Is there sufficient evidence to claim muscle units are not localised and functionally grouped within the human gastrocnemius?

Similar results, conflicting interpretations and equivocal implications: this is the issue triggered by the recent, interesting study published by Héroux et al. (2015).

To identify the region confining the majority of the muscle fibres of single motor units, we triggered and averaged surface EMGs detected along the proximo‐distal gastrocnemius axis with the firing instants of motor units identified from intramuscular recordings (Vieira et al. 2011). We concluded the fibres of individual units are spatially localised in gastrocnemius. Héroux et al. (2015), on the other hand, used the firing instants of motor units identified from intramuscular recordings to trigger multi‐unit, intramuscular EMGs detected from different, proximo‐distal gastrocnemius regions. They concluded the muscle fibres of single units are not localised in gastrocnemius. In this letter we present arguments that current and previous results are not in conflict or disagreement with spatially localised and functionally grouped muscle units in the human gastrocnemius.

The first crucial point to consider is the fact that regional activation has been recorded in both intramuscular and surface EMG from the gastrocnemii. What could be the cause of such a localised activity pattern? It is well established that (i) the presence of a motor unit action potential in the recorded EMG indicates its parent fibres are located near the recording electrode, be it on the skin surface or in the muscle (Lynn et al. 1978; Roeleveld et al. 1997) – the closer the action potential is to the pick‐up source, the larger it appears in the EMG (Lynn et al. 1978) – and (ii) fascicles in the gastrocnemius muscle are pinnate (angled along deep‐superficial axis). Differently from other muscles (e.g. biceps brachii), in which action potentials travel from end‐plates to the muscle origin and insertion, gastrocnemius action potentials travel toward the deep and superficial aponeuroses. Consequently, the distance between the action potential and the skin surface decreases progressively with propagation, being minimal when the action potentials reach the muscle superficial aponeurosis. For these two reasons, the action potential propagating along any gastrocnemius fibre appears differently in EMGs detected by small surface electrodes, positioned serially and with small centre‐to‐centre distance over the gastrocnemius superficial aponeurosis; action potentials appear well defined in EMGs detected by electrodes located in proximity of the superficial attachment of fibres along which they propagate (Mesin et al. 2011). Contrary to the view of Héroux and colleagues, the position of active fibres in relation to surface electrodes – rather than regional differences in pinnation angle and subcutaneous tissue, as well as the position of innervation zones and tendinous regions – seems the main factor accounting for regional variations in the amplitude of action potentials. To our knowledge, there is no evidence available for questioning the ability of surface EMGs to capture the distribution of active fibres within the gastrocnemius volume.

A second, central issue is the distinction between spatial localisation and small motor unit territory. Conceptually, the territory of a motor unit is defined by the perimeter of the physiological cross sectional area confining all its muscle fibres. How fibres distribute within the territory and not the territory size itself is, however, the key issue determining spatial localisation. Indeed, regardless of how large the territory of a given motor unit may be, muscle fibres may group locally within the territory. From our experimental intramuscular and surface EMGs we observed the action potential of individual motor units was represented at a skin region spanning about 4 cm of the gastrocnemius proximo‐distal axis (Vieira et al. 2011). In view of the large gastrocnemius length (∼25 cm; Narici et al. 1996), and given our arguments above, we therefore believe the spatial localisation of individual muscle units is the most likely interpretation for spatially localised action potentials in surface EMGs. Most importantly, it must be noted our findings and interpretation are not in conflict with the notion of spatially localised muscle units and of motor units with large territories; both may coexist.

Bearing these two points in mind, the elegant data presented by Héroux et al. (2015) are not in disagreement with the notion of spatially localised muscle units in gastrocnemius. These authors recorded EMGs with fine‐wire electrodes inserted into seven equally spaced sites, spanning most of the gastrocnemius proximo‐distal axis. In at least four consecutive detection sites, the amplitude of action potentials of about half of the motor units identified were greater than the noise level, leading the authors to state gastrocnemius muscle units are not spatially localised. Even though the results presented by Héroux et al. (2015) suggest the territory of single motor units may span proximo‐distal regions as large as the gastrocnemius length, they did not access the distribution of fibres within the territory. Interestingly, in relation to the noise level, the amplitude of action potentials of individual motor units was different across detection sites; potentials with greater amplitude seem to have been detected locally, i.e. by adjacent recording sites (cf. their Fig. 3). Supposing the pick‐up volume of their multi‐unit, fine‐wire recording was similar across sites, such relative difference in amplitude may indeed suggest the fibres of gastrocnemius motor units do not distribute equally across the muscle.

Having reviewed potential reasons for different conclusions, the question remains as to which of the two conclusions describes more accurately the function of gastrocnemius motor units. The human gastrocnemius contributes to a wide number of locomotor and postural tasks, and hence responds to a wide range of functional, mechanical demands. To fully explore the task dependence of motor unit activation therefore requires an experimental design that samples from a physiological range of tasks. In previous work, cycling has offered a valuable paradigm for the manipulation of both load (resistance) and velocity (cadence) demands placed on the neuromuscular system. Within this context, the amplitude of surface EMG recordings from different quadrants within gastrocnemii has shown a little over half of their activity patterns are in common (Wakeling, 2009). This evidence is consistent with Héroux et al. (2015), where 32 out of 69 motor units spanned at least 6.9 cm, reported as being half the muscle length. Critically, however, Wakeling (2009) found manipulating the mechanical demands of the task resulted in considerable regional differences in EMG amplitude in the lateral gastrocnemius (an anatomically compartmentalised muscle; Segal et al. 1991), and measureable proximo‐distal differences in the medial gastrocnemius, which were largely speed dependent (Wakeling, 2009). The detailed work presented by Héroux et al. (2015) clearly provides good insight into the behaviour of small populations of motor units along medial gastrocnemius during relatively low level isometric contractions. Employing intramuscular EMG to study individual motor units does not, however, enable the study of a spectrum of functionally significant, mechanically demanding tasks, nor does it allow the full spectrum of motor units to be included in the evaluation (i.e. torques up to 38% MVC will likely not result in activation of higher threshold units). This means the results should not be generalised to draw conclusions regarding the functional significance of spatial localisation.

Cautiously shaping interpretations is a mandatory exercise when reporting EMGs, regardless of whether they were collected with intramuscular or surface electrodes. To date, and according to the arguments presented in this letter, we believe no evidence can be advanced contradicting the localised distribution of fibres of single muscle units within the human gastrocnemius.

Additional information

Competing interests

The authors declare no conflicts of interest.

Author contributions

Conception and design: T.M.V., J.W. and E.F.H. Drafting the article or revising it critically for important intellectual content: T.M.V., J.W. and E.F.H. All authors read and approved the final version of the manuscript and all authors listed qualify for authorship.

Funding

This study was supported by a national research project funded by the Italian Ministry of Education, Universities and Research (Protocol number: 2010R277FT), and co‐funded by Compagnia di San Paolo and Fondazione C.R.T.