Bispectral index controlled anaesthesia for electroconvulsive therapy

Robert Hanss, MD*; Martin Bauer, MD*; Bertholt Bein, MD*; Robert Göder, MD#; Björn Buttgereit, MD*; Anna Christina Schulz-Du Bois, MD#; Markus Steinfath, MD+; Jens Scholz, MD§
*: Staff member
+: Professor of Anaesthesiology
§: Professor of Anaesthesiology and Chair
&: Professor of Anaesthesiology and Vice-Chair
Department of Anaesthesiology and Intensive Care Medicine
#: Staff member, Department of Psychiatry and Psychotherapy
University-Hospital Schleswig-Holstein, Campus Kiel, Germany
Correspondence to: Robert Hanss, Department of Anaesthesiology and Intensive Care Medicine, University-Hospital Schleswig-Holstein, Campus Kiel, Schwanenweg 21, 24105 Kiel, Germany. E-mail:; Tel.: +49-431-597 2971, Fax: +49-431-597 2230.
No author has a conflict of interest in regard of devices discussed in this publication.

Presented, in part, at the 2003 ESA annual meeting in Glasgow, GB.


Keywords: Electroconvulsive Therapy; Electroencephalography; Anaesthesia; Barbiturates


Background: The recommended induction dose of methohexital (MD) for electroconvulsive therapy (ECT) varies in a wide range influencing efficacy of ECT and safety of anaesthesia. Bispectral index (BIS), a monitor of consciousness, may be useful to assure adequate hypnosis with optimized MD.
Methods: Patients with medical resistant major depression (ICD10), scheduled for multiple ECT’s, were studied. Depth of anaesthesia was BIS controlled. ECT was not performed until BIS dropped below 50. Initially anaesthesia was induced with MD of 1.0 mg·kg-1, and stepwise reduced by 0.1 mg·kg-1 during consecutive treatments. If BIS did not drop below 50, MD was supplemented by further boli of 0.5 mg·kg-1, until level of hypnosis was reached. The adequacy of the anaesthetic recommendation for MD (1.0 - 1.5 mg·kg-1) as well as the psychiatric recommendation (0.75 - 1 mg·kg-1) was investigated.
Results: 109 ECT’s in 14 patients were studied. The anaesthetic recommendation was inadequately in 40 % of the treatments, with 12 % exceeding 1.5 mg·kg-1, and 28 % below 1.0 mg·kg-1. Psychiatric recommendation was inadequately in 49 %, with 39 % exceeding 1.0 mg·kg-1, 10 % undershooting 0.75 mg·kg-1.
Conclusions: MD for ECT showed a great variability, exceeding as well as undershooting the dosage recommendations in a wide range. BIS monitoring may be useful to secure adequate hypnosis during muscle relaxation and treatment and may optimize ECT efficacy.


Electroconvulsive therapy (ECT) has an established and important role in the management of medical resistant depression and life threatening conditions such as depressive stupor, catatonia, and neuroleptic malignant syndrome (Fink 2003). ECT is performed after administration of a short acting intravenous anaesthetic. The barbiturate methohexital remains the most widely used general anaesthetic for ECT and is considered the drug of choice (Avramov, Husain et al. 1995; Gurmarnik, Young et al. 1996; Cook, Stevenson et al. 2000; Ding and White 2002).
The anaesthetic drug itself as well as the dosage influence the duration of ECT induced seizure activity and can adversely affect the efficacy of ECT treatment (Nguyen, Chhibber et al. 1997; Conca, Germann et al. 2003). Interestingly the methohexital induction dosage (MD), recommended by the American Psychiatric Association (P = 0.75-1.0 mg·kg-1), is particularly lower than the anaesthetic recommendation (A = 1.0-1.5 mg·kg-1) (Fink 1990; Cook, Stevenson et al. 2000).
It is unlikely that in patients undergoing ECT depth of hypnosis after injection of the anaesthetic at a fixed dose is equal among all subjects. On the one hand patients may receive a barbiturate overdose decreasing ECT efficacy; on the other hand the dose administered may be inadequate for muscle relaxation and ECT with inadequate anaesthesia. Awareness may adversely influence the course of the underlying psychiatric disease (Osterman and van der Kolk 1998; Gurman 2000; Ho 2001). The risk of awareness is pronounced if only clinical criteria like loss of eye-lash reflex, blood pressure, heart rate, and perspiration are applied (Schneider and Sebel 1997).
The bispectral index (BIS) is an electroencephalogram derived multivariant scale that reflects the level of hypnosis in anesthetized patients (Rosow and Manberg 2001). BIS has been reported to correlate with both loss of consciousness and awakening from anaesthesia (Flaishon, Windsor et al. 1997; Glass, Bloom et al. 1997; Liu, Singh et al. 1997; Sebel, Lang et al. 1997). Therefore, it might be a tool to ensure adequate depth of anaesthesia with the optimal MD and may optimize seizure duration during ECT.

