Objectives: Manual methods of measuring duration of cortical silent periods (CSP) evoked by transcranial magnetic stimulation (TMS) depend upon subjective visual estimation of onset and offset. Because of this, the measurements are susceptible to poor rater reliability. We describe a graphical method to measure silent periods with greater precision. The statistical process underlying this new method is simple and particularly suited to signal detection in serially dependent data.
Methods: TMS-evoked silent periods were recorded in 13 healthy subjects. Two investigators subjectively measured silent period duration on each subject to estimate rater reliability. Using the graphical method, the mean and 99.76% variation limits of pre-stimulus electromyogram (EMG) activity were computed. Each averaged trial was displayed and CSP onset and offset detected when post-stimulus EMG activity moved outside the 99.76% limits.
Results: Maximum variation in silent period duration was 21.8 ms between the two investigators' subjective measurements. Silent period duration measured with the graphical method closely approximated measurements obtained using the manual method. It was possible to automate the procedure.
Conclusions: This graphical method allowed precise measurement of CSP duration, independent of subjective estimations of onset or offset points. Further studies are necessary to determine if this method can provide a framework for other physiologic measures.