Assessing the Intra- and Inter-Subject Reliability of the Perturbational Complexity Index (PCI) of Consciousness for Three Brain Regions Using TMS-EEG

Abstract

BackgroundThe perturbational complexity index (PCI) is a useful measure of consciousness that combines transcranial magnetic stimulation (TMS) with electroencephalography (EEG). However, the PCI has not been assessed for reliability between sessions nor is there a clear best stimulation target to acquire a PCI. Objective/HypothesisWe assessed the reliability of within-subject PCIs between 3 sessions with stimulation over the same premotor, motor, and parietal targets between visits, hypothesizing that we could determine a most reliable TMS-EEG target to acquire PCIs in healthy, conscious adults. MethodsPCIs were acquired for 9 participants (5 women) over 3 sessions within a week. A 64 channel EEG system was used with all electrode impedances [≤] 10k{Omega}. Premotor, motor, and parietal stimulation targets were identified using a real-time Matlab graphical user interface (GUI). Neuronavigation using an MRI-template brain ensured that every TMS pulse was delivered within 3.0mm and 5{degrees} of the stimulation target. ResultsPremotor, motor, and parietal PCIs all had significant PCIs (all p < 0.05). However, parietal and motor PCIs had excellent reliability (ICCs = 0.927 and 0.857 respectively) whereas premotor PCIs had good reliability (ICC = 0.737). PCIs were similar between brain sites within each subject in a single visit, but with only a moderate effect (p = 0.024, ICC = 0.480). PCIs on a group level did not differ between brain sites (p = 0.589). ConclusionsThe PCI is a reliable measure over this timeframe within each subject for single brain targets. PCIs for parietal and motor sites are most similar between visits. Due to only moderate similarity between PCIs from three brain sites within each session, PCIs should be acquired over at least two brain sites, with parietal and motor regions as the top candidates.