Performance of the Bispectral Index During Electrocautery

Background: The electroencephalogram contains small electrical signals that are vulnerable to contamination from high-frequency noise during electrocautery. The bispectral index (BIS) monitor incorporated hardware and software changes to eliminate artifacts, thus allowing BIS monitoring even in the presence of electrocautery. We evaluated the accuracy of BIS to measure anesthetic effect during electrocautery interference. Methods: Anesthesia was induced and maintained with target-controlled infusions of propofol (3 mu g/mL) and remifentanil (4 ng/mL). After baseline BIS recordings, "simulated" electrocautery interference was induced continuously for 20 minutes. Five minutes after the start of electrocautery, propofol infusion was increased to achieve an effect site concentration of 6 mu g/mL. Patients remained undisturbed during the study. BIS values and signal quality index were recorded continuously. Results: During electrocautery, there was a significant decrease in signal quality index (mean difference: 16.9; 95% confidence intervals: 15.9-17.9; P<0.001). There was, however, no change in BIS value even after a step increase in propofol infusion from 3 to 6 mu g/mL (P= 0.93). In 22% of the patients there was a paradoxical increase in BIS values after doubling of propofol concentration. Following cessation of electrocautery, there was a prompt decrease in BIS (P< 0.001), indicating a lack of response to the change in anesthetic depth during electrocautery. Conclusions: Rejecting and filtering artifacts from electrocautery interference reduced the ability of BIS to respond to a change in anesthetic depth. BIS values during electrocautery should be interpreted with caution.