same recording interval. The simulation experiments indicate that a considerable number of motor units are active at low contraction levels.
, 百拇医药
The uniform size of motor unit potentials at low SEMG amplitude is probably due to the sheet-like appearance of the trapezius muscle, located on average between 7 and 12 mm below the surface (Jensen et al. 1994). The motor units presented muscle fibres at the midpoint of the transverse extent of the surface electrodes, ensuring a location within the transverse pick-up area of the SEMG electrode. Distance from the recorded motor units to the SEMG electrode is thereby largely eliminated as a source of variation for STA-derived potentials in this study.
, 百拇医药
Motor unit synchrony clearly contributes to the increase in motor unit potentials with contraction amplitude. As an example, the motor units in Fig. 5B fired 223 times in synchrony (above the number expected by chance) during trains of 5100 and 5400 firings. The 4.4% (4.1% for second motor unit) added number of firings within ±2 ms of the trigger unit increases the STA-derived motor unit potential accordingly. The contribution from firing synchrony to the increase in STA area is for the most part defined by the narrow-peak PSTHs, as there will be increasing cancellation from opposite-phase motor unit potentials at wider dispersion of the firing events. This is consistent with the observed increase in width of approximately 4 ms for STA-derived potentials at increasing SEMG amplitude.
, 百拇医药
Although the increase in motor unit potentials with increasing SEMG amplitude is, to a major extent, due to motor unit synchrony, changes to the signal source may contribute (McComas et al. 1994). The mechanism is increased Na+–K+ pump activity due to elevated K+ concentration in the interstitial fluid, causing membrane hyperpolarization and larger muscle fibre action potentials both in active and inactive muscle fibres (Hicks & McComas, 1989; Kuiack & McComas, 1992). Several studies indicate an increase of 10–30% following tetanic stimulation, assessed as transient increase in M-wave (West et al. 1996; Harrison & Flatman, 1999) and in motor unit potentials elicited by nerve stimulation (Thomas et al. 2004). A third, less likely explanation of the increase in motor unit potentials with SEMG amplitude is more efficient signal transmission from the muscle fibres to the recording electrode. Contact pressure to the shoulder had the effect of reducing motor unit potentials at relatively high SEMG amplitude.
, 百拇医药
Contrasting the immediate effect of contraction amplitude on the size of STA-derived motor unit potentials, there was little or no time-dependent change in size and duration for constant-amplitude contractions within the time period studied. This indicates that the active motor units are little affected by fatigue. However, the sensitivity to detecting changes in motor unit potential is low, as the triggering motor unit may contribute as little as 25–30% to the STA-derived potentials, if the potential derived at low contraction amplitude is the true representation of the motor unit.
, http://www.100md.com
The consistent and largely invariant size and firing rate (Westad et al. 2004) of low-threshold motor units in sustained contractions, together with the simulation experiments allows an examination of motor unit estimates based on the presumed effects of firing synchrony on motor unit size. The mean STA-derived motor unit area was 0.06% EMGmax s at SEMG amplitude 5% EMGmax, while the area of singular firing motor units was ~ 0.02% EMGmax s. If the apparent increase in motor unit area is due to synchronization at the mean narrow-peak value (2.8%), about 70 motor units are required to achieve the observed effect. Mean firing rate of motor units in a sustained contraction at 5% EMGmax is 12.5 pps. The simulation indicates SEMG amplitude of 3.4–4.4% EMGmax. This is in reasonable agreement with the 5% EMGmax starting point, as several factors may perturb estimates both for the simulation trials and the synchronization effects. Such factors include changes to the motor unit source potentials and effects due to the distribution of narrow-peak and wide-peak synchronization. The simulation experiments are simplistic as they do not consider the effects of e.g. firing rate variability and firing rate synchronization. However, a recent simulation study (Keenan et al. 2004) concluded that both these variables may not exert much influence on the synthesized SEMG signal, while the width of the motor unit potential had a major influence, as is also shown by our simulations.
, 百拇医药
This study has practical implications for the interpretation of results using the STA technique. Shortcomings of the STA technique have been recognized for a long time (Kirkwood, 1979). Motor unit synchronization is suggested as a cause of deviant results in the comparison of motor unit force determined by STA versus nerve microstimulation (Keen & Fuglevand, 2004). Synchronization is usually considered less of a problem in spike-triggered averaging to determine motor unit potentials, due to the restricted width of these potentials, and the method has been used to quantify motor unit size (Conwit et al. 1997; Jensen et al. 2000), motor unit numbers (Boe et al. 2004), and muscle fibre conduction velocity by use of electrode arrays (Farina et al. 2002). However, it appears that motor unit synchrony has considerable influence on STA-extracted motor unit potentials, especially for muscles presenting wide potentials. This restricts the applicability of the STA technique in studies of large muscles located away from the skin surface and with large numbers of motor units active at low contraction levels. This restriction is clearly valid for the trapezius and is probably a general problem in STA-based studies of motor units in postural muscles. A reduction in the separation of the electrode contact areas from 20 to 10 mm does not appreciably reduce the width of the motor unit potential at the skin surface, as motor unit potentials extracted from trapezius recordings with electrode bars separated by 10 mm (Westgaard & De Luca, 1999) had similar width to those presented in this study.
, 百拇医药
Trapezius motor units are expected to increase in size with increasing recruitment threshold, according to the Henneman size principle (Henneman et al. 1965), which is established and later confirmed in studies of limb muscles (e.g. Milner-Brown et al. 1973). Less is known about the motor control of postural muscles, which do not lend themselves to this type of experimentation, due to the difficulty of performing reliable motor unit force recordings, but it is generally assumed that motor control principles valid for extremity muscles hold for (compartments of) postural muscles. The size of trapezius motor units is only indirectly determined by the area of STA-derived motor unit potentials. As the observed motor units probably contribute only a fraction of the apparent size of the motor unit potential, and the added contribution is the result of highly non-linear processes, an estimate of motor unit size by the STA technique cannot be performed. However, if the non-linear effects due to cancellation of opposite-phase motor unit potentials equally influence the size of simultaneous active early- and late-recruited units, the relative size of motor units derived in the same time interval should allow inferences concerning their proportional size. Table 1 shows there is no trend of increase in size of trapezius motor units with increasing recruitment threshold for contraction amplitudes <10% EMGmax. The proportional variation in size is similar to the range of variation for the 10 lowest-threshold motor units. On this basis, it may be speculated that trapezius motor units recruited below 10% EMGmax are of equal size. With due caution in consideration of the possible sources of error that would modify this conclusion, such a motor unit organization is quite different from that of FDI, which shows progressive increase in motor unit size with recruitment threshold also for low-threshold motor units (Milner-Brown et al. 1973). It may further be argued that such a difference in the organization of motor units makes aetiological sense in that there is little need for fine control of trapezius force, beyond controlling the number of motor units recruited. Instead, the habitual control requirement is sustained low-amplitude, long-duration force generation, which in part is maintained by load sharing through motor unit substitution.
