The distribution and reliability of TMS-evoked short- and long-latency afferent interactions

Abstract

Short-latency afferent inhibition (SAI) and long-latency afferent inhibition (LAI) occur when the motor evoked potential (MEP) elicited by transcranial magnetic stimulation (TMS) is reduced by the delivery of a preceding peripheral nerve stimulus. The intra-individual variability in SAI and LAI is considerable, and the influence of sample demographics (e.g., age and biological sex) and testing context (e.g., time of day) is not clear. There are also no established normative values for these measures, and their reliability varies from study-to-study. To address these issues and facilitate the interpretation of SAI and LAI research, we pooled data from studies published by our lab between 2014 and 2020 and performed several retrospective analyses. Patterns in the depth of inhibition with respect to age, biological sex and time of testing were investigated, and the relative reliability of measurements from studies with repeated baseline SAI and LAI assessments was examined. Normative SAI and LAI values with respect to the mean and standard deviation were also calculated. Our data show no relationship between the depth of inhibition for SAI and LAI with either time of day or age. Further, there was no significant difference in SAI or LAI between males and females. Intra-class correlation coefficients (ICC) for repeated measurements of SAI and LAI ranged from moderate (ICC = 0.526) to strong (ICC = 0.881). The mean value of SAI was 0.71 ± 0.27 and the mean value of LAI was 0.61 ± 0.34. This retrospective study provides normative values, reliability estimates, and an exploration of demographic and testing influences on these measures as assessed in our lab. To further facilitate the interpretation of SAI and LAI data, similar studies should be performed by other labs that use these measures.

Investigating the intra-session reliability of short and long latency afferent inhibition

“…Given that ICC values are a function of the between-subject variability in a sample, with large amounts of between-subject variability leading to higher estimates of relative reliability, this likely explains the trends in reliability seen across the literature and in the present study ( Bruton et al, 2000 , Weir, 2005 ). We recently published a large retrospective analysis of SAI and LAI data which also showed larger amounts of between-subject variability for LAI compared to SAI, with a plausible reasoning being the different neurological pathway that LAI may traverse ( Toepp et al, 2021 ). We theorize that because LAI is evoked at longer ISIs, there is possible activation of additional brain regions including the basal-ganglio-thalamocortical loop, the posterior parietal cortex, and the secondary somatosensory cortex ( Toepp et al, 2021 ).…”

“…We recently published a large retrospective analysis of SAI and LAI data which also showed larger amounts of between-subject variability for LAI compared to SAI, with a plausible reasoning being the different neurological pathway that LAI may traverse ( Toepp et al, 2021 ). We theorize that because LAI is evoked at longer ISIs, there is possible activation of additional brain regions including the basal-ganglio-thalamocortical loop, the posterior parietal cortex, and the secondary somatosensory cortex ( Toepp et al, 2021 ). The potential activation of these various regions introduces several avenues of variability between individuals, which is not seen in SAI.…”

“…Importantly, despite their emerging use in the sensorimotor research, the reliability of afferent inhibition ranges from low to strong in most cases ( Brown et al, 2017 , Toepp et al, 2021 , Turco et al, 2019 ). As such, it is difficult to interpret changes in afferent inhibition within an experimental setting as real (i.e.…”

“…Previous work has only explored the intersession reliability of SAI and LAI, with the three studies reporting the relative reliability of both measures as ranging from low to strong ( Brown et al, 2017 , Toepp et al, 2021 , Turco et al, 2019 ). Even more limited are quantifications of the absolute reliability of these measures, with a single study reporting low absolute reliability as indicated by high SEMeas and high SDC values, indicating large deviations in the measure are needed in order to be significantly different from measurement error ( Turco et al, 2019 ).…”

Experimental environment improves the reliability of short-latency afferent inhibition

“…MEPs and resting motor threshold have demonstrated good test re-test reliability in contrast to measures of intracortical facilitation [ 36 ]. Researchers have also explored the reliability of afferent inhibition and concluded that SAI and LAI demonstrate poor to moderate reliability [ 37 , 38 ]. The relative reliability of afferent inhibition measures across studies also seem to vary substantially, indicating that there may be methodological factors including coil placement, electrode placement [ 39 ] that are modulating reliability [ 40 ].…”