Secondary Logo

Journal Logo

Original Article

Bispectral index monitoring in patients undergoing cardiac surgery under cardiopulmonary bypass

Puri, G. D.; Murthy, S. S.

Author Information
European Journal of Anaesthesiology: June 2003 - Volume 20 - Issue 6 - p 451-456


Maintenance of a suitable depth of anaesthesia throughout surgery is essential to achieve fast recovery. Recently, the bispectral index (BIS) – a variable derived from the electroencephalogram (EEG) – has been shown to measure the hypnotic component of the anaesthetic state [1–3]. Intraoperative titration of anaesthetic drugs using the BIS facilitates early return to consciousness after surgery [4,5]. In cardiac surgery, a successful anaesthetic technique requires both haemodynamic stability and prevention of awareness. Cardiac surgical patients are more vulnerable to the harmful effects of either too light or too deep anaesthesia in respect of their often limited cardiovascular reserve: too deep an anaesthetic may produce cardiovascular depression, while one too light may induce undesirable neurohumoral responses. We decided to see how useful the BIS monitor behaves in this regard. The present study was planned to evaluate the effect of BIS-controlled anaesthetic titration on haemodynamic stability and recovery characteristics.


The study was conducted at the Department of Anaesthesia and Intensive Care, Postgraduate Institute of Medical Education and Research, Chandigarh, India. Our institutional Review Committee approved the study. Thirty patients (18–70 yr) undergoing either coronary artery grafting (CABG) or valve replacement under cardiopulmonary bypass (CPB) were randomized into either a study group (using BIS) or a control group (no BIS) using computergenerated numbers. Subjects with known neurological disorders, poor ventricular function (ejection fraction <40%), New York Heart Association (NYHA) Grade IV, diabetes mellitus, and those with impaired renal or hepatic function were excluded.

Patients were premedicated with diazepam 0.1 mg kg−1 orally the night before and on the morning of surgery, and morphine 0.1 mg kg−1 intramuscularly was given 45 min before surgery. Peripheral intravenous access and other invasive monitoring catheters (arterial cannula and pulmonary artery catheter) were placed after administration of midazolam 0.05 mg kg−1. Baseline haemodynamic variables were recorded. Patients were monitored by continuous invasive blood pressure determination, HR, electrocardiograph, pulse oximetry and train-of-four responses to neuromuscular stimulation (Biometer®; Myotest, Biometer International, Denmark). Anaesthesia was induced with morphine 0.2 mg kg−1, midazolam 0.05 mg kg−1 and thiopental titrated to loss of the eyelash reflex; endotracheal intubation was facilitated with vecuronium 0.08 mg kg−1.

Four silver chloride electrodes were applied to the skin of the forehead – one over each outer malar bone, the third at the centre of the forehead and the fourth (ground electrode) on one side of the centre electrode. Impedance was checked and kept at <2000 Ω. The EEG signal was acquired using this frontal-Cz montage and the BIS was displayed on the Aspect A-1000® EEG monitor (Aspect Medical Systems, Inc., Natik, MA, USA) on real-time basis (BIS v.3.1). In the study group, the anaesthesiologist was allowed to see and use the monitor and accordingly he adjusted the isoflurane concentration or performed any necessary haemodynamic interventions. The anaesthesiologist for the patients in the control group was blinded to the BIS scores as the EEG monitor was outwith his viewpoint.

Anaesthesia was maintained with isoflurane; N2O (66% in O2 before CPB) and morphine 0.025 mg kg−1 h−1. The administration of isoflurane was titrated to keep BIS between 45 and 55 throughout the procedure except for the last 30 min when it was titrated to 65–75 in the study group. In the control group, the inhaled isoflurane concentration was adjusted according to standard clinical signs of adequacy of anaesthesia (tachycardia, hypertension, sweating, lacrimation, etc.) and with the aim of achieving rapid recovery from its effects. The increase in the vapour concentration was permitted when episodes of inadequate anaesthesia were detected by the anaesthesiologist. Central venous pressure and pulmonary artery pressures were monitored during surgery to optimize preload status. Inotropes were used if the cardiac index was <2L min−1 m−2 after ascertaining the adequate volume load from measurements of the central venous pressure and pulmonary artery wedge pressure. Hypotension and bradycardia were treated by adjusting the doses of anaesthetic drugs, intravenous fluids (preload) or other necessary drugs. In the study group, if the BIS was within the normal range and there was hypertension or tachycardia, morphine 0.05–0.1 mg kg−1 was given slowly intravenously before using vasodilators or β-adrenoceptor blocking drugs.