, 百拇医药
In conclusion, STA-derived trapezius motor unit potentials show strong dependence on SEMG amplitude due to motor unit synchrony and, possibly, increased amplitude of muscle fibre action potentials. Motor unit potentials in trapezius and postural muscles with relatively long signal transmission distance are especially sensitive to the effects of motor unit synchrony. Finally, trapezius motor units active in contractions with amplitude <10% EMGmax may have a uniform size.
, http://www.100md.com
References
Boe SG, Stashuk DW & Doherty TJ (2004). Motor unit number estimation by decomposition-enhanced spike-triggered averaging: control data, test-retest reliability, and contractile level effects. Muscle Nerve 29, 693–699.
Buchthal F, Guld C & Rosenflack P (1957). Multielectrode study of the territory of a motor unit. Acta Physiol Scand 39, 83–103.
Conwit RA, Tracy B, Jamison C, McHugh M, Stashuk D, Brown WF & Metter EJ (1997). Decomposition-enhanced spike-triggered averaging: contraction level effects. Muscle Nerve 20, 976–982.
, http://www.100md.com
Day SJ & Hulliger M (2001). Experimental simulation of cat electromyogram: evidence for algebraic summation of motor-unit action-potential trains. J Neurophysiol 86, 2144–2158.
De Luca CJ & Adam A (1999). Decomposition and analysis of intramuscular electromyographic signals. In Modern Techniques in Neuroscience Research, ed. Windhorst U, Johansson H, pp. 757–776. Springer, Heidelberg.
De Luca CJ, LeFever RS, McCue MP & Xenakis AP (1982). Behaviour of human motor units in different muscles during linearly varying contractions. J Physiol 329, 113–128.
, http://www.100md.com
Ellaway PH (1978). Cumulative sum technique and its application to the analysis of peristimulus time histograms. Electroencephalogr Clin Neurophysiol 45, 302–304.
Farina D, Arendt-Nielsen L, Merletti R & Graven-Nielsen T (2002). Assessment of single motor unit conduction velocity during sustained contractions of the tibialis anterior muscle with advanced spike triggered averaging. J Neurosci Meth 115, 1–12.
Harrison AP & Flatman JA (1999). Measurement of force and both surface and deep M wave properties in isolated rat soleus muscles. Am J Physiol Regul Integr Comp Physiol 277, R1646–R1653.
, http://www.100md.com
Henneman E, Somjen G & Carpenter DO (1965). Excitability and inhibitability of motoneurons of different sizes. J Neurophysiol 28, 599–620.
Hicks A & McComas AJ (1989). Increased sodium pump activity following repetitive stimulation of rat soleus muscle. J Physiol 414, 337–349.
Jensen BR, Jrgensen K & Sjgaard G (1994). The effect of prolonged isometric contractions on muscle fluid balance. Eur J Appl Physiol 69, 439–444.
, http://www.100md.com
Jensen BR, Pilegaard M & Sjgaard G (2000). Motor unit recruitment and rate coding in response to fatiguing shoulder abductions and subsequent recovery. Eur J Appl Physiol 83, 190–199.
Jensen C, Vasseljen O & Westgaard RH (1993). The influence of electrode position on bipolar surface electromyogram recordings of the upper trapezius muscle. Eur J Appl Physiol 67, 266–273.
Keen DA & Fuglevand AJ (2004). Distribution of motor unit force in human extensor digitorum assessed by spike-triggered averaging and intraneural microstimulation. J Neurophysiol 91, 2515–2523.
, http://www.100md.com
Keenan KG, Farina D, Maluf KS, Merletti R & Enoka RM (2005). Influence of amplitude cancellation on the simulated surface electromyogram. J Appl Physiol 98, 120–131.
Kern DS, Semmler JG & Enoka RM (2001). Long-term activity in upper- and lower-limb muscles of humans. J Appl Physiol 91, 2224–2232.
Kirkwood PA (1979). On the use and interpretation of cross-correlation measurements in the mammalian central nervous system. J Neurosci Meth 1, 107–132.
, 百拇医药
Kirkwood PA, Sears TA, Tuck DL & Westgaard RH (1982). Variations in the time course of the synchronization of intercostal motoneurones in the cat. J Physiol 327, 105–135.
Kuiack S & McComas A (1992). Transient hyperpolarisation of non-contracting muscle fibres in anaesthetized rats. J Physiol 454, 609–618.
LeFever RS & De Luca CJ (1982). A procedure for decomposing the myoelectric signal into its constituent action potentials – Part 1: technique, theory and implementation. IEEE Trans Biomed Eng 29, 149–157.
, http://www.100md.com
McComas AJ, Galea V & Einhorn RW (1994). Pseudofacilitation: a misleading term. Muscle Nerve 17, 599–607.
Mambrito B & De Luca CJ (1984). A technique for the detection, decomposition and analysis of the EMG signal. Electroencephalogr Clin Neurophysiol 58, 175–188.
Milner-Brown HS, Stein RB & Yemm R (1973). The orderly recruitment of human motor units during voluntary isometric contractions. J Physiol 230, 359–370.
, http://www.100md.com
Roeleveld K, Stegeman DF, Vingerhoets HM & van Oosterom A (1997). Motor unit potential contribution to surface electromyography. Acta Physiol Scand 160, 175–183.
Taylor AM, Steege JW & Enoka RM (2002). Motor-unit synchronization alters spike-triggered average force in simulated contractions. J Neurophysiol 88, 265–276.
Thomas CK, Johansson RS & Bigland-Ritchie B (2004). Changes in thenar motor unit electromyographic activity and force with repeated activation. In Proceedings of the 15th ISEK Congress, ed. Roy SH, Bonato P, Meyer J, p. 160. Boston.
, 百拇医药
Thorn S, Forsman M, Zhang Q & Taoda K (2002). Low-threshold motor unit activity during a 1-h static contraction in the trapezius muscle. Int J Ind Erg 30, 225–236.
Wrsted M, Bjrklund RA & Westgaard RH (1994). The effect of motivation on shoulder-muscle tension in attention-demanding tasks. Ergonomics 37, 363–376.