Hypothesis: This study was designed to test two hypotheses: Firstly: MD for ECT varies in a wide range not covered by the recommendations of both psychiatric and anaesthesia societies. Secondly: BIS controlled induction of anaesthesia for ECT may help the anaesthesiologist to set the optimal MD for adequate hypnosis during ECT avoiding drug overdose as well as awareness.

Material and Methods

Patient: Approval of the internal review board of the University Hospital Schleswig-Holstein, Campus Kiel (Germany) and informed written consent of all patients or their appropriate relative were obtained. The study was performed as a prospective clinical trial. Patients fulfilling the following criteria were included: medical resistant major depressive episodes (ICD10), confirmed by a consultant psychiatrist not involved in the study, scheduled for multiple ECT’s, age over 18 years. Patients were excluded if they suffered from uncontrolled hypertension, had a history of coronary artery, or cerebro-vascular disease, or met ASA physical status > III.
Hemodynamic and BIS monitoring: On arrival at the anaesthesia induction area, standard monitoring (Datex Ohmeda S/5, Helsinki, Finland) was employed: ECG, non-invasive blood pressure, peripheral oxygen saturation.

BIS (BIS-XP, Aspect Medical Systems, Natick, MA, USA) was recorded from a single frontal derivation using disposable electrodes (BIS Sensor, Aspect Medical Systems, Natick, MA, USA) according to the manufacturer’s instructions (BIS-XP Monitoring system Operating Manual). We accepted the manufacturer’s default settings for electrode impedance (< 7.5 kΩ). The monitor’s impedance check feature was kept continuously turned on during recording. The signal quality index was recorded to assure BIS signal quality. The EEG was sampled at 128 Hz; high pass and low pass filters for BIS calculation were at 0.5 and 48 Hz, respectively. BIS latency was set at 15sec.
ECT: ECT was carried out with Thymatron-IV (Somatics Inc., Lake Bluff, Ill USA). Duration of EEG seizure activity was measured by EEG recordings of polyspike-activity via single bifrontal leads. Additionally the postictal suppression index as a measure for seizure quality was calculated. The investigator was blinded for BIS values and MD.

Study protocol: Anaesthesia was induced with methohexital following the study protocol. Muscle relaxation was achieved with succinylcholin (1 mg·kg-1). ECT induced bradycardia was treated with atropine (0.5 mg), hypertension and tachycardia were treated with esmolol (1 mg·kg-1), and urapidil (10 to 25 mg) until blood pressure reached preoperative limits. BIS was monitored continuously, from arrival in the anaesthesia induction area, until discharge to the post anaesthesia care unit. Muscle relaxation and ECT were performed only when a BIS value of less than 50 was reached after injection of MD. The first session was done with 1 mg·kg-1 methohexital in all patients. In every further treatment, MD was stepwise reduced by 0.1 mg·kg-1 (1 mg·kg-1 => 0.9 mg·kg-1 => 0.8 mg·kg-1). If BIS did not decrease below 50 after the initial MD, it was supplemented by repeated boli of 0.5 mg·kg-1 methohexital, until defined BIS value was reached. If ECT energy or psychiatric medication were changed between two treatments, baseline measurement was repeated and MD of further sessions were stepwise reduced as described above. Total MD per session was recorded.

