Secondary Logo

Share this article on:

Pre-Ictal Bispectral Index Has a Positive Correlation with Seizure Duration During Electroconvulsive Therapy

Nishihara, Fumio MD; Saito, Shigeru MD

doi: 10.1097/00000539-200205000-00037

Propofol anesthesia increases the seizure threshold of patients receiving electroconvulsive therapy. Excessive neuronal suppression could result in an unacceptably short seizure. We sought to identify the correlation between the pre-ictal bispectral index (BIS) score and seizure duration in patients receiving electroconvulsive therapy under propofol anesthesia. BIS was monitored in 38 psychotically depressed patients. Anesthesia was induced by a bolus injection of 1 mg/kg of propofol. The duration of muscular and electroencephalographic seizure was measured during the therapy. The BIS immediately before the electrical shock was 54 ± 13. Both muscular and electroencephalographic seizure durations had a positive correlation with pre-ictal BIS (r = 0.68 and 0.73, respectively;P < 0.01). After the electrically induced seizure, BIS decreased to 30 ± 8, reflecting post-ictal suppression. BIS scores when the patients had awakened after the seizure had a wide variation (range, 29–81; mean, 45; sd, 13). In conclusion, seizure duration has a positive correlation with BIS immediately before electrical shock; however, BIS may not be an accurate predictor of awakening after electrical shock.

Department of Anesthesiology & Reanimatology, Gunma University School of Medicine, Maebashi, Japan

Supported by Grant-in-Aid 11770834 for Scientific Research from the Ministry of Education, Science and Culture of Japan (Tokyo).

January 16, 2002.

Address correspondence and reprint requests to Shigeru Saito, MD, Department of Anesthesiology & Reanimatology, Gunma University School of Medicine, 3-39-22, Showamachi, Maebashi, 371-8511, Japan. Address e-mail to

Electroconvulsive therapy (ECT) is effective for the treatment of drug therapy-resistant severe depression. Because the seizure itself is believed to be important to the efficacy of the therapy, the anesthetics used should not interfere with electrical seizures (1). Until recently, short-acting barbiturates, such as methohexital and thiopental, were often used for anesthesia (1). More recently, propofol at <1 mg/kg has also been recommended for ECT anesthesia (2,3). Several studies have demonstrated that systemic hemodynamics during ECT are more stable under propofol anesthesia than under barbiturate anesthesia (2–5). In a previous study, we also showed that drastic cerebral hemodynamic changes during ECT, measured by transcranial Doppler sonography, were ameliorated by propofol anesthesia (6). However, seizure duration tends to be shorter under propofol anesthesia than under barbiturate or etomidate anesthesia.

In several previous studies, the relationship between propofol dose and seizure duration was carefully investigated. Simpson et al. (7) reported that 1.3 mg/kg of propofol significantly reduced seizure duration, and they concluded that this dose is not acceptable for ECT. Avramov et al. (3) also demonstrated that propofol, at doses larger than 1 mg/kg, causes 45% decreases in ECT-induced seizure duration. Although the view that seizure duration is a primary determinant of treatment efficacy is changing, seizure durations of <25 s are still believed to be ineffective (8).

The level of hypnosis in ECT patients after injection of propofol at a fixed dose is not necessarily identical among study subjects. This is because several pharmacokinetic factors, such as circulating blood volume, clearance rate, and sensitivity to propofol, are not identical among patients. It is possible that, in the patients whose seizure duration after electrical shock is unacceptably short, effects of propofol are more profound than in others because of pharmacokinetic or pharmacodynamic characteristics of these patients. Recent studies of anesthesia depth, using bispectral index (BIS) monitoring, have demonstrated that the BIS score under propofol anesthesia and sedation correlates with the level of hypnosis (9). It is possible that the level of hypnosis measured by BIS before electrical shock correlates with seizure duration after electrical shock. However, there has been no study in which BIS value before electrical shock was examined, nor has there been a study in which the relationship between the BIS value and the neurological response after the shock was investigated. In this study, we continuously measured BIS value during ECT under propofol anesthesia. The dose of propofol was 1 mg/kg, as in our previous study.

Back to Top | Article Outline


Informed consent was obtained from the patient or, where necessary, the appropriate relative. This study protocol was approved by the local Clinical Study Committee, which considers the ethics and legal aspects of clinical investigations. ECT was prescribed to 38 patients experiencing endogenous depression. The patients ranged from 27 to 70 yr of age and were in good physical heath. No patient had cardiovascular or cerebrovascular complications or drug allergies. All patients were treated more than six times (three times per week at 2-day intervals). The data were obtained in the second ECT trial in each case. All persons present at the ECT session were blinded to the displayed value of the BIS monitor. The data obtained in this study were analyzed later by an independent individual.