West W, Hicks A, McKelvie R & O'Brien J (1996). The relationship between plasma potassium, muscle membrane excitability and force following quadriceps fatigue. Pflugers Arch 432, 43–49.
, 百拇医药
Westad C, Mork PJ & Westgaard RH (2004). Firing patterns of low-threshold trapezius motor units in feedback-controlled contractions and vocational motor activities. Exp Brain Res 158, 465–473.
Westad C, Westgaard RH & De Luca CJ (2003). Motor unit recruitment and derecruitment induced by brief increase in contraction amplitude of the human trapezius muscle. J Physiol 552, 645–656.
Westgaard RH & Bjrklund R (1987). Generation of muscle tension additional to postural muscle load. Ergonomics 30, 911–923.
, 百拇医药
Westgaard RH & De Luca CJ (1999). Motor unit substitution in long-duration contractions of the human trapezius muscle. J Neurophysiol 82, 501–504.
Westgaard RH & De Luca CJ (2001). Motor control of low-threshold motor units in the human trapezius muscle. J Neurophysiol 85, 1777–1781.
Yao W, Fuglevand AJ & Enoka RM (2000). Motor-unit synchronization increases EMG amplitude and decreases force steadiness of simulated contractions. J Neurophysiol 83, 441–452., http://www.100md.com(C Westad and R. H Westgaa)
Abstract
The spike-triggered averaged (STA) technique was used to examine trapezius motor unit potentials and their dependence on contraction amplitude and firing history. Individual motor unit firings were identified by a fine-wire intramuscular electrode, while STA-derived potentials were extracted from the simultaneously recorded surface electromyographic (SEMG) signal. Amplitude-controlled contractions and contractions with typing tasks and mental stress were carried out. STA potentials were mostly derived from 20 s intervals of firing. Motor unit synchrony was estimated by peristimulus time histograms (PSTHs). An association between SEMG amplitude and STA-derived motor unit potentials was found: motor unit area showed a four-fold increase when SEMG amplitude increased from 1.5 to 10.5% of the root mean square-detected SEMG signal at maximal voluntary contraction (%EMGmax). Low- and higher threshold motor unit potentials, all with recruitment thresholds <10% EMGmax, had similar area at the same contraction amplitude. A significant increase in the STA-derived potentials was observed after 3 min of constant-amplitude contractions; however, this difference was reduced after 10 min and no longer present after 30 min of contraction. Motor unit synchrony accounted for, on average, 2.8% additional firings within 2 ms of the triggering motor unit. We conclude that the increase in STA-derived potentials with contraction amplitude is, to a major extent, due to motor unit synchrony, limiting the applicability of this method in postural muscles presenting wide motor unit potentials. The similar area of motor units at same SEMG amplitude may indicate that trapezius motor units recruited below 10% EMGmax are of similar size and thus not organized according to the Henneman size principle.
, 百拇医药
Introduction
Most studies of motor activity in postural muscles use surface (SEMG) or multiunit intramuscular electromyographic recordings, while it is tacitly assumed that motor units are organized and controlled in a similar way to extremity muscles (e.g. De Luca et al. 1982). One of the governing principles of motor control is the Henneman size principle (Henneman et al. 1965); motor units become progressively larger and develop more force with increasing recruitment threshold. However, the data to support this principle all derive from experimentation on extremity muscles. The force requirement of postural muscles, such as sustained force development over long time periods, is quite different from that of limb muscles. It is therefore pertinent to query whether the size principle is applicable to postural muscles.
, 百拇医药
Trapezius motor units can be continuously active for up to one hour (Thorn et al. 2002). Whole-day SEMG recordings have shown that trapezius has a more sustained activity pattern than upper and lower limb muscles (Mork and Westgaard, unpublished observations; Kern et al. 2001). Trapezius motor unit firing pattern deviates from motor units of a distal hand muscle (first dorsal interosseus; FDI) in slow ramp contractions, by sudden onset and higher firing rates of new relative to previously recruited motor units (Westgaard & De Luca, 2001). Abrupt changes in recruitment threshold during sustained contractions have been observed, causing ‘silent’ periods without motor unit firing, and ‘substitution’ by other motor units with initially higher threshold (Westgaard & De Luca, 1999; Westad et al. 2003). Altogether, this shows that the control of low-threshold trapezius motor units has characteristics that are different from those of limb muscles.
, http://www.100md.com
As it is not possible to determine motor unit force in a postural muscle like trapezius with the available technology, we wanted to examine whether motor unit size could be deduced from SEMG recordings. The representation of single motor units in the SEMG signal was determined by spike triggered averaging (STA). This was anticipated to generate relatively consistent motor unit potentials, as trapezius has a sheet-like appearance with mean thickness 5 mm at the preferred SEMG electrode location (range 3–9 mm; Jensen et al. 1994). A muscle thickness of 5 mm corresponds well to the typical extent of motor unit territory (Buchthal et al. 1957; Roeleveld et al. 1997).
, 百拇医药
The first aim of this study was to examine the hypothesis that low-threshold trapezius motor units show high consistency in their contribution to the SEMG signal, but with an increase in amplitude (i.e. size) of motor unit potentials with higher recruitment threshold as formulated in the Henneman size principle. However, it appeared the STA-derived potentials of the same motor unit were highly dependent on SEMG amplitude. A second aim was therefore to examine the cause of the increase in the area of STA-derived motor unit potentials with contraction amplitude. A possible explanation is short-term synchrony in motor unit firing, which causes an increase in the apparent SEMG amplitude of STA-derived motor unit potentials (Kirkwood, 1979; Yao et al. 2000; Taylor et al. 2002). Motor unit synchrony was examined by constructing peristimulus time histograms (PSTHs) of firing events for pairs of motor units.
, 百拇医药
Methods
Material primarily collected for other study purposes was analysed (Westad et al. 2003; Westad et al. 2004). This included trapezius electromyographic recordings from 16 healthy subjects, six males and ten females, age ranging from 20 to 57 years. Experimental procedures and methods are described in detail in the previous papers; here the description of methods and procedures is limited to those aspects relevant to the present study. The experiments were carried out according to the Declaration of Helsinki. Each subject read and signed an informed consent form approved by the local Institutional Review Board prior to participating in the study.
, http://www.100md.com
Experimental procedures
Simultaneous electromyographic (EMG) recordings by surface and intramuscular electrodes were performed. The indwelling EMG signal was used to detect firing events of identified trapezius motor units while STA-derived potentials were extracted from the surface EMG (SEMG) signal. The trapezius SEMG signal was also root-mean-square (RMS) detected and used to indicate and, in some procedures, to control trapezius contraction amplitude by displaying signal amplitude on a visual display screen in front of the subject. The SEMG signal was calibrated in technical units (μV) and as a percentage of the RMS-detected EMG activity at maximal voluntary contraction (%EMGmax).
