Composite auditory evoked potential index versus bispectral index to estimate the level of sedation in paralyzed critically ill patients: a prospective observational study.

Electromyographic activity (EMG) has been reported to elevate the Bispectral Index (BIS) in patients not receiving neuromuscular blockade while under sedation in the intensive care unit (ICU). We investigated the change of the composite A-line autoregressive index (AAI) and BIS after administration of muscle relaxants in sedated surgical ICU patients. We prospectively investigated 38 patients who required administration of a muscle relaxant while continuously sedated with midazolam hydrochloride and fentanyl citrate to achieve a Ramsay Sedation Scale value equal to 5. BIS, EMG activity of BIS (EMG-BIS), signal quality index of BIS, AAI, EMG activity of AAI (EMG-AAI), and acceleromyography at the adductor pollicis muscle were recorded simultaneously every 5 min for 30 min before and after neuromuscular blockade. Student's t-test, the Wilcoxon's signed ranks test, and the Spearman test were calculated using the standard statistics software SPSS 10.0 (SPSS Inc., Chicago, IL). After administration of a muscle relaxant, BIS (58.61+/-7.45 vs 44.68+/-6.65, P<0.001), EMG-BIS (37.33+/-7.15 vs 27.24+/-1.51, P<0.001), AAI (34.11+/-10.96 vs 15.97+/-6.69, P<0.001), and EMG-AAI (59.58+/-9.57 vs 1.00+/-0.00, P<0.001) decreased significantly. Significant correlations between BIS and EMG-BIS (rs=0.75, P<0.001) and AAI and EMG-AAI (rs=0.87, P<0.001) were also found during the baseline period. This study demonstrated that, in sedated ICU patients, BIS and AAI markedly decreased after administration of myorelaxant, and the decreased BIS and AAI values after neuromuscular blockade were correlated to those usually seen in the state of surgical anesthesia, respectively.