The extracorporeal circuit included a bubble oxygenator primed with Ringer's lactate. After the aortic root had been cannulated, the necessary cannulae were placed and moderate hypothermic cardiopulmonary was initiated. Body temperature was maintained to between 28 and 30°C in all patients. The oxygenator pump flows were maintained at 2.1–2.3 L min-1 m2 with a mean systemic arterial pressure in the range 40–80 mmHg. Myocardial protection was achieved with cold potassium-enriched blood cardioplegia solution and topical application of cold saline and ice slush. Morphine 3 mg, midazolam 1 mg and vecuronium 0.5 mg were added to the priming solution before the start of CPB.

Neuromuscular junction monitoring was performed to maintain train-of-four twitches to ≤2 throughout surgery and no vecuronium was administered during the last 15 min of surgery. This monitoring was continued after operation until all four twitches reappeared. After sternal closure, the BIS was titrated to 65–75 in the study group. The isoflurane was discontinued once the skin suturing had been completed.

After operation, the lungs of all patients were mechanically ventilated and consciousness was assessed every 5 min until all systems were stable and verbal commands were obeyed. Patients were weaned from controlled ventilation of the lungs using synchronized intermittent mandatory ventilation in the pressure support mode as soon as they were conscious, haemodynamically stable and could sustain adequate respiratory efforts. They were extubated after ascertaining standard extubation criteria.

Heart rate and MAP were recorded every 5 min throughout surgery. The haemodynamic disturbances were defined as: hypertension – MAP > 20% of baseline or >80 mmHg during CPB; hypotension – MAP < 20% of baseline or <40 mmHg during CPB; tachycardia – HR > 20% of baseline before CPB and >30% above baseline after CPB; bradycardia (absolute) – HR < 40 beats min−1.

An episode of alteration of HR or MAP was defined as any of the above-mentioned disturbances of these variables for >5 min. Though monitoring was continuous, we compared sustained haemodynamic disturbances for ≥5 min in this study. All gross fluctuations in arterial pressure and HR were treated immediately by adjusting the amounts of volatile anaesthetic, analgesics, intravenous fluids, β-adrenoceptor-blocking drugs, vasodilators or vasoconstrictors.

Haemodynamic disturbances in the window period of 5 min post-tracheal intubation and 15 min whilst instituting and terminating CPB were not considered. BIS, MAP and HR – at the following predetermined times – were compared between and within the two groups. Preinduction (PreI), 15, 30, 45 and 60 min after induction (I + 15, I + 30, I + 45 and I + 60, respectively); start of CPB (B); 5, 30, 60, 90 and 120 min after the start of CPB (B + 5, B + 30, B + 60, B + 90 and B + 120, respectively); the end of bypass; 5, 30 and 60 min after CPB (EB + 5, EB + 30, EB + 60, respectively), and at the end of surgery.

The time to recovery was defined as the time interval from switching off the anaesthetic vaporizer to when the patient opened his or her eyes and obeyed verbal commands. Any patient who required supplementation of sedation after surgery, but before reaching the primary recovery end-point due to unprecedented events, was excluded from the study. Every patient was interviewed to determine any recall on the first postoperative day after the trachea had been extubated.

Statistical analysis

Parametric data was analysed with unpaired and paired t-tests. Non-parametric variables were analysed with Fischer's exact test and the U-test as appropriate. Data are the mean ± SD or the median and range. P < 0.05 was considered as statistically significant.


Patient characteristics distribution in each group is shown in Table 1. There was no significant difference between the two groups (P > 0.05). MAP and HR remained stable throughout the periods of surgery and bypass and did not differ between treatment and control groups at the predetermined time intervals. There was a significant fall in MAP at the initiation of CPB (P < 0.5) and a rise in MAP on termination of bypass (P < 0.05) in both groups.

Table 1
Table 1:
Patient characteristics data (mean ± SD).