, 百拇医药
The feedback-controlled contractions consisted of slow ramp contractions (nominally from 0 to 10% EMGmax in 3 min, rise time 0.055% EMGmax s–1; some contractions reached lower amplitudes) and constant amplitude contractions of one to 30 min duration, amplitudes ranging from 1 to 8% EMGmax. All feedback-controlled contractions were carried out with the subject seated and straps placed over the shoulders to provide resistance to the attempted movement of elevating the shoulders. Shoulder elevation was performed bilaterally, with EMG data collected from the left trapezius. Three 30-min and four 10-min contractions with imposed brief periods of transient increase in SEMG activity (Westad et al. 2003) and 10 contractions of 10 min duration without transient increase in amplitude (Westad et al. 2004) provided long-duration, constant-amplitude recordings. These contractions were used to examine the stability of STA-derived potentials over time.
, 百拇医药
Motor tasks mimicking muscle activation in daily living, such as motor activity in typing and trapezius motor response to mental stress were carried out (Westad et al. 2004). These tasks were of 10 min duration and were performed without shoulder straps or any other device (except EMG electrodes) that could influence shoulder geometry or shoulder movement. Contraction amplitude varied within a range of 2–4% EMGmax during these contractions, which had mean amplitudes from <1–9% EMGmax. The mental strain task consisted of a complex, attention-demanding two-choice reaction test presented on the computer screen. An open (‘frame’) and a solid (‘brick’) quadrangle were placed in a square pattern, and an alphanumeric suggestion on how to move the brick to superimpose on the frame was given (Westgaard & Bjrklund, 1987). The subject responded by pressing one of two keys, ‘correct’ or ‘wrong’, with the right or left index finger. A new position of the ‘brick’ and ‘frame’ in the square pattern and a new suggestion then appeared. The execution of the test was self-paced, but the subject first carried out the test for 2 min at a steady pace while attempting to maintain a low failure rate. On the basis of this performance the subject was offered a small monetary reward to perform 10% faster with the same or lower failure rate for 10 min. Feedback provided on the computer screen informed the subjects about the response speed (very slow, slow, OK, fast, or very fast) and whether the answer was correct (Wrsted et al. 1994).
, http://www.100md.com
Physiological recordings and analyses
The SEMG signal was detected by an active differential electrode with two circular recording surfaces (6 mm in diameter, 20 mm interelectrode distance). The electrode was positioned with the medial recording surface 20 mm lateral to the midpoint of a line between the C7 spinous process and the acromion (Jensen et al. 1993). The SEMG signal was band-pass filtered at 10–1000 Hz. The RMS-detected SEMG signal was averaged at a time resolution of 0.2 s. The intramuscular EMG signal was recorded with quadrifilar wire electrodes, constructed by bonding together four 50-μm nylon-coated nickel–chrome alloy wires (‘Stablohm 800A’, California Fine Wire Co, Grover Beach, CA, USA). The wire bundle was cut transversely, exposing only the cross-section of the wires. The wire bundle was placed in a 27-gauge needle and a hook was formed at approximately one millimetre from the exposed end of the wire. The needle was inserted to a depth of approximately 10 mm at a location approximately 10 mm medial to the midpoint of a line between the C7 spinous process and the acromion, along the direction of the muscle fibres recorded by the SEMG electrode. The insertion aimed for the midpoint of the transverse extent of the SEMG electrode pick-up area. The needle was removed and the wire bundle remained lodged in the muscle. Three pairs were chosen as the differential input to the amplifiers. The signals were band-pass filtered from 1 to 10 kHz. All EMG signals were stored on a digital recorder (DATaRec-A160, Racal-Heim Systems GmbH, Bergisch Gladbach, Germany). The signals were subsequently reconverted to an analog form and digitized at a sampling rate of 50 kHz on a PC.
, 百拇医药
The intramuscular EMG signals were resolved into individual motor unit firing trains using the Precision Decomposition technique (LeFever & De Luca, 1982; De Luca & Adam, 1999). This technique uses template matching, template updating, firing probabilities and superposition resolution to identify the individual firing times of the motor units (Mambrito & De Luca, 1984). The firing rates of the motor units were obtained by inverting the time series of the interpulse intervals. Firing rates were low-pass filtered at 0.5 Hz for display purposes, but STA motor unit potentials and estimates of motor unit synchrony were derived from the unfiltered firing rates.
, http://www.100md.com
Data analysis
Firing events of identified motor units were used as triggers in the STA analyses, which were based on recording intervals of 20 s, corresponding to 180–250 firings, unless otherwise stated. The intramuscular electrode was located medial and the SEMG electrode lateral to the endplate region (Jensen et al. 1993), making the STA potential sometimes appear to start before the firing event. The size of the STA potential was quantified as the area underneath the curve traced by the rectified STA potential. All derivations of STA area for a continuous recorded motor unit used the same start and end points. In long-duration, constant-amplitude contractions, STA potentials were extracted at the start of the experiment and 3, 10 and 30 min into the recording. Additional epochs were used to examine STA potentials at different SEMG amplitudes or at marked changes in firing patterns, e.g. STA potential before and after motor unit silent periods.
, 百拇医药
Motor unit synchronization was determined by constructing peristimulus time histograms between simultaneously firing motor units (PSTHs; e.g. Kirkwood et al. 1982). The CUSUM method (Ellaway, 1978) was used to determine the number of extra triggering events synchronous with the reference motor unit and the width of the central peak in the PSTH. CUSUM values were normalized by number of firings of the motor unit with the lowest firing rate.
Synthesized SEMG responses from STA-derived motor unit potentials
, 百拇医药
Ten STA-derived motor unit potentials were extracted at low SEMG amplitude (range 0.6–1.6% EMGmax). The rectified area of the motor unit potentials differed by a factor of 2 (from 0.017 to 0.037% EMGmax s). Trains of firings at 10 and 12.5 pulses per second (pps), corresponding to mean firing rates in constant-amplitude contractions of approximately 1 and 5% EMGmax (Westad et al. 2004), were constructed. The start of the individual trains of firings was decided by random draw within the intervals 0–100 and 0–80 ms for motor unit trains at 10 and 12.5 pps, respectively. The SEMG response for the set of motor unit firings was determined by algebraic summation of the motor unit trains (Day & Hulliger, 2001) and average rectified value (ARV) over 0.2 s intervals determined. The simulation of the SEMG signal was repeated, using a STA-extracted motor unit potential from the first dorsal interosseus (FDI), recorded by an electrode with bar-shaped recording surfaces that had a separation of 10 mm (Westgaard & De Luca, 2001). Ten trains of firing were constructed, with the motor unit potentials scaled to mimic the variation in area of the trapezius motor unit potentials.
