Background: The use of electroencephalogram (EEG) monitoring devices for assessing the depth of hypnosis is most difficult in children under 5 years of age.
Methods: Forty five children aged 0-60 months were included in a prospective observational study. A direct comparison of the processed EEG variables Bispectral Index (BIS, version 3.4) and Narcotrend Index (NI, version 2.0AF) was to be achieved by simultaneous recording. The ability of these parameters to differentiate between various clinical states was evaluated by using the prediction probability (P(k)). Age-related effects on the BIS and NI were analyzed by dividing the children into three age groups: 0-6, 7-18 and 19-60 months.
Results: The preanesthesia, conscious children were differentiated from anesthetized patients by the BIS and NI with no overlap (P(k) = 1.0). In the awake period the BIS was superior to the NI (P(k) to differentiate 'end of anesthesia' from 'awakening' was 0.97 vs 0.73 respectively; P = 0.002). Patients aged 7-18 months showed higher BIS and NI values in the course of anesthesia than the younger and older children (P = 0.001). On awakening, children aged 0-6 months showed the lowest mean BIS (n.s.) and NI (P = 0.006) values.
Conclusions: The BIS currently seems to be superior to the NI, but age-related processing algorithms of the raw EEG must be implemented in both BIS and NI in order to be useful in children younger than 5 years of age.