Serial Bispectral index scores in patients undergoing observation for sedative overdose in the emergency department

Study Objective: Many patients who overdose on sedatives experience a declining mental status and eventually require endotracheal intubation. The goal of this study was to determine if serial bedside Bispectral index (BIS) scores monitoring can be used to detect the eventual need for intubation in overdosed patients who are undergoing observation in the ED. Methods: This was a prospective, observational study of a convenience sample of patients who presented to the Hennepin County Medical Center ED between June and November 2002. Patients being treated and observed for a suspected sedative ingestion were eligible. Upon presentation, a Bispectral electroencephalographic probe was applied to the patient's forehead, and a BIS score was recorded at 0 and 20 minutes. The Altered Mental Status scale was used to describe the patient's clinical status. Data were collected by trained research assistants. Data are described with descriptive statistics. The mean changes in BIS score between patients who did and did not require intubation are compared with t tests, and the outcome of patients with stable vs declining BIS scores were compared with χ² tests. Results: Seventy-six patients were enrolled. The mean initial BIS score was 83.9 (95% CI, 79.7-88.1; range, 9-99). The mean change in BIS scores during the 20-minute observation period for the patients who required intubation was -13.5 (95% CI, -30.2 to 3.2) and was +6.7 (95% CI, 3.3-10.1) for those who were not intubated. Sixteen patients had an initial BIS score below 70. Of these patients, 6 were intubated. All intubations occurred during the 20 minutes, and this group had a mean initial BIS of 47.2 (95% CI, 35.6-58.8). The 10 patients with an initial BIS below 70 who were not intubated had a mean increase in BIS score of 23.3 (95% CI, 11.7-33.9) during the 20 minutes. Of the 60 patients whose first BIS score was above 70, 5 were eventually intubated during their ED treatment. The mean change in BIS was -36.4 (95% CI, -18.7 to -54.1) for the intubated patients vs +7.9 (95% CI, 4.4-11.3) for nonintubated patients during the first 20 minutes. Conclusion: The overdosed patients who required intubation during their ED treatment experienced a mean decrease in BIS during the first 20 minutes, compared with those who did not. Bispectral index scores monitoring may prove useful for earlier ED treatment and decision making regarding sedative overdose patients.