Background

Transcranial Magnetic Stimulation (TMS) can quantify corticospinal tract function by extracting parameters from stimulus-response (recruitment) curves. Monitoring some of these parameters, such as slopes (k), maximum motor evoked potential (MEP) amplitudes (MEP_sat), and intensities at the steepest part of the slope (S₅₀), could help provide more individualised prognoses following neurological injury. However, current TMS protocols for generating recruitment curves are often restricted to measuring only one muscle at a time, via stimulation at the cortical hotspot corresponding to each muscle, making them time-consuming and clinically impractical. The aim of this study was to determine whether parameters derived for two muscles simultaneously from a single cortical location would be consistent, to provide evidence for optimised data acquisition times.

 

Methods

With ethical approval from the Imperial College Research Ethics Committee, thirteen healthy participants (7 male, 6 female; mean ± SD age: 26.54 ± 9.33 years, range 21-52 years) underwent neuronavigation-guided TMS of the motor cortex. Surface electromyography was recorded from the first dorsal interosseus (FDI) and triceps brachii (TRI) muscles. Recruitment curves were generated for both muscles at each hotspot and modelled with a four-parameter Boltzmann sigmoidal function (MEP_sat, EMG_base, S₅₀, k). Within-subject comparisons were carried out between on-hotspot and off-hotspot conditions using paired t-tests or Wilcoxon signed-rank tests. The relationship between Euclidean hotspot distance and the change in each parameter between stimulation conditions was assessed using Spearman’s correlation analysis. Significance was defined as p < 0.05.

Results

For FDI, off-hotspot stimulation increased median S₅₀ from 46.03% (IQR 43.94–60.71) to 72.84% (IQR 63.79–85.36) of maximum stimulator output (W = 87, p = 0.0007, r_c = 0.96), indicating that higher stimulus intensities were required to achieve equivalently sized responses. No significant effect was observed for the other FDI parameters (k: p = 0.84; MEP_sat: p = 0.50), indicating that off-hotspot FDI stimulation preserved the shape of recruitment curves. For TRI, no significant differences were observed between parameters derived from on- or off-hotspot stimulation (S₅₀: p = 0.62; k: p = 0.77; MEP_sat: p = 0.13).

Across participants, the change in FDI S₅₀ between stimulation conditions was positively correlated with the distance between FDI and TRI hotspots (r_s = 0.670, p = 0.015, 95% CI [0.172, 0.896]). No significant correlations were observed for any other FDI or TRI parameters.

 

Conclusion

Boltzmann recruitment curve parameters, particularly MEP_sat and k, were largely robust to some deviation from the optimal stimulation site, i.e. the hotspot. Although off-hotspot stimulation only increased FDI S₅₀, suggesting that muscle-specific variables such as differences in cortical representations may modulate the impact of spatial deviations, the overall preservation of Boltzmann parameters demonstrated that recruitment curves can be derived for two muscles simultaneously from a single cortical location. The findings of this study in healthy individuals support the feasibility of more efficient acquisition protocols and provide proof-of-concept for the assessment of multiple muscles at once. This represents an important first step towards clinically practical application, where rapid characterisation of corticospinal tract function could improve prognostication following neurological injury.