P27-F Point estimation and uninterrupted monitoring of resting motor threshold by threshold-tracking transcranial magnetic stimulation

Abstract

Conventionally, RMT is defined as stimulus intensity eliciting a response in 50% of consecutive trials and probabilistic approaches are commonly used for its estimation. Alternatively, threshold can be defined as stimulation intensity required to maintain a response of a specific size and obtained by threshold-tracking. Here, we compare threshold-tracking to well-established RMT estimation methods. RMT with conventional cut-off value of 0.05 mV was measured in 24 healthy volunteers (11 men; median age 22 (range 18–55) years; all self-reported right-handed) from the dominant first dorsal interosseous muscle using three. Methods (i) Relative frequency (RF, 10/20 positive trial rule), (ii) adaptive maximum-likelihood procedure ‘best PEST’ (Parameter Estimation by Sequential Testing), and (iii) threshold-tracking (TT). Measurements were repeated twice by a single operator on the same day. Data is presented as mean ± standard deviation. No difference in mean group RMT estimates was observed between the methods (RF 50.6 ± 8.3, PEST 50.3 ± 8.2, TT 50.9 ± 9.1% maximum stimulator output (MSO); rmANOVA, p = 0.228), but the duration of procedure differed considerably (number of stimuli: RF 60 ± 26, PEST 19 ± 2, TT 12 ± 4; rmANOVA, p In conclusion, all methods provide the same value of RMT, but threshold-tracking offers an improved speed without fundamentally compromising reliability of repeated measurements. While probabilistic methods provide point estimates only, threshold-tracking allows uninterrupted monitoring of RMT and therefore opens new avenues in TMS research and its clinical application.