, 百拇医药
Statistics
Pearson correlation coefficients were determined for regression of STA-derived motor unit area versus SEMG amplitude. A Student t test was used to compare STA potentials obtained with and without the use of shoulder straps. Wilcoxon signed ranks test was used to compare STA areas extracted at different times during sustained, constant-amplitude contractions to initial STA area. A P-value of 0.05, two-tailed, was considered to indicate significant differences.
, 百拇医药
Results
Figure 1 shows trapezius SEMG activity and firing rates of three motor units (1–3) recruited at different amplitudes (i.e. recruitment thresholds) of a slow ramp contraction increasing from 0 to 5% EMGmax in 180 s. Surface representations of the motor unit potentials were extracted by STA at intervals marked by horizontal bars on top of the time plot of SEMG activity in the upper left panel. The STA-derived potentials for each motor unit are shown as superimposed waveforms to the right of the plots of motor unit firing rates. The top right scatter plot shows rectified STA area as a function of SEMG amplitude. There was a marked increase in amplitude and area of the STA potentials with increasing SEMG amplitude. A near linear (r = 0.99), three-fold increase in STA area (from 0.22 to 0.66 μV s) was observed for the lowest threshold motor unit for SEMG amplitude increasing from 1.4 to 5.0% EMGmax. Although the higher-threshold motor units presented larger STA potentials than the low-threshold motor unit at recruitment of the respective motor units, the STA areas were closely similar at same contraction amplitude. A widening of the STA potential of the lowest threshold motor unit, in the order of 4 ms, from low to high SEMG amplitude was observed.
, 百拇医药
Three motor units with different recruitment thresholds were detected during the contraction (numbered 1–3); time plots of firing rates are shown below the SEMG recording. Spike-triggered averaged (STA) motor unit potentials were derived at intervals indicated by horizontal bars above the SEMG recording (20 s duration, >200 firings). Superimposed plots of the STA-derived potentials are shown to the right of each firing rate plot. STA area is determined as the area underneath the curve between vertical broken lines. The top right scatter plot shows STA area versus SEMG amplitude (1: , 2: , 3: ).
, http://www.100md.com
STA-derived motor unit potentials in other slow ramp contractions behaved as illustrated in Fig. 1. STA area always increased with contraction amplitude and in most instances there was a widening of the STA potential similar to that shown in Fig. 1. The width of STA-derived potentials at low contraction amplitude was relatively constant: mean width of the biphasic wave for 10 motor units was 22 ms (range 17–29 ms), with SEMG amplitude ranging from 0.6 to 1.6% EMGmax. Figure 2A shows regression lines of STA area versus SEMG amplitude for 15 low-threshold motor units in slow ramp contractions. The regression lines are based on four or five derivations of the STA potential. The extent of the lines delineates the range of STA derivations. Regression coefficients were near unity (mean value 0.96, range 0.84–1.0). The regression lines are re-plotted in Fig. 2B with STA amplitude calibrated in% EMGmax; intersubject variability is presumably reduced because the re-calibration compensates for intersubject variation in signal transmission to the surface electrode. Finally, regression of STA area (% EMGmax s) versus SEMG amplitude was established for all STA areas derived in these experiments (470 values from 163 motor unit recordings; Fig. 2C). Separate regressions were determined for STA potentials obtained in shoulder elevation (with straps providing resistance to the movement) and in trials without contact pressure to the shoulders. STA potentials were sorted by SEMG amplitude in intervals spanning 2% EMGmax. Mean and S.D. of STA area were determined. The increase in STA area with SEMG amplitude was nearly linear, with low S.D. especially at low SEMG amplitude. STA area was reduced in contractions with shoulder straps, relative to contractions without straps for SEMG amplitude >8% EMGmax (P < 0.01; STA areas normalized to 10% EMGmax by regression). There was no difference between the two conditions at lower SEMG amplitudes.
, http://www.100md.com
A, regression lines of STA area (μV s) versus contraction amplitude (%EMGmax) in slow ramp contractions. The extent of the regression lines indicates the range of intervals used to extract STA potentials. B, regression lines in A are re-plotted with STA area calibrated as (%EMGmax s). C, scatter plot of STA area versus SEMG amplitude of all STA-derived motor unit potentials (470 potentials from 163 recordings). Two groups are distinguished: potentials derived in experiments with shoulder straps providing resistance to the attempted elevation of the shoulders (), and contractions without restrictions on shoulder movement (). Mean value and S.D. of STA area (%EMGmax s) are shown for STA potentials within SEMG amplitude intervals of 2% EMGmax. Horizontal position shows mean value of SEMG amplitude for the different samples (n = 14–99).
, 百拇医药
A four-fold increase in STA area was observed in trials without the use of straps, from 0.03 to 0.12% EMGmax s, for an increase in SEMG amplitude from 1.5 to 10.5% EMGmax. Extrapolation of the regression lines to zero SEMG amplitude indicates STA areas of 0.014 (no straps) and 0.016% EMGmax s (straps) for low-threshold motor units firing in isolation. Five motor units recorded at SEMG amplitude <1% EMGmax had a mean area of 0.023%EMGmax·s.
STA potentials of motor units with high and low recruitment threshold were compared by quantifying STA area of higher threshold motor units in percentage of the simultaneously determined STA area of the lowest threshold motor unit. Table 1 shows the results for lowest-threshold motor units recruited in the intervals 0–2 and 2–4% EMGmax. There was no systematic change in STA area with recruitment threshold for motor units with thresholds <10% EMGmax.
, 百拇医药
The stability of the STA potential with time was examined in constant-amplitude contractions of long duration (Fig. 3). The motor units were firing during most of the recordings, but silent periods of up to a few minutes occurred (Westad et al. 2003). A significant increase in
Mean values, statistical significance and numbers of motor units are shown in the diagrams.
STA area after 3 min (mean area 118% relative to start of contraction; P < 0.0001) was observed. However, there was also a small increase in mean SEMG amplitude (mean difference 0.3% EMGmax) and the proportional change in STA area at 3 min versus start of contraction was significantly correlated with the corresponding change in SEMG amplitude (r = 0.46, P = 0.0013). The increase in STA area was reduced but still higher than at the start of contraction when normalizing the STA area to initial SEMG amplitude (108%, P = 0.02). The STA area was lower than 3-min values at 10 min (110%), and showed no difference from the initial values at 30 min (99%). At the later times there was no net change in SEMG amplitude relative to the start of contraction.
