Spatial variation of compound muscle action potentials across human gastrocnemius medialis

Abstract

The massed action potential (M wave) elicited through nerve stimulation underpins a wide range of physiological and mechanical understanding of skeletal muscle structure and function. Although systematic approaches have evaluated the effect of different factors on M waves, the effect of the location and distribution of activated fibers within the muscle remains unknown. By detecting M waves from the medial gastrocnemius (MG) of 12 participants with a grid of 128 electrodes, we investigated whether different populations of muscle units have different spatial organization within MG. If populations of muscle units occupy discrete MG regions, current pulses of progressively greater intensities applied to the MG nerve branch would be expected to lead to local changes in M-wave amplitudes. Electrical pulses were therefore delivered at 2 pps, with the current pulse amplitude increased every 10 stimuli to elicit different degrees of muscle activation. The localization of MG response to increases in current intensity was determined from the spatial distribution of M-wave amplitude. Key results revealed that increases in M-wave amplitude were detected somewhat locally, by 10-50% of the 128 electrodes. Most importantly, the electrodes detecting greatest increases in M-wave amplitude were localized at different regions in the grid, with a tendency for greater stimulation intensities to elicit M waves in the more distal MG region. The presented results indicate that M waves recorded locally may not provide a representative MG response, with major implications for the estimation of, e.g., the maximal stimulation levels, the number of motor units, and the onset and normalization in H-reflex studies.

Keywords: M wave; electrical stimulation; medial gastrocnemius; motor units; surface EMG.

Fig. 1.

Staircase stimulation protocol. A schematic illustration of the stimulation protocol as well as of the procedure for the calculation of incremental M waves is shown. Stimulation intensity increased equally from zero to the current intensity (I_max), after which no changes in the amplitude of gastrocnemius twitches were observed. For each stimulation level, 10 stimuli were applied at a rate of 2 pps. Incremental M waves were calculated from M waves averaged over these 10 stimuli, as shown at bottom. Note that the amplitude of incremental M waves is large only when an increase in stimulation intensity leads to marked changes in the amplitude/shape of M waves.

Fig. 2.

Incremental M waves and their local representation. A: root mean square (RMS) values computed from M waves obtained for a single participant. M waves were averaged over 10 consecutive stimuli (top), separately for the third (21 mA) and fourth (28 mA) stimulation steps (cf. Fig. 1) and for each of the 64 electrodes (8 rows × 8 columns) positioned over the muscle superficial aponeurosis. The RMS amplitude of incremental M waves is shown at bottom. Electrodes providing incremental M waves with greatest amplitudes are represented with small circles, whereas the coordinate of their barycenter along rows and columns is represented with a large crossed circle. B: RMS amplitude of incremental M waves obtained for the 9 progressive increases in stimulation intensity. Changes in the stimulation intensity leading to each of the 9 increments are indicated at top of each plot. Ranges in bold above array plots denote significant difference in the RMS amplitude of M waves with the increase in stimulation intensity (cf. Fig. 3).

Fig. 3.

Changes in M-wave amplitude with stimulation intensity. A: raw M waves averaged over 10 consecutive stimuli, for each stimulation level. Only M waves detected by electrodes in the third column of the grid and positioned over the gastrocnemius superficial aponeurosis are shown (from row 1 to row 8). B: RMS values averaged over the 64 mean M waves obtained from the grid (8 rows × 8 columns), separately for each stimulation intensity. *Significant differences between RMS values obtained for 2 consecutive stimulation steps [Tukey honestly significant difference (HSD) post hoc; P < 0.005; n = 100: 10 stimuli × 10 stimulation levels].

Fig. 4.

Incremental muscle activation and response. Changes in the peak plantar flexion torque and the RMS amplitude of incremental M waves are shown for each of the 12 subjects tested. The Pearson correlation coefficient, its P value, and the slope of regression (dashed lines) fitted to individual data are shown.

Fig. 5.

Location of incremental M waves along the muscle proximo-distal axis. The RMS amplitude of mean M waves (black circles) and the proximo-distal coordinate of the barycenter (cf. Fig. 2) of incremental M waves are shown for each of the 12 subjects tested. Gray circles indicate the barycenter coordinates obtained in correspondence to changes in stimulation intensity leading to a statistically significant increase in the RMS amplitude of mean M waves (cf. Fig. 3). Spearman ρ and its P value are indicated within each plot. Regression lines are shown so as to provide a qualitative indication of how much the location of incremental M waves (i.e., in their barycenter) changed with increases in stimulation intensity. Only gray circles were considered for the correlation and regression analyses.

Fig. 6.

Location of incremental M waves along the muscle medio-lateral axis. The RMS amplitude of mean M waves (black circles) and the medio-lateral coordinate of the barycenter (cf. Fig. 2) of incremental M waves are shown for each of the 12 subjects tested. Gray circles indicate the barycenter coordinates obtained in correspondence to changes in stimulation intensity leading to a statistically significant increase in the RMS amplitude of mean M waves (cf. Fig. 3). Spearman ρ and its P value are indicated within each plot. Regression lines are shown so as to provide a qualitative indication of how much the location of incremental M waves (i.e., in their barycenter) changed with increases in stimulation intensity. Only gray circles were considered for the correlation and regression analyses.

Fig. 7.

M-wave amplitude in different gastrocnemius regions. Raw M waves detected from the proximal [channel (Ch) 2], central (Ch 6), and distal (Ch 10) portions of the medial gastrocnemius muscle are shown separately for the 10 stimulation amplitudes. Data from 2 representative participants are shown (subject 8, A; subject 2, B). The RMS amplitude of M waves detected by each channel is shown at bottom. Note that while the amplitude of M waves detected from different portions for subject 8 changes equally with stimulation amplitude, for subject 2 the increase in M-wave amplitude with stimulation level is strongly dependent on the muscle region from which M waves are detected.