Alternative electrode placement in (automatic) sleep scoring (Fpz-Cz/Pz-Oz versus C4-A1)

The purpose of this study is to investigate whether the international standard electrode placement (C4-A1) can be replaced by an alternative placement (Fpz-Cz/Pz-Oz) in an automatic sleep monitoring system without losing Rechtschaffen and Kales (R-K) balances. Single night-sleep polygraphic recordings of 10 patients, screened in a clinical sleep disorder setting, were recorded simultaneously with both placements, and visual sleep classification was performed separately by two independent observers. Interobserver and interplacement agreement were evaluated by way of average (dis)agreement matrices and kappa values computed for overall and individual stage scoring. Interobserver agreement for both the test and the standard electrode placements and interplacement agreement for both observers were assessed as fair to good or excellent. Scoring differences were evaluated by the rank sign test applied to clinical and theoretical difference scores. It appears that the interplacement differences are about equal to the interobserver differences, except for a slight tendency for sleep to be scored in a deeper stage with the proposed alternative placement. The data are presented and discussed in relation to current literature concepts.