The mean baseline (preinduction) BIS values were 92.17 ± 1.95 and 92.43 ± 4.35 in the study and control groups, respectively (Table 2). There was a significant rise in the BIS after initiation and termination of bypass in the control group (P < 0.05, paired t-test). At the end of surgery, the BIS was lower in the control group (67.42 ± 15.24) than in the study group (75 ± 5.59) (n.s.).

Table 2
Table 2:
Bispectral index, mean arterial pressure and heart rate variables at various intervals during the study (mean ± SD).

There were no differences in the time to reach the defined recovery end-point or the time to tracheal extubation between the two groups (Table 3). Only one patient in the control group recovered within 10 min of the isoflurane vaporizer being switched off, whereas five patients in the study group did so. None of the patients in the study group had recall of intraoperative events when interviewed on the first postoperative day, while one patient in the control group had awareness during the sternotomy. This patient had episodes of tachycardia during this period of stimulus and his BIS at that time was 75.

Table 3
Table 3:
Time needed to achieve the recovery end-point and tracheal extubation (mean ± SD).

There was no significant difference in the number of patients suffering haemodynamic disturbances (Table 4). The number of episodes of hypertension and tachycardia were significantly higher in the control group (P < 0.05, U-test).

Table 4
Table 4:
Haemodynamic disturbances (median and range).


Anaesthesiologists titrate doses of anaesthetic drugs based on the haemodynamic responses to noxious stimulation – which may not necessarily mean awareness – nor does a lack of any haemodynamic change guarantee unconsciousness [6]. This approach is feasible in normal healthy individuals but does not guarantee safety in patients with compromised cardiopulmonary systems. Ideally, the safe approach would be to titrate drugs according to their presumed effect-site concentrations or any measure that reflects this effect. In clinical practice, it would be easier to rely upon a monitor that reflects effect-site changes produced by drugs rather than on complex mathematical exponentials involved in calculating effect-site concentrations. Differences in drug sensitivity between individuals can also occur despite identical effect-site concentrations. This can be caused by different receptor sub-types. This is one of the reasons why a monitor of anaesthetic depth, e.g. the BIS monitor, may improve anaesthesia. Thus, effect-site concentrations may not necessarily be a reliable measure of drug effect.

The BIS has been demonstrated to be safe and efficient as a pharmacodynamic measure of the central effects of anaesthetics during short surgical procedures [4]. It is desirable to ascertain how the index performs during long surgical procedures, where there will be changes in the pharmacokinetics of drugs and because both ‘too deep’ and ‘too light’ anaesthesia can be detrimental to patients. The pharmacokinetics of drugs are altered in cardiac surgical patients because of altered haemodynamics, concurrent cardiovascular medications and partly because of the effects of CPB. Furthermore, anaesthetic titration – based on HR and arterial pressure responses – may be hindered due to vasoactive medication, e.g. β-adrenoceptor blocking drugs, calcium channelblocking drugs and other cardioactive drugs that the patient may be receiving. A decreased incidence of tachycardia and hypertension in the BIS-controlled group may be due to improvements in the titration of drugs in the study group. As the BIS provides additional information of the hypnotic state, the anaesthesiologist should be able to react before such haemodynamic changes occur. Though monitoring was continuous we compared haemodynamic disturbances for periods of ≥5 min in this study. All gross fluctuations in arterial pressure and HR were treated immediately with a standard protocol of adjustment of the volatile anaesthetic concentration, and the administration of hypnotics, analgesics, intravenous fluids, β-adrenoceptor blocking drugs, and vasodilators. Five minutes is a reasonable time to adjust these conditions according to the patient requirement based on feedback. The endotracheal intubation period and the periods immediately associated with the institution or termination of CPB were excluded from consideration. Such periods of gross haemodynamic change are associated with alterations in drug concentration, a change in body temperature, vasoactive drug administration and direct cardiac manipulation, which may be primarily responsible and thus unrelated to anaesthetic depth exclusively. The arterial pressure and HR at the time of termination of CPB were comparable between the two groups, although the BIS was higher in the control group. This demonstrates that the anaesthesiologist was mainly reacting to the haemodynamic variables and trying to restore them to normal while unknowingly accepting a lightly anaesthetized patient.

The tracheal extubation timings were similar in both groups. This may be because the time to tracheal extubation depends on many other factