Our data demonstrate that the effect of motor ES (30 Hz, ramped) on corticospinal excitability as measured by TMS, depends on application duration in a sample of healthy subjects. The novel findings are that although 20- and 40-min of motor ES increased corticospinal excitability, 60-min of stimulation had no effect. The similar magnitude of increase in corticospinal excitability between the 20- and 40-min conditions suggests that 20-min of motor ES is sufficient to increase corticospinal excitability.
Previous research examining the effect of motor ES on corticospinal excitability has used frequencies of 1–10 Hz and constant stimulus amplitudes to produce simple muscle twitches [20, 22–26, 33]. Data are conflicting with some suggesting increased corticospinal excitability with 60–120 min of stimulation [20, 22, 26], and others reporting increases with application times as short as 10–30 min [24, 40]. The only study to systematically examine the effect of application time of motor ES (10 Hz) on corticospinal excitability reported the greatest increase in excitability with 45–60 min of stimulation . As motor ES applied with a constant stimulus amplitude at 10 Hz (twitch) and that with a ramped stimulus amplitude at 30 Hz (functional) have differing effects on corticospinal excitability , differences in the effects of stimulation duration are possible. Recent work comparing a 30-min application of 10 Hz and ramped 30-Hz motor ES demonstrated increased corticospinal excitability only for the 30 Hz ramped protocol . This suggests motor ES (30 Hz, ramped) designed to mimic a voluntary contraction can more effectively increase corticospinal excitability with short application durations, consistent with other data from short applications [2, 5, 18].
Similar to previous reports , Mmax amplitude was reduced (indicating fatigues of the peripheral apparatus) in APB following motor ES. This effect was present regardless of ES application time. To account for these peripheral changes, MEPs were expressed relative to Mmax in the current study. As MEPs increased with 20- and 40-min of motor ES, despite a reduction in Mmax, increased MEP amplitudes following these interventions can be attributed to excitabtility changes at the corticospinal level. However, one consideration is whether changes in corticospinal excitability following ES occur at the motor cortex or spinal motoneurones. Although not tested here, previous research has demonstrated that H-reflexes [41, 42], F-waves  and cervicomedullary evoked potentials [19, 43] are unchanged following peripheral ES. As these techniques probe motoneurone excitability, it is suggested that changes induced by ES are most likely to occur at the cortex. Several mechanisms are thought to underlie plastic change in the motor cortex following motor ES. These include unmasking of silent synaptic connections and long term potentiation (LTP) of synaptic efficacy [28–31].
Why application of ramped motor ES at 30 Hz for 60-min did not increase corticospinal excitability is unclear. One possible explanation is that time-dependent homeostatic plasticity mechanisms acted to prevent destabilisation of the nervous system and maintain neural activity within a specific range [44–47]. The long-term potentiation (LTP) and long-term depression (LTD) of synaptic efficacy, that are thought to underlie increased or decreased corticospinal excitability during ES applications, operate via a positive feedback mechanism . If large increases in corticospinal excitability are induced by motor ES the potential exists for runaway excitability and destabilisation of cortical neuronal networks . To ensure neural activity is maintained within a stable, physiological range homeostatic plasticity adjusts the threshold for synaptic modifications based on the history of neuronal activity [44–47]. A history of high activity biases synaptic modifications towards LTD (linked to decreased corticospinal excitability), and a history of low activity biases synapses towards LTP (linked to increased corticospinal excitability) [46, 49, 50]. In the current study it is possible that the first half of the 60 minute motor ES application induced an increase in corticospinal excitability sufficient to be interpreted by the system as “high activity”. This would trigger homeostatic plasticity and reduce or reverse the effect on corticospinal excitability towards that of depression. Support for this theory is drawn from a recent study by Gamboa and colleagues , using prolonged theta burst stimulation. Similar to our findings, a short stimulation period resulted in LTP and increased corticospinal excitability while a prolonged period of stimulation resulted in a reversal of the response towards LTD and decreased corticospinal excitability . Taken together with the results of the current study, these findings suggest that longer periods of stimulation have the potential to invoke homeostatic plasticity mechanisms, reducing the effectiveness of the intervention. This novel interpretation has been overlooked in previous ES work.
Alternatively, the difference in effect of shorter and longer durations of stimulation may be explained by difference in the ability of subjects to maintain attention to the stimulus. Attention to the stimulation and contraction may modulate the effect of the intervention on corticospinal excitability [51–54]. Despite instruction and reminders to focus on the stimulation every 5 min, attention may have been less in the 60-min protocol leading to a decreased response.
This study is the first to examine the effect of duration of motor ES (30 Hz, ramped) on corticospinal excitability. Our finding that 20-min of motor ES (30 Hz, ramped) is sufficient to produce a significant increase in corticospinal excitability that lasts at least 20-min supports the use of shorter stimulation periods in rehabilitation settings. Such duration of application is likely to be easier to administer and more efficient. The period of increased corticospinal excitability is likely to provide therapists with a window of opportunity to assist patients to learn novel tasks and aid skill acquisition [2, 5, 19–21]. However, further studies examining the link between increased corticospinal excitability, motor ES and learning are required before conclusions regarding potential clinical application can be made.
This study has few limitations, however, two issues should be mentioned. One is the relatively short follow-up time of 20 min. Studies of 1–10 Hz motor ES report increased corticospinal excitability for up to 120 min after cessation of stimulation . We are not aware of any evidence that temporal changes in corticospinal excitability depend on duration of motor ES. Further work should investigate this question. Second, this study used a relatively small sample size of 14 healthy subjects. Small sample sizes are common in TMS studies due to the novel and explorative aspect of TMS research. Caution must be exercised when interpreting these findings and extrapolating to the wider population. In addition, further testing of the duration of motor ES paradigms requires in-depth exploration on subjects with neurological pathology before clinical recommendations can be made.