A study of the Bispectral Index Monitor during procedural sedation and analgesia in the emergency department.

STUDY OBJECTIVE: The Bispectral Index Monitor has been validated as an objective measure of sedation depth in the operating room; however, its value for states other than general anesthesia remains unclear. We hypothesized that bispectral index monitoring would reliably correlate with traditional definitions of sedation depth in emergency department patients undergoing procedural sedation and analgesia.

METHODS: In this prospective observational study we measured the Bispectral Index score and a modified Ramsay Sedation Scale score every 5 minutes in a convenience sample of adult ED patients undergoing procedural sedation and analgesia. Investigators and treating physicians were blinded to Bispectral Index scores, which were later correlated with modified Ramsay Sedation Scale scores. We constructed receiver operating characteristic curves to determine the most discriminatory Bispectral Index score thresholds for sedation depth.

RESULTS: Two hundred seventy paired readings were obtained from 37 patients. Despite being statistically significant (P <.0005), the correlation between the Bispectral Index Monitor and modified Ramsay Sedation Scale scores was only moderate (Spearman rho=-0.690) and displayed wide variability. A modified Ramsay Sedation Scale score consistent with mild-to-moderate sedation corresponded to Bispectral Index scores of 34 (general anesthesia) to 98 (fully awake), whereas a modified Ramsay Sedation Scale score consistent with deep sedation corresponded to Bispectral Index scores of 40 to 98. Areas under receiver operating characteristic curves demonstrated moderate (0.83 to 0.86) discriminatory power at all modified Ramsay Sedation Scale thresholds, with the exception of a modified Ramsay Sedation Scale score of 7 or less, which demonstrated high discriminatory power (0.95). A Bispectral Index score of greater than 83 reliably excluded general anesthesia, and a Bispectral Index score of 70 or less reliably identified general anesthesia.

CONCLUSION: Bispectral index monitoring reliably predicted patients undergoing procedural sedation and analgesia who were sedated to the point of general anesthesia from those with lesser degrees of sedation but did not discriminate mild-to-moderate sedation or moderate-to-deep sedation, as measured by the Ramsay Sedation Scale score for the patients undergoing procedural sedation and analgesia in our ED.