Data acquisition: BIS values, blood pressure, heart rate, and peripheral oxygen saturation were analyzed at the following events: awake patient (AWAKE), after induction of anaesthesia (Post I), at ECT (ECT), two minutes after ECT (ECT+2), five minutes after ECT (ECT+5), and before discharge to the post anaesthesia care unit (DIS). To evaluate the ECT success, the Hamilton depression score based on 21 items was calculated in all patients before the first treatment and after the last by an investigator blinded for the study results. A score of more than 20 is known as major depressive episode and decrease to levels of 50 % or less from baseline reflects remission.
Statistics: The data were analyzed using GraphPad PRISM statistic and graphic software (GraphPad Software, San Diego, CA). Normally distributed data (demographics, hemodynamic data) are expressed as mean ± SD. Non-normally distributed data (BIS, MD) are expressed as median, 25th – 75th percentile and range.


A total of 15 Patients were studied. One patient was excluded from analysis because ECT was cancelled due to severe cognitive dysfunctions after the first session. In the remaining 14 patients a total of 109 ECT’s were performed (8 ± 3). Demographic data and the Hamilton depression index are shown in table 1, the Hamilton depression index showed a value of major depression before treatment in all patients (26 ± 4 [32/24]). After completion of ECT sessions the index decreased in 8 of the 14 patients to levels of 50 % or less from baseline, reflecting remission (11 ± 6 [5/22]).
Hemodynamic data and BIS values are shown in table 2.

Changes due to hemodynamic side effects of ECT were moderate due to medical interventions as described above. Because muscle relaxation and ECT was not performed unless BIS decreased to 50 or less, this level was reached in all patients. After ECT BIS remained compromised, not reflecting the level of hypnosis any more. Discharge values of BIS showed a great variation between 25 and 98.
Individual MD’s are shown in figure 1. MD showed great inter-individual differences with levels from 0.5 up to 2.7 mg·kg-1 and intra-individual changes during the therapy from 0.9 to 2.7 mg·kg-1. MD recommended by the American Psychiatry Association was adequate in only 56 sessions (51 %). In 53 sessions (49 %) this dosage was inadequate. In 43 sessions (39 %) MD was increased above 1.0 mg·kg-1, 10 sessions (10 %) required a dosage less than 0.75 mg·kg-1.
The MD recommended by anaesthesiologists was adequate in 65 treatments (60 %). In 44 sessions (40 %) this dosage was inadequate. In 13 sessions (12 %) the MD was increased above 1.5 mg·kg-1, 31 sessions (28 %) required a MD less than 1.0 mg·kg-1. Due to overlapping recommendations the sum of adequate and inadequate sessions of psychiatric and anaesthetic recommendations was greater than the total number of ECT’s.


In a prospective clinical trial the MD needed for adequate hypnosis was investigated in 14 patients, scheduled for multiple ECT’s. MD was BIS controlled adapted to the individual needs aiming the lowest MD possible, which still ensures a BIS level of 50 or less, corresponding with adequate hypnosis. A total of 109 ECT’s were investigated. MD showed a great inter-individual and intra-individual variation. The MD recommended by the American Psychiatry Association was found to be inadequate in 49 % of the ECT’s (Fink 1990). The MD recommended in the anaesthesia textbooks was inadequate in 40 % of the ECT’s (Cook, Stevenson et al. 2000).

How to set the induction dose for ECT?