To avoid an unfavorable parasympathetic reflex, atropine 0.01 mg/kg IM was given as premedication. Arterial blood pressure was measured continuously at the right radial artery by using a tonometric blood pressure monitor (CBM-7000; Colin Co. Ltd., Komaki, Japan). The BIS (Aspect Medical Systems, Natick, MA) electrode was attached to the forehead of the patients as instructed by the manufacturer. Single-lead electroencephalography (EEG) was recorded on the same monitor. General anesthesia was induced with propofol (1 mg/kg). Propofol was administered over 15 s through an indwelling IV catheter. After loss of consciousness, succinylcholine chloride (1 mg/kg) was administered, and ventilation was assisted with a face mask and 100% oxygen. One minute after the injection, an electrical current was applied bilaterally for 5 s at the minimal stimulus intensity, which had been determined in the first ECT trial by a stepwise increase in electrical intensity. The electroshock stimulus was delivered by a trained psychiatrist using an ECT stimulator (CS-1; Sakai Iryo Co. Ltd., Tokyo, Japan). The efficacy of electrical stimulation was determined by the so-called tourniquet technique—that is, by observation of convulsive movements of the distal leg, around which an inflated tourniquet was set to block the distribution of muscle relaxant. Consciousness was assessed by calling the patient’s name every 30 s after the start of spontaneous respiration. The end-expiratory CO2 partial pressure (end-tidal CO2) at the nostrils and Spo2 were monitored by a respiration monitor (Capnomac Ultima; Datex Co. Ltd., Helsinki, Finland), and end-tidal CO2 tension was maintained at 30–35 mm Hg and the Spo2 value (measured at left index) at >98% by manual ventilation assistance throughout the therapy.

The data are expressed as mean ± sd. BIS scores were compared by one-way analysis of variance with the Scheffépost hoc test. To evaluate the correlation between BIS scores before electrical shock and seizure durations or circulatory variables, simple regression analysis was performed with a computer program (StatView 5.0; SAS Institute Inc., Cary, NC). A P value <0.05 was considered statistically significant.

Back to Top | Article Outline


Demographics of the patients were as follows: 54 ± 16 yr in age, 49 ± 8 kg in weight, and 157 ± 8 cm in height. Patients in this study had been prescribed multiple psychiatric medications at various doses. These drugs were imipramine, setiptiline, flunitrazepam, levomepromazine, amitriptyline, trazodone, etizolam, phenobarbital, clomipramine, alprazolam, mianserin, cloxazolam, chlorpromazine, and paroxetine.

All patients could not respond to verbal command within 2 min after injection of propofol. The stimulus intensity was 106 ± 8 V. No patient could recall ECT procedures, and no complaint was reported after ECT regardless of the BIS score before electrical shock. Seizure duration measured by muscle movement was 29 ± 10 s, and the duration measured by EEG was 39 ± 15 s.

BIS scores decreased from the initial values (92 ± 5, immediately after the start of recording) after injection of propofol and gradually increased from the minimum value. During the electrical current application, the BIS monitor indicated an “artifact” message showing interference by the electrical stimulation. During the seizure, the monitor further indicated an “artifact” message or questionable values. After the EEG seizure, the score decreased drastically to 10–49. Patients opened their eyes at 7.3 ± 2.3 min after electrical shock. In some cases, the BIS scores abruptly increased after their eye opening. However, in other cases, the score did not increase further or slowly increased until discharge to the ward (Table 1). BIS score before electrical stimulation had positive statistically significant correlations with motor seizure duration (Fig. 1A) and EEG seizure duration (Fig. 1B).

Table 1

Table 1

Figure 1

Figure 1

Back to Top | Article Outline


The influence of antipsychotic drugs, such as antidepressants and major tranquilizers, on BIS score has not been well understood. Minor tranquilizers such as diazepam and midazolam apparently affect BIS score (10). Patients in this study had been medicated with multiple drugs, including diazepam-related drugs, and it is possible that the BIS score before anesthesia was affected by these drugs. However, BIS scores after the administration of 1 mg/kg of propofol in this study were comparable to those of previous reports (9,11). Although the dose of propofol was determined by body weight, the minimum BIS score after administration ranged from 31 to 69. It is possible that the BIS score was affected by the pharmacokinetic variation of patients, such as variation in circulating blood volume or drug clearance, or by variation in the sensitivity of the central nervous system. In this study, propofol was administered by bolus injection, not by continuous infusion. This method of anesthesia is considered standard practice in many reports (2–6). The BIS score gradually increased after reaching the minimum value. The BIS score immediately before electrical stimulation varied among the patients (range, 30–80). This variation in recovery of BIS score may be due to differences in pharmacokinetic characteristics among the patients (such as drug clearance) or to minor differences in the interval between anesthesia induction and electrical stimulation.

