Abstract
Percutaneous electrical stimulation over tibialis anterior and triceps surae was performed in 14 patients with traumatic SCI to look for evidence that increased force can develop, beyond that due to activation of the motor axons beneath the stimulating electrodes. Criteria for the increased force included marked asymmetry of force with respect to stimulation, progressively rising force during stimulation of constant amplitude and frequency, and force remaining high after stimulation frequency had returned to the control level following a high-frequency burst. Twelve of the patients showed evidence of such behaviour, more frequently in triceps surae than tibialis anterior. There was no apparent correlation between the completeness or level of injury and the ability to induce the behaviour. Evidence of cyclic force potentiation and habituation was also seen. Eleven of the patients exhibited hyper-reflexia and reported spontaneous spasms, but there was no obvious association with the increased force. It is concluded that the neurones within the spinal cord can contribute to the extra force by operating in a sustained firing mode associated with plateau potentials.
1 Introduction
FES potentially offers a solution for restoring function in patients suffering paralysis in conditions such as SCI. Currently its practical application is less promising due to the related problems of muscle fatigue, habituation and low muscle force output [1]. The spinal cord contains complex circuitry for the control of movement and operates in a different way to produce muscle contraction, potentially obviating the problems mentioned above. Voluntary movement activates small, fatigue-resistant motor units first, with larger units activated as more force is required [2]. In contrast, conventional FES activates initially large diameter motor axons that innervate large, more fatiguable muscle fibres. Smaller, fatigue-resistant muscle fibres are only recruited at higher stimulus levels. When motoneurones are recruited via dendritic synapses at a sufficiently high rate and level of stimulus, the neurone may change to a sustained firing mode, characterised by a persistent inward current across the cell membrane [3]. This sustained firing mode may persist for a few seconds, if the synaptic input is maintained, and is characterised by a high firing-frequency to input-stimulus gain ratio [4]. To take advantage of these spinal cord properties, FES needs to activate Ia afferents and to do so in such a way as to drive motoneurones into their sustained firing mode. This can be done by using wide stimulating pulses, to excite Ia afferents preferentially, and to use high frequency stimulation to provide sufficient synaptic excitation [5]. Much of the synaptic excitation comes from higher centres and is lost following SCI. One of the objectives of this study was to determine if sufficient excitation could be given to patients with SCI so as to trigger motoneurone sustained firing. Potentially the additional force provided by the combination of sustained firing and recruitment of additional motoneurones via Ia afferents, may solve the present problems with conventional FES.
2 Methods
14 patients (11 male, 5 clinically complete)
reclined in a comfortable chair with one foot strapped to a plate that was
hinged co-axially with the ankle joint (Figure 1). A force transducer on the
plate recorded the force produced by stimulation of the triceps surae (TS) or
tibialis anterior (TA) using 6 or 7s trains containing 1-ms duration
constant-amplitude pulses. Three types of test trains were given: frequency
increasing linearly from 4 to 100Hz over 3s then decreasing to 4Hz; 25Hz for
2s, 100Hz for 2s, then 25Hz for 3s; 100Hz for 7s. In addition trains of 25Hz
for 7s were used as a control. Stimulus amplitude was set so that a train of 5
pulses at 100Hz produced a force 6 - 10% of the maximal stimulated contraction
force (MSC) using this train frequency and duration.
Figure 1:
Experimental method
Based on previous studies in able-bodied subjects, the following criteria were judged to indicate the presence of neurone sustained firing: with triangle stimulation, there was marked asymmetry in the force response such that forces were relatively larger when the frequency was declining; with superimposed burst, the force was greater after the 100Hz burst than before; with long trains, force increased with time [6][7].
3 Results
Evidence of neurone sustained firing was seen in 12 of the 14 patients in a manner qualitatively similar to that seen in able-bodied subjects [6][7] and reported in greater detail elsewhere [8]. Typical responses are shown in Figure 2 for one patient with T4 incomplete injury. Triangle stimulation (upper panels) showed asymmetry of the force, with greater force as stimulus frequency was decreasing than when it was increasing. With superimposed 100Hz burst stimulation (middle panels), the force was greater after the 100Hz burst than before it. When a constant 100Hz stimulus was used (bottom panels), the force continued to increase throughout the stimulation.
Evidence of sustained firing was found more frequently in TS than TA (P < 0.05). No one stimulation pattern was more efficacious than the others. There was no significant correlation between injury level or completeness, nor the frequency of spasms and the frequency of evidence of sustained neurone firing. Stimulation in some patients resulted in the frequent appearance of contractions in antagonists, either before or after the increased agonist contraction, and contractions in other muscle groups of the same or opposite leg. Both potentiation and habituation of the response was seen with an oscillation between the two, upon repeated stimulation, with a period of ~80s.
Figure 2:
Reproducibility of extra contractions arising from neurone sustained firing.
4 Discussion and Conclusions
This study confirms that the phenomenon of increased muscle force, possibly associated with sustained neurone firing, can be induced in patients with SCI, much as it can in able-bodied people. Although it is difficult to be certain that a lesion is truly complete on clinical grounds, the fact that extra contractions, indicating the possibility of sustained firing, were seen in patients with apparently complete SCI argues against the brain playing an essential role in the generation of these phenomena.
A possible explanation for the increased muscle force is that stimulation over the motor point not only caused an initial muscle contraction due to direct stimulation of motor axons but also directly induced action potentials in afferent fibres. The afferent signals recruited other motoneurones monosynaptically and through interneurones and other spinal neurones. As the integrated excitatory dendritic synaptic input increased, an increasing number of spinal neurones would have been driven from their resting state to a sustained firing state, characterised by plateau potentials [4][9]. Integration of input by the synapses means that the motoneurones would continue to increase their output even as the stimulation decreased, in some cases continuing to fire after stimulation had ceased. Eventually the neurones would spontaneously revert from their sustained firing state to their resting state or be driven back to the resting state by inhibitory synaptic input. It is possible that other spinal neurones could also be operating in a sustained firing mode in addition to, or instead of the motoneurones. Harnessing this central mechanism to generate extra force by FES of muscle may afford several advantages over conventional stimulation techniques that primarily activate the largest motor axons beneath the stimulating electrodes. The central mechanism that we believe to be responsible for the “extra” contraction force in the present study activates muscle fibres asynchronously, with the smallest units recruited first. The contractions should then be more fatigue-resistant. Less battery power would be needed in a portable FES system as not all motor axons need to be stimulated directly. Recruitment of a wider range of muscle fibre sizes may obviate the muscle atrophy that often accompanies chronic SCI.
References
Acknowledgements
We wish to thank Drs Stella Engel and Leanne Hunt of the Prince of Wales and Prince Henry Hospitals, Sydney for their assistance in recruiting subjects for the studies. This work was supported by the National Health and Medical Research Council of Australia (3206) and the Alberta Heritage Foundation for Medical Research (DFC).