Normally, calculation of the induction dose of hypnotics is based on patient’s body weight, age, sex, and concomitant diseases. Adequacy of hypnosis is evaluated by hemodynamic, respiratory, and autonomic signs (Schneider and Sebel 1997; Halliburton 1998; Dodds 1999). These parameters are indirect and nonspecific, and therefore, may result in an over- or underdose of anaesthetic drugs. Especially patients with a history of chronic psychiatric disorders and long term medical therapy may show even greater variability of the required induction dose (Eranti and McLoughlin 2003; Rasmussen 2003). Furthermore, during ECT hemodynamic parameters are useless for evaluating depth of hypnosis due to hemodynamic side effects of the treatment.

BIS is an established monitor of depth of anaesthesia (Schneider and Sebel 1997; Halliburton 1998; Dodds 1999). It displayed the great variability of MD in depressive patients scheduled for ECT. The reason of this wide range might be the absence of co-analgesics, chronic treatment with benzodiazepines which induce the cytochrom-p450 enzymes as barbiturates (see table 3) and the underlying mental disorders (Hoen, Bijsterbosch et al. 2001). As described in the current literature postictal BIS values varied in a wide range not reflecting the clinical level of consciousness any more (Gunawardane, Murphy et al. 2002).

Based on the current recommendations a considerable proportion of patients (A: 28 % vs. P: 10 %) would have received a methohexital overdose. It has been shown that the total anaesthetic induction dose has a negative correlation with seizure duration (Nguyen, Chhibber et al. 1997; Smith, Angst et al. 2003; Wajima, Shiga et al. 2003). Therefore, anaesthetic overdose may adversely affect the efficacy of ECT treatment (Nguyen, Chhibber et al. 1997; Conca, Germann et al. 2003). In a recently published study, a combination of propofol and sevoflurane was compared to the administration of one of the anaesthetics alone. The combination of both drugs led to the shortest seizure duration compared to propofol or sevoflurane alone (Wajima, Shiga et al. 2003). Another study compared two different methohexital dosages: 1.25 mg·kg-1 vs. 0.625 mg·kg-1. The latter barbiturate dose was supplemented with remifentanil. Seizure duration in the low dose group was significantly higher than in the high dose methohexital group (Smith, Angst et al. 2003). Comparable results were published by others, investigating induction of anaesthesia with methohexital (0.75 mg·kg-1 vs. 0.5 mg·kg-1) or propofol (0.75 mg·kg-1 vs. 0.5 mg·kg-1) with or without remifentanil supplementation (Nguyen, Chhibber et al. 1997). Both lower anaesthetic dosages provided better conditions for ECT. None of the published studies established an objective measure of hypnosis. Our data suggest that some of the patients, especially those in the low-dose groups, were treated under inadequate hypnosis with a potential risk of awareness.
Based on the current recommendations, a considerable number of patients (A: 13 % vs. P: 40 %) would have received muscle-relaxation and therapy inadequately anesthetized. Though lowest possible MD is crucial for successful ECT, some patients needed a higher dose, eventually due to the underlying disease. MD underdose is associated with awareness. Awareness is a traumatic event and may lead to post-traumatic stress disorders (Osterman and van der Kolk 1998).

Awareness itself was found to result in serious psychological sequelae in some patients and may influence patient’s outcome (Gurman 2000; Ho 2001). Therefore, especially in patients with known psychological disorders it is crucial to ensure adequate hypnosis during therapy to avoid potential additional trauma. The advantages of light anaesthesia must be balanced against the risk of awareness induced post-traumatic stress disorders. BIS monitoring might be a tool to detect awareness as shown in patients without psychiatric disorders (Arbour 2004). Usually, clinical signs of inadequate anaesthesia combined with postoperative questionnaire are used to objectively assess awareness (Spitellie, Holmes et al. 2002). These are impossible measures in ECT patients, because hemodynamic changes are an inadequate measure of hypnosis during ECT, movements are restrained by muscle relaxation, and postoperative questioning is impossible due to post ECT amnesia.
Some limitations of our study should be noted. The psychiatric disease itself may have an influence on BIS readings. Recently it was shown that BIS is compromised in patients suffering from dementia (Renna, Handy et al. 2003). However, EEG impairments in depressive patients are unlikely and have not been described yet. Chronic psychiatric medication may affect the EEG, and no studies have been published on changes of BIS values due to this medication. Due to postictal amnesia a questioning after ECT about awareness was impossible. Thus, despite adequate BIS values awareness during ECT cannot completely ruled out.

