Comparison of Rossini–Rothwell and adaptive threshold-hunting methods on the stability of TMS induced motor evoked potentials amplitudes

Several methods can be used to determine the resting motor threshold (RMT) and by that recording transcranial magnetic stimulation (TMS) induced motor evoked potentials (MEPs). However, no research has compared the test retest reliability of these methods. Thus, the aim of this study was to determine intra- and inter-session reliability of Rossini–Rothwell (R–R) and parameter estimation by sequential testing (PEST) methods on TMS-induced MEPs and comparison of these two methods on RMT. Twelve healthy individuals participated in this study three times (T1, T2 and T3) over two days. TMS was applied using both R–R and PEST to estimate RMT and average of 25 MEPs were acquired at each of the three time points. The intra-class correlation coefficient indicated high intra-session reliability in the MEP amplitudes for both methods (0.79 and 0.88, R–R and PEST respectively). The RMT and MEP amplitudes had higher inter-session reliability in both methods (0.99 and 0.998, R–R and PEST respectively; 0.84 and 0.76, R–R and PEST respectively). There was no significant difference between methods for RMT at both T1 (maximum stimulator output of R–R vs. PEST, 33.7% ± 7.7% vs. 33.8% ± 7.6%, p = 0.75) and T3 (maximum stimulator output of R–R vs. PEST, 33.5% ± 7.3% vs. 33.7% ± 7.3%, p = 0.19). There was a significant positive correlation between the methods' estimates of RMT, with PEST requiring significantly fewer stimuli. This study shows that the R–R and PEST methods have high intra-and inter-session reliability and the same precision, with PEST having the advantage over R–R in speed of estimation of RMT.