Can the Bispectral Index Be Used to Predict Seizure Time and Awakening After Electroconvulsive Therapy? Paul F. White, PhD, MD, FANZCA*, Shivani Rawal, MD*, Alejandro Recart, MD*, Larry Thornton, MD†, Mark Litle, MD†, and Louis Stool, MD* Departments of *Anesthesiology and Pain Management and †Psychiatry, University of Texas Southwestern Medical Center at Dallas The electroencephalogram (EEG) bispectral index (BIS) measures the hypnotic component of the anesthetic state and correlates with emergence from general anes- thesia. Therefore, we hypothesized that the BIS would be useful in predicting electroconvulsive therapy (ECT)-induced seizure times and awakening from methohexital anesthesia. Twenty-five consenting pa- tients with major depressive disorders underwent 100 maintenance ECT treatments. All patients were pre- medicated with glycopyrrolate 0.2 mg IV, and anesthe- sia was induced with methohexital 1 mg/kg IV. The BIS was monitored continuously, and the values were re- corded at specific end-points, including before anesthe- sia (baseline), after the induction of anesthesia (pre- ECT), at the end of ECT (peak), after ECT (suppression), and on awakening (eye opening). The pre-ECT BIS value correlated with the duration of both the motor (r = 0.3) and EEG (r = 0.4) seizure activity (P  0.05). The peak post-ECT BIS value correlated with the duration of the EEG seizure activity (r = 0.5) (P  0.05). A posi- tive correlation was also found between the EEG sei- zure duration and the time to eye opening (r = 0.4) (P  0.05). However, the BIS values on awakening from methohexital anesthesia varied from 29 to 97 and were 60 in 75% of the cases. We conclude that the BIS value before the ECT stimulus is applied could be useful in predicting the seizure time. However, the BIS values on awakening were highly variable, suggesting that it re- flects both the residual depressant effects of methohexi- tal and post-ictal depression. (Anesth Analg 2003;96:1636 –9) T he bispectral index (BIS) is an electroencephalo- gram (EEG)-derived multivariate scale that re- flects the level of hypnosis in anesthetized pa- tients (1). The BIS has been reported to correlate with both loss of consciousness and awakening from IV anesthesia (2–5). However, a recent report by Nishi- hara and Saito (6) questioned the reliability of the BIS monitor in predicting awakening after electroconvul- sive therapy (ECT) with propofol anesthesia. ECT is widely used to treat severe depression in patients who have not responded to standard phar- macotherapeutic drugs (7). Methohexital is the most commonly used IV anesthetic for ECT because it is less likely to interfere with ECT-induced seizure activity than propofol and thiopental (8). However, the rela- tionship between the EEG-BIS values and clinical end- points in depressed patients undergoing ECT with methohexital anesthesia has not been studied. We designed this study to test the hypothesis that BIS values immediately before ECT would be useful in predicting the duration of seizure activity and awak- ening after ECT with methohexital anesthesia. A sec- ondary objective of this study was to assess the corre- lation between the changes in the BIS value during and after ECT and the durations of both motor and EEG seizure activity. Finally, we analyzed the rela- tionship between the severity of the pre-ECT depres- sive symptoms and the baseline BIS value. Methods A total of 25 consenting patients, ranging in age from 30 to 81 yr and undergoing 100 maintenance ECT treatments for major depressive disorders, were en- rolled in this prospective IRB-approved study. In ad- dition to the standard monitoring devices (which in- cluded a noninvasive blood pressure cuff, five-lead The BIS monitoring equipment and supplies were provided by Aspect Medical Systems, Inc. (Newton, MA). Dr. White receives salary support from endowment funds provided by the Margaret Milam McDermott Distinguished Chair. Accepted for publication February 24, 2003. Address correspondence and reprint requests to Paul F. White, PhD, MD, FANZCA, Department of Anesthesiology and Pain Man- agement, University of Southwestern Medical Center at Dallas, 5323 Harry Hines Blvd., F2.208, Dallas, TX 75390-9068. Address e-mail to paul.white@utsouthwestern.edu. DOI: 10.1213/01.ANE.0000066018.13553.08 ©2003 by the International Anesthesia Research Society 1636 Anesth Analg 2003;96:1636–9 0003-2999/03