In conclusion, anaesthesia for ECT has to assure best conditions with the optimal MD for adequate hypnosis during muscle relaxation and treatment. Therefore, an objective monitoring of the level of hypnosis should be established, especially because traditional measurements of hypnotic adequacy can not be applied in these patients. Concerning the wide range of MD and the vast number of patients exceeding or undershooting the recommended doses of both anaesthetic textbooks and psychiatric societies BIS may be an objective measure of hypnosis and a helpful tool to adjust the induction dose to individual needs. Therefore, BIS monitoring should be applied for ECT.


Arbour, R (2004). Using bispectral index monitoring to detect potential breakthrough awareness and limit duration of neuromuscular blockade. Am J Crit Care 13(1): 66-73.
Avramov, MN, Husain, MM, White PF (1995). The comparative effects of methohexital, propofol, and etomidate for electroconvulsive therapy. Anesth Analg 81(3): 596-602.
Conca, A, Germann R, et al. (2003). Etomidate vs. thiopentone in electroconvulsive
therapy. An interdisciplinary challenge for anesthesiology and psychiatry. Pharmacopsychiatry 36(3): 94-7.
Cook, A, Stevenson G, et al. (2000) A survey of methohexitone use by anesthetists in the clinical practice of ECT in Edinburgh. J Ect 16(4): 350-5.
Ding, Z, White, PF (2002). Anesthesia for electroconvulsive therapy. Anesth Analg 94(5): 1351-64.
Dodds, C (1999) General anaesthesia: practical recommendations and recent advances. Drugs 58(3): 453-67.
Eranti, SV, McLoughlin, DM (2003) Electroconvulsive therapy - state of the art. Br J Psychiatry 182: 8-9.
Fink, M (1990) The Practice of ECT: Recommendations for Treatment, Training, and Privileging -- Background. Convuls Ther 6(2): 82-4.
Fink M (2003) Electroconvulsive therapy update: recognizing and treating psychotic depression. J Clin Psychiatry 64(3): 232-4.
Flaishon, R, Windsor, A et al. (1997) Recovery of consciousness after thiopental or propofol. Bispectral index and isolated forearm technique. Anesthesiology 86(3): 613-9.
Glass, PS., Bloom M et al. (1997) Bispectral analysis measures sedation and memory effects of propofol, midazolam, isoflurane, and alfentanil in healthy volunteers. Anesthesiology 86(4): 836-47.
Gunawardane, PO, Murphy PA et al. (2002) Bispectral index monitoring during electroconvulsive therapy under propofol anaesthesia. Br J Anaesth 88(2): 184-7.
Gurman, G. M. (2000). Awareness during general anesthesia. Is this any implication for the outcome? Minerva Anestesiol. 66: 177-81; discussion 181-2.
Gurmarnik, S, Young R et al. (1996) Divided doses of methohexitone improves ECT outcome. Can J Anaesth 43(5 Pt 1): 535.
Halliburton, JR (1998) Awareness during general anesthesia: new technology for an old problem. Crna 9(2): 39-43.
Ho, AM (2001) 'Awareness' and 'recall' during emergence from general anaesthesia. Eur J Anaesthesiol. 18: 623-5.
Hoen, PA, Bijsterbosch MK et al. (2001) Midazolam is a phenobarbital-like cytochrome p450 inducer in rats. J Pharmacol Exp Ther 299(3): 921-7.
Liu, J, Singh, H et al. (1997) Electroencephalographic bispectral index correlates with intraoperative recall and depth of propofol-induced sedation. Anesth Analg 84(1): 185-9.
Nguyen, TT, Chhibber AK, et al. (1997) Effect of methohexitone and propofol with or without alfentanil on seizure duration and recovery in electroconvulsive therapy. Br J Anaesth 79(6): 801-3.
Osterman, JE, van der Kolk, BA (1998) Awareness during anesthesia and posttraumatic stress disorder. Gen Hosp Psychiatry. 20: 274-81.
Rasmussen, KG(2003) Clinical applications of recent research on electroconvulsive therapy. Bull Menninger Clin 67(1): 18-31.
Renna, M, Handy J, et al. (2003) Low baseline Bispectral Index of the electroencephalogram in patients with dementia. Anesth Analg 96(5): 1380-5, table of contents.
Rosow, C, Manberg, PJ (2001) Bispectral index monitoring. Anesthesiol Clin North America 19(4): 947-66, xi.
Schneider, G, Sebel, PS (1997) Monitoring depth of anaesthesia. Eur J Anaesthesiol Suppl. 15: 21-8.
Schneider, G, Sebel, PS (1997) Monitoring depth of anaesthesia. Eur J Anaesthesiol Suppl 15: 21-8.
Sebel, PS, Lang, E et al. (1997) A multicenter study of bispectral electroencephalogram analysis for monitoring anesthetic effect. Anesth Analg 84(4): 891-9.
Smith, DL, Angst, MS et al. (2003) Seizure duration with remifentanil/methohexital vs. methohexital alone in middle-aged patients undergoing electroconvulsive therapy. Acta Anaesthesiol Scand 47(9): 1064-6.
Spitellie, PH, Holmes, MA et al. (2002) Awareness during anesthesia. Anesthesiol Clin North America. 20: 555-570.
Wajima, Z, Shiga T et al. (2003) Propofol alone, sevoflurane alone, and combined propofol-sevoflurane anaesthesia in electroconvulsive therapy. Anaesth Intensive Care 31(4): 396-400.

