Use of bispectral index to monitor the depth of sedation in mechanically ventilated patients in the prehospital setting.

Background: Sedative drug administration is a challenging aspect of the management of mechanically ventilated patients in the out-of-hospital critical care medicine. We hypothesised that the bispectral index of the EEG (BIS) could be a helpful tool in evaluating the depth of sedation in this difficult environment. The main objective of the present study was to assess the agreement of BIS with the clinical scales in the out-of-hospital setting.

Methods: This prospective study included mechanically ventilated patients. BIS values were blindly recorded continuously. A Ramsay score was performed every 5 min. The main judgement criterion was the correlation between BIS values and the Ramsay score.

Results: 72 patients were included, mostly presenting with toxic coma (36%) or neurological coma (21%). The median (IQR) BIS value was 85 (84-86) when the Ramsay score was 3, 80 (76-84) when the Ramsay score was 4, 61 (55-80) when the Ramsay score was 5 and 45 (38-60) when the Ramsay score was 6. According to Receiver operating characteristic (ROC) curves, BIS was categorised into three classes (BIS<54 corresponding to Ramsay score 6, 54≤BIS<72 for Ramsay score 5 and BIS≥73 for Ramsay score ≤4). Based on these categories, the proportion of appropriate BIS values was 67% (217/323). The concordance correlation coefficient for repeated measurements was 0.54 (0.43-0.64). The agreement between BIS and the Ramsay score is moderate.

Conclusions: Prehospital measured BIS values appear poorly correlated with clinical assessment of the depth of sedation. For this reason, the use of BIS to guide prehospital sedation cannot be recommended.