Seizure duration in this study was comparable to durations seen in previous studies in which 1 mg/kg of propofol was used as the anesthetic (3,6), and all of the patients were effectively treated by ECT. Although the view that seizure duration is a primary determinant of treatment efficacy is changing, an extremely short seizure is still believed to be ineffective. However, no study has determined the threshold of effective seizure duration under propofol anesthesia. Accordingly, it is impossible to predict inadequate seizure by BIS score. Further assessment of seizure quality and the effects of ECT may be indispensable to predict inadequate seizure by BIS monitoring.

There has been no study regarding BIS score after seizure. Only one case report described a decrease in BIS after electrical shock (12). In this study, as predicted by the crude EEG pattern (13), during the post-ictal suppression phase after electrical shock, the BIS score was suppressed to a very low range. However, even after the patients had awakened, the BIS score was comparable to the minimum value reached immediately after propofol injection. Although the BIS score slowly increased before discharge to the ward, the final value before discharge was still lower than the initial value and ranged widely, from 30 to 95. These findings indicate that the BIS score after ECT might not reflect the level of consciousness. Although there are differences between electrically induced seizures and other types of seizures, it is possible that the BIS score after other types of seizures also does not reflect the level of hypnosis. Because BIS was designed to evaluate the hypnotic level induced by certain anesthetics and its reliability is restricted to adult standard cases, caution should be exercised when using a BIS monitor in nonstandard cases (9). Watcha (14) recently reported that the BIS value might not be directly applicable to young children. Stanski (10) proposed that care should be taken when using the BIS monitor in special situations, such as pediatrics, pregnancy, and some disease states. The results of this study further suggest that the BIS score after seizure should be cautiously evaluated.

In conclusion, seizure duration has a positive correlation with BIS value immediately before electrical shock. However, BIS score may not be an accurate predictor of awakening time after ECT.

Back to Top | Article Outline


1. Gaines GY, Rees DI. Anesthetic considerations for electroconvulsive therapy. South Med J 1992; 85: 469–82.
2. Fredman B, d’Etienne J, Smith I, et al. Anesthesia for electroconvulsive therapy: effects of propofol and methohexital on seizure activity and recovery. Anesth Analg 1994; 79: 75–9.
3. Avramov MN, Husain MM, White PF. The comparative effects of methohexital, propofol, and etomidate for electroconvulsive therapy. Anesth Analg 1995; 81: 596–602.
4. Boey WK, Lai FO. Comparison of propofol and thiopentone as anaesthetic agents for electroconvulsive therapy. Anaesthesia 1990; 45: 623–8.
5. Geretsegger C, Rochowanski E, Dartnig C, Unterrainer AF. Propofol and methohexital as anesthetic agents for electroconvulsive therapy (ECT): a comparison of seizure-quality measures and vital signs. J ECT 1998; 14: 28–35.
6. Saito S, Kadoi Y, Nara T, et al. The comparative effects of propofol versus thiopental on middle cerebral artery blood flow velocity during electroconvulsive therapy. Anesth Analg 2000; 91: 1531–6.
7. Simpson KH, Halsall PJ, Carr CME, Stewart KG. Propofol reduces seizure duration in patients having anaesthesia for electroconvulsive therapy. Br J Anaesth 1988; 61: 343–4.
8. Mackenzie RA, Southorn PA, Stensrud PE. Anesthesia at remote locations. In: Miller RD, ed. Anesthesia. 5th ed. New York: Churchill Livingstone, 2000: 2241–69.
9. Reves JG, Glass PSA, Lubarsky DA. Nonbarbiturate intravenous anesthetics. In: Miller RD, ed. Anesthesia. 5th ed. New York: Churchill Livingstone, 2000: 228–70.
10. Stanski DR. Monitoring depth of anesthesia. In: Miller RD, ed. Anesthesia. 5th ed. New York: Churchill Livingstone, 2000: 1087–116.
11. Flaishon R, Windsor A, Sigl J, Sebel PS. Recovery of consciousness after thiopental or propofol: bispectral index and isolated forearm technique. Anesthesiology 1997; 86: 613–9.
12. Tanabe T, Sakai T, Mi WD, Matsuki A. Electroconvulsive therapy decreased bispectral index: a case report. Masui 1998; 47: 1096–8.
13. Saito S, Yoshikawa D, Nishihara F, et al. The cerebral hemodynamic response to electrically induced seizures in man. Brain Res 1995; 673: 93–100.
14. Watcha MF. Investigations of the bispectral index monitoring in pediatric anesthesia: first things first. Anesth Analg 2001; 92: 805–7.
© 2002 International Anesthesia Research Society