Figure 1: Methohexital dosages P = psychiatric recommendation, A = anesthesiologic recommendation, n = number of treatments,
MD, anaesthetic recommendation, MD, psychiatric recommendation


Table 1: Demographic data, hamilton depression index.

Height (cm) Weight (kg) Age (y) Sex (m/f) ASA II/III HAMD-BEF HAMD-AFTER
171 ± 5 68 ± 11 59 ± 8 4/10 9/5 26 ± 4 (20/32) 11 ± 6 (5/22)

Kg = kilogram, y = years, m = male, f = female, ASA = ASA physical status, HAMD-BEF = hamilton depression index before ECT, HAMD-AFTER = hamilton depression index at the end of ECT. Mean ± SD, hamilton depression index is presented as mean ± SD with minimum and maximum in brackets.

Table 2: BIS values and hemodynamic data at events 1 to 6


BIS 93 [76/99] 53 [25/99] 45 [25/48] 44 [23/96] 46 [21/97] 57 [25/98]
HR(bpm) 84 ± 21 92 ± 17 95 ± 19 94 ± 21 83 ± 14 80 ± 14
BP(mmHg) 136 ± 23 136 ± 23 151 ± 30 160 ± 30 152 ± 20 145 ± 18


BIS = Bispectral index, HR = heart rate, BP = systolic blood pressure, AWAKE = baseline awake patient, POST-I = direct after induction of anaesthesia, ECT = electroconvulsive therapy, ECT + 2 = 2 minutes after ECT, ECT + 5 = 5 minutes after ECT, DIS = Right before discharge from the OR. HR and BP: means ± SD, BIS: median, range.

First Published: February 2005

Click on these links to visit our Journals:
Priory Bookshop | Search for papers and articles  | Psychiatry On-Line 
Dentistry On-Line
 |  Vet On-Line | Chest Medicine On-Line 
GP On-Line | Pharmacy On-Line | Anaesthesia On-Line | Medicine On-Line
Family Medical Practice On-Line

Home • Journals • Search • Rules for Authors • Submit a Paper • Sponsor us
Rules for Authors
Submit a Paper
Sponsor Us

Google Search

Advanced Search



Default text | Increase text size