Comparing the bispectral index and suppression ratio with burst suppression of the electroencephalogram during pentobarbital infusions in adult intensive care patients

Pharmacotherapy. 2003 Sep;23(9):1087-93. doi: 10.1592/phco.23.10.1087.32766.

Abstract

Objective: The bispectral index (BIS), a processed variable derived from the raw electroencephalogram (EEG) used to guide sedation in the intensive care unit (ICU), has not been tested during barbiturate therapy for elevated intracranial pressure. We determined the BIS and suppression ratio (SR) values during traditional burst monitoring of the raw EEG during pentobarbital infusions.

Design: Prospective, observational cohort study.

Setting: A 42-bed multidisciplinary ICU in a tertiary care medical center.

Patients: Twelve consecutive patients with elevated intracranial pressure treated with pentobarbital infusions.

Intervention: All patients were monitored continuously with the Aspect Medical Systems A-1050 bedside EEG monitor using a bilateral referential montage. Pentobarbital doses were titrated based on the raw EEG to attain a burst-suppression pattern with a goal of 3-5 bursts/minute. Drug dosage, intracranial pressure, cerebral perfusion pressure values, EEG bursts/minute, BIS version 3.2, and SR were recorded daily.

Measurements and main results: The 12 patients were monitored for 62 patient-days. Mean +/- SD age was 32 +/- 15 years, seven (58%) patients were male, mean Acute Physiology and Chronic Heath Evaluation II score was 17.0 +/- 5.0, and hospital mortality was 42%. The mean pentobarbital infusion rate was 124 +/- 49 mg/hour or 2.3 +/- 1.3 mg/kg/hour, and mean pentobarbital serum concentration was 29.7 +/- 13 microg/ml. The mean BIS value was 18 +/- 14, mean SR 56% +/- 36%; BIS correlated well with SR (r=-0.99, p<0.001). For patient-days with a burst-suppression pattern, BIS 3.2 (r=0.90, p<0.001) and SR (r=-0.89, p<0.001) strongly correlated with the number of bursts/minute. The mean BIS value corresponding to 3-5 bursts/minute was 15 (95% confidence interval [CI] 10-20); SR value was 71 (95% CI 61-80).

Conclusion: The Aspect A-1050 applied to patients and monitored by nurses and physicians works well as a bedside EEG monitor, providing a raw EEG signal to titrate barbiturate therapy. The continuous data trend and real-time digital output for the BIS and SR quantify the degree of EEG suppression well and may prove helpful in facilitating titration of barbiturate infusions.

Publication types

  • Comparative Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Cohort Studies
  • Electroencephalography / drug effects*
  • Electroencephalography / methods
  • Humans
  • Infusions, Intravenous
  • Inpatients
  • Intensive Care Units*
  • Intracranial Hypertension / diagnosis
  • Intracranial Hypertension / drug therapy*
  • Male
  • Middle Aged
  • Pentobarbital / administration & dosage*
  • Pentobarbital / blood*
  • Prospective Studies
  • Status Epilepticus / diagnosis

Substances

  • Pentobarbital