, 百拇医药
STA potentials at start and after 3, 10 and 30 min of constant-amplitude contraction are shown in Fig. 4A for a continuously firing motor unit. The stability of these potentials is contrasted by the 2.5-fold increase in STA area during brief periods of elevated SEMG amplitude (Fig. 4B). STA area was also examined immediately before and after silent periods of more than 1 min duration. In the example of Fig. 4C there was a 10% reduction in STA area following the silent period; however, both a slight increase and decrease of the STA potential was observed following re-recruitment.
, 百拇医药
A, motor unit with sustained firing for 30 min in constant-amplitude contraction (motor unit 3 in Fig. 2 of Westad et al. 2003); motor unit potentials shown at start and after 3, 10, and 30 min. B, STA-derived potentials of the same motor unit before, during, and after transient elevation of SEMG amplitude from 6.5 to peak value 15.3% EMGmax (inset, horizontal bars show intervals used to extract STA potentials). Ten elevations from the first 10 min of contraction are averaged. Intervals with elevated SEMG amplitude, ranging from 2 to 4 s, were individually chosen from the contraction profiles (mean amplitude of included intervals 11.6% EMGmax). C, motor unit potentials derived immediately before and after a 2.5 min silent period in a motor unit train, from 20.5 to 23 min after the start of contraction (motor unit 1 in Fig. 2 of Westad et al. 2003).
, 百拇医药
PSTHs of motor unit pairs active for 10 or 30 min were constructed to determine the strength of trapezius motor unit synchrony (Fig. 5A and B). A total of 125 motor unit pairs from 28 experiments (13 subjects) were included. Most PSTH histograms presented a narrow peak on top of a wider shoulder, detected as a steep rising phase in the CUSUM plot. In the example of Fig. 5A, CUSUM is 0.045 for the wide peak, i.e. an excess of 4.5% firings in synchrony with the reference unit. A narrow central peak within the wide peak, 5 ms wide with CUSUM = 0.025, is also defined. Some PSTHs presented a narrow peak only; in the example of Fig. 5B, peak width is 5 ms and CUSUM = 0.044. Histograms show CUSUM values and width of intervals with elevated probability of firing (Fig. 5C and E) and the corresponding values for the narrow central peak (Fig. 5D and F). Strength of synchronization was examined with respect to experimental condition (constant-amplitude, mental stress, typing), but no clear differences emerged. A comparison of the first versus the last 10 min of 30 min contractions did not show any indication of time-dependent change in strength of synchronization.
, 百拇医药
The size of the central peak is quantified by the cumulative sum (CUSUM) technique. A, histogram showing a combination of wide and narrow peak synchrony; the size and width of the peaks are determined by inflections in the CUSUM plot above the histogram (single arrows indicate narrow-peak values, feathered arrows wide-peak values). Axis on right shows CUSUM counts. B, motor units showing only narrow peak synchrony. C and E, normalized amplitude (C) and width (E) of central region with elevated probability of firing in the PSTH histograms, detected by CUSUM analysis. D and F, amplitude (D) and width (F) of narrow peak components in the PSTH histograms. If only a narrow-peak component is defined (e.g. example in B), this is included in all diagrams C–F. In 5 out of 125 histograms, no narrow-peak component could be defined within the relatively wide central peaks (11–27 ms). These are not included in D and F.
, 百拇医药
A simulation of the compound SEMG signal was performed by constructing trains of trapezius motor unit firings at 10 and 12.5 pps on the basis of 10 low-threshold, STA-derived potentials. Increasing from 1 to 40 motor units firing at 10 pps increased the SEMG signal six-fold. A similar simulation on the basis of an ensemble of low-threshold FDI motor units (width of the biphasic wave 7 ms versus mean width 22 ms for trapezius motor units) showed a 14-fold increase in ARV amplitude for 40 motor units firing at 10 pps. This demonstrates substantially more cancellation of opposite-phase components for the wider trapezius motor unit potentials (amplitude cancellation 85 versus 65% for FDI-type motor unit potentials). When increasing the number of trapezius motor units to 70 and the firing rate to 12.5 pps, a 17-fold increase in ARV was observed, relative to one motor unit firing at 10 pps. The mean area of trapezius motor unit potentials, extrapolated to zero SEMG amplitude, is close to 0.02% EMGmax s. A single motor unit, firing in isolation at 10 pps, thus contributes approximately 0.2% EMGmax to the SEMG amplitude. On this basis, the simulation indicates that 70 motor units firing at 12.5 pps generate a SEMG response of 3.4% EMGmax. If the actual area of the 10 low-threshold motor units (which can be spotted as singular events in the SEMG recordings) is used as basis for this calculation, the simulated SEMG amplitude is 4.4% EMGmax.
, 百拇医药
Discussion
The main findings of this study are, first, STA-derived trapezius motor unit potentials show strong dependence on SEMG amplitude in low-amplitude contractions. Second, STA-derived motor unit potentials recruited below 10% EMGmax are of similar size, regardless of recruitment threshold, when estimated from the same recording interval. The simulation experiments indicate that a considerable number of motor units are active at low contraction levels.
, 百拇医药
The uniform size of motor unit potentials at low SEMG amplitude is probably due to the sheet-like appearance of the trapezius muscle, located on average between 7 and 12 mm below the surface (Jensen et al. 1994). The motor units presented muscle fibres at the midpoint of the transverse extent of the surface electrodes, ensuring a location within the transverse pick-up area of the SEMG electrode. Distance from the recorded motor units to the SEMG electrode is thereby largely eliminated as a source of variation for STA-derived potentials in this study.
, 百拇医药
Motor unit synchrony clearly contributes to the increase in motor unit potentials with contraction amplitude. As an example, the motor units in Fig. 5B fired 223 times in synchrony (above the number expected by chance) during trains of 5100 and 5400 firings. The 4.4% (4.1% for second motor unit) added number of firings within ±2 ms of the trigger unit increases the STA-derived motor unit potential accordingly. The contribution from firing synchrony to the increase in STA area is for the most part defined by the narrow-peak PSTHs, as there will be increasing cancellation from opposite-phase motor unit potentials at wider dispersion of the firing events. This is consistent with the observed increase in width of approximately 4 ms for STA-derived potentials at increasing SEMG amplitude.
, 百拇医药
Although the increase in motor unit potentials with increasing SEMG amplitude is, to a major extent, due to motor unit synchrony, changes to the signal source may contribute (McComas et al. 1994). The mechanism is increased Na+–K+ pump activity due to elevated K+ concentration in the interstitial fluid, causing membrane hyperpolarization and larger muscle fibre action potentials both in active and inactive muscle fibres (Hicks & McComas, 1989; Kuiack & McComas, 1992). Several studies indicate an increase of 10–30% following tetanic stimulation, assessed as transient increase in M-wave (West et al. 1996; Harrison & Flatman, 1999) and in motor unit potentials elicited by nerve stimulation (Thomas et al. 2004). A third, less likely explanation of the increase in motor unit potentials with SEMG amplitude is more efficient signal transmission from the muscle fibres to the recording electrode. Contact pressure to the shoulder had the effect of reducing motor unit potentials at relatively high SEMG amplitude.
, 百拇医药
Contrasting the immediate effect of contraction amplitude on the size of STA-derived motor unit potentials, there was little or no time-dependent change in size and duration for constant-amplitude contractions within the time period studied. This indicates that the active motor units are little affected by fatigue. However, the sensitivity to detecting changes in motor unit potential is low, as the triggering motor unit may contribute as little as 25–30% to the STA-derived potentials, if the potential derived at low contraction amplitude is the true representation of the motor unit.
, http://www.100md.com
The consistent and largely invariant size and firing rate (Westad et al. 2004) of low-threshold motor units in sustained contractions, together with the simulation experiments allows an examination of motor unit estimates based on the presumed effects of firing synchrony on motor unit size. The mean STA-derived motor unit area was 0.06% EMGmax s at SEMG amplitude 5% EMGmax, while the area of singular firing motor units was ~ 0.02% EMGmax s. If the apparent increase in motor unit area is due to synchronization at the mean narrow-peak value (2.8%), about 70 motor units are required to achieve the observed effect. Mean firing rate of motor units in a sustained contraction at 5% EMGmax is 12.5 pps. The simulation indicates SEMG amplitude of 3.4–4.4% EMGmax. This is in reasonable agreement with the 5% EMGmax starting point, as several factors may perturb estimates both for the simulation trials and the synchronization effects. Such factors include changes to the motor unit source potentials and effects due to the distribution of narrow-peak and wide-peak synchronization. The simulation experiments are simplistic as they do not consider the effects of e.g. firing rate variability and firing rate synchronization. However, a recent simulation study (Keenan et al. 2004) concluded that both these variables may not exert much influence on the synthesized SEMG signal, while the width of the motor unit potential had a major influence, as is also shown by our simulations.
, 百拇医药
This study has practical implications for the interpretation of results using the STA technique. Shortcomings of the STA technique have been recognized for a long time (Kirkwood, 1979). Motor unit synchronization is suggested as a cause of deviant results in the comparison of motor unit force determined by STA versus nerve microstimulation (Keen & Fuglevand, 2004). Synchronization is usually considered less of a problem in spike-triggered averaging to determine motor unit potentials, due to the restricted width of these potentials, and the method has been used to quantify motor unit size (Conwit et al. 1997; Jensen et al. 2000), motor unit numbers (Boe et al. 2004), and muscle fibre conduction velocity by use of electrode arrays (Farina et al. 2002). However, it appears that motor unit synchrony has considerable influence on STA-extracted motor unit potentials, especially for muscles presenting wide potentials. This restricts the applicability of the STA technique in studies of large muscles located away from the skin surface and with large numbers of motor units active at low contraction levels. This restriction is clearly valid for the trapezius and is probably a general problem in STA-based studies of motor units in postural muscles. A reduction in the separation of the electrode contact areas from 20 to 10 mm does not appreciably reduce the width of the motor unit potential at the skin surface, as motor unit potentials extracted from trapezius recordings with electrode bars separated by 10 mm (Westgaard & De Luca, 1999) had similar width to those presented in this study.
, 百拇医药
Trapezius motor units are expected to increase in size with increasing recruitment threshold, according to the Henneman size principle (Henneman et al. 1965), which is established and later confirmed in studies of limb muscles (e.g. Milner-Brown et al. 1973). Less is known about the motor control of postural muscles, which do not lend themselves to this type of experimentation, due to the difficulty of performing reliable motor unit force recordings, but it is generally assumed that motor control principles valid for extremity muscles hold for (compartments of) postural muscles. The size of trapezius motor units is only indirectly determined by the area of STA-derived motor unit potentials. As the observed motor units probably contribute only a fraction of the apparent size of the motor unit potential, and the added contribution is the result of highly non-linear processes, an estimate of motor unit size by the STA technique cannot be performed. However, if the non-linear effects due to cancellation of opposite-phase motor unit potentials equally influence the size of simultaneous active early- and late-recruited units, the relative size of motor units derived in the same time interval should allow inferences concerning their proportional size. Table 1 shows there is no trend of increase in size of trapezius motor units with increasing recruitment threshold for contraction amplitudes <10% EMGmax. The proportional variation in size is similar to the range of variation for the 10 lowest-threshold motor units. On this basis, it may be speculated that trapezius motor units recruited below 10% EMGmax are of equal size. With due caution in consideration of the possible sources of error that would modify this conclusion, such a motor unit organization is quite different from that of FDI, which shows progressive increase in motor unit size with recruitment threshold also for low-threshold motor units (Milner-Brown et al. 1973). It may further be argued that such a difference in the organization of motor units makes aetiological sense in that there is little need for fine control of trapezius force, beyond controlling the number of motor units recruited. Instead, the habitual control requirement is sustained low-amplitude, long-duration force generation, which in part is maintained by load sharing through motor unit substitution.
, 百拇医药
In conclusion, STA-derived trapezius motor unit potentials show strong dependence on SEMG amplitude due to motor unit synchrony and, possibly, increased amplitude of muscle fibre action potentials. Motor unit potentials in trapezius and postural muscles with relatively long signal transmission distance are especially sensitive to the effects of motor unit synchrony. Finally, trapezius motor units active in contractions with amplitude <10% EMGmax may have a uniform size.
, http://www.100md.com
References
Boe SG, Stashuk DW & Doherty TJ (2004). Motor unit number estimation by decomposition-enhanced spike-triggered averaging: control data, test-retest reliability, and contractile level effects. Muscle Nerve 29, 693–699.
Buchthal F, Guld C & Rosenflack P (1957). Multielectrode study of the territory of a motor unit. Acta Physiol Scand 39, 83–103.
Conwit RA, Tracy B, Jamison C, McHugh M, Stashuk D, Brown WF & Metter EJ (1997). Decomposition-enhanced spike-triggered averaging: contraction level effects. Muscle Nerve 20, 976–982.
, http://www.100md.com
Day SJ & Hulliger M (2001). Experimental simulation of cat electromyogram: evidence for algebraic summation of motor-unit action-potential trains. J Neurophysiol 86, 2144–2158.
De Luca CJ & Adam A (1999). Decomposition and analysis of intramuscular electromyographic signals. In Modern Techniques in Neuroscience Research, ed. Windhorst U, Johansson H, pp. 757–776. Springer, Heidelberg.
De Luca CJ, LeFever RS, McCue MP & Xenakis AP (1982). Behaviour of human motor units in different muscles during linearly varying contractions. J Physiol 329, 113–128.
, http://www.100md.com
Ellaway PH (1978). Cumulative sum technique and its application to the analysis of peristimulus time histograms. Electroencephalogr Clin Neurophysiol 45, 302–304.
Farina D, Arendt-Nielsen L, Merletti R & Graven-Nielsen T (2002). Assessment of single motor unit conduction velocity during sustained contractions of the tibialis anterior muscle with advanced spike triggered averaging. J Neurosci Meth 115, 1–12.
Harrison AP & Flatman JA (1999). Measurement of force and both surface and deep M wave properties in isolated rat soleus muscles. Am J Physiol Regul Integr Comp Physiol 277, R1646–R1653.
, http://www.100md.com
Henneman E, Somjen G & Carpenter DO (1965). Excitability and inhibitability of motoneurons of different sizes. J Neurophysiol 28, 599–620.
Hicks A & McComas AJ (1989). Increased sodium pump activity following repetitive stimulation of rat soleus muscle. J Physiol 414, 337–349.
Jensen BR, Jrgensen K & Sjgaard G (1994). The effect of prolonged isometric contractions on muscle fluid balance. Eur J Appl Physiol 69, 439–444.
, http://www.100md.com
Jensen BR, Pilegaard M & Sjgaard G (2000). Motor unit recruitment and rate coding in response to fatiguing shoulder abductions and subsequent recovery. Eur J Appl Physiol 83, 190–199.
Jensen C, Vasseljen O & Westgaard RH (1993). The influence of electrode position on bipolar surface electromyogram recordings of the upper trapezius muscle. Eur J Appl Physiol 67, 266–273.
Keen DA & Fuglevand AJ (2004). Distribution of motor unit force in human extensor digitorum assessed by spike-triggered averaging and intraneural microstimulation. J Neurophysiol 91, 2515–2523.
, http://www.100md.com
Keenan KG, Farina D, Maluf KS, Merletti R & Enoka RM (2005). Influence of amplitude cancellation on the simulated surface electromyogram. J Appl Physiol 98, 120–131.
Kern DS, Semmler JG & Enoka RM (2001). Long-term activity in upper- and lower-limb muscles of humans. J Appl Physiol 91, 2224–2232.
Kirkwood PA (1979). On the use and interpretation of cross-correlation measurements in the mammalian central nervous system. J Neurosci Meth 1, 107–132.
, 百拇医药
Kirkwood PA, Sears TA, Tuck DL & Westgaard RH (1982). Variations in the time course of the synchronization of intercostal motoneurones in the cat. J Physiol 327, 105–135.
Kuiack S & McComas A (1992). Transient hyperpolarisation of non-contracting muscle fibres in anaesthetized rats. J Physiol 454, 609–618.
LeFever RS & De Luca CJ (1982). A procedure for decomposing the myoelectric signal into its constituent action potentials – Part 1: technique, theory and implementation. IEEE Trans Biomed Eng 29, 149–157.
, http://www.100md.com
McComas AJ, Galea V & Einhorn RW (1994). Pseudofacilitation: a misleading term. Muscle Nerve 17, 599–607.
Mambrito B & De Luca CJ (1984). A technique for the detection, decomposition and analysis of the EMG signal. Electroencephalogr Clin Neurophysiol 58, 175–188.
Milner-Brown HS, Stein RB & Yemm R (1973). The orderly recruitment of human motor units during voluntary isometric contractions. J Physiol 230, 359–370.
, http://www.100md.com
Roeleveld K, Stegeman DF, Vingerhoets HM & van Oosterom A (1997). Motor unit potential contribution to surface electromyography. Acta Physiol Scand 160, 175–183.
Taylor AM, Steege JW & Enoka RM (2002). Motor-unit synchronization alters spike-triggered average force in simulated contractions. J Neurophysiol 88, 265–276.
Thomas CK, Johansson RS & Bigland-Ritchie B (2004). Changes in thenar motor unit electromyographic activity and force with repeated activation. In Proceedings of the 15th ISEK Congress, ed. Roy SH, Bonato P, Meyer J, p. 160. Boston.
, 百拇医药
Thorn S, Forsman M, Zhang Q & Taoda K (2002). Low-threshold motor unit activity during a 1-h static contraction in the trapezius muscle. Int J Ind Erg 30, 225–236.
Wrsted M, Bjrklund RA & Westgaard RH (1994). The effect of motivation on shoulder-muscle tension in attention-demanding tasks. Ergonomics 37, 363–376.
West W, Hicks A, McKelvie R & O'Brien J (1996). The relationship between plasma potassium, muscle membrane excitability and force following quadriceps fatigue. Pflugers Arch 432, 43–49.
, 百拇医药
Westad C, Mork PJ & Westgaard RH (2004). Firing patterns of low-threshold trapezius motor units in feedback-controlled contractions and vocational motor activities. Exp Brain Res 158, 465–473.
Westad C, Westgaard RH & De Luca CJ (2003). Motor unit recruitment and derecruitment induced by brief increase in contraction amplitude of the human trapezius muscle. J Physiol 552, 645–656.
Westgaard RH & Bjrklund R (1987). Generation of muscle tension additional to postural muscle load. Ergonomics 30, 911–923.
, 百拇医药
Westgaard RH & De Luca CJ (1999). Motor unit substitution in long-duration contractions of the human trapezius muscle. J Neurophysiol 82, 501–504.
Westgaard RH & De Luca CJ (2001). Motor control of low-threshold motor units in the human trapezius muscle. J Neurophysiol 85, 1777–1781.
Yao W, Fuglevand AJ & Enoka RM (2000). Motor-unit synchronization increases EMG amplitude and decreases force steadiness of simulated contractions. J Neurophysiol 83, 441–452., http://www.100md.com(C Westad and R. H Westgaa)