Background: Transcranial magnetic stimulation (TMS) is widely used to probe corticospinal excitability through Motor Evoked Potential (MEP) amplitude measurements. The input-output (I/O) curve is a sigmoid-shaped relation between the MEP amplitude at incremented TMS intensities. The aim of this study was to examine the relationships between seven parameters derived from the sigmoid function. Methods: Principal Component Analysis and Spearman's rank correlation matrices were used to determine if the seven I/O curve parameters capture similar or, conversely, different aspects of the corticospinal excitability in 24 healthy subjects and 40 stroke survivors with a hand motor impairment. Results: Maximum amplitude (MEPmax), peak slope, area under the I/O curve (AUC), and MEP amplitude recorded at 140% of the resting motor threshold showed strong linear relationships with each other (ρ > 0.72, p < 0.001). Results were found to be similar in healthy subjects and in both hemispheres of stroke patients. Our results did not support an added benefit of sampling entire I/O curves in both healthy subjects and stroke patients, with the exception of S50, the stimulus intensity needed to obtain half of MEPmax amplitude. Conclusions: This demonstrates that MEP elicited at a single stimulus intensity allows to capture the same characteristics of the corticospinal excitability as measured by the AUC, MEPmax and the peak slope, which may be of interest in both clinical and research settings. However, it is still necessary to plot I/O curves if an effect or a difference is expected at S50.
Keywords: corticospinal excitability; input-output curve; motor evoked potentials; stroke; transcranial magnetic stimulation.