ABSTRACT
OBJECTIVE: To determine the influence of different factors on test-retest reliability of frequently used transcranial magnetic stimulation (TMS) parameters while controlling for potential confounders in healthy subjects. METHODS: TMS was applied in 93 healthy volunteers (61% male) twice (mean retest interval of 34.0 ± 25.6 (SD) days) between 7 am and 2 pm by four investigators (sessions n investigator A=47, investigator B=95, investigator C=28, investigator D=16). Women were assessed in their follicular phase. Test stimulus (TS), resting motor threshold (RMT), short latency intracortical inhibition (SICI), intracortical facilitation (ICF) and cortical silent period (SCP) were analyzed. RESULTS: Good test-retest reliabilities were observed for TS (r=.880) and RMT (r=.826), moderate for visual and automated analyzed CSP durations (resp. r=.466, r=.486), and poor for ICF (r=-.159). Reliable change indexes are reported. Gender (e.g. automated CSP women: r=.538 vs. men: r=.422), re-test interval and method of CSP-analysis did not influence reliabilities. CONCLUSIONS: In a large sample of healthy volunteers we found good to moderate test-retest reliabilities in all but one TMS-parameter. Automated analysis of the CSP did not prove to be more reliable than visual determination. SIGNIFICANCE: This study contains analyses of re-test reliability in TMS considering several confounding factors. For the first time it presents reliable change indices for all frequently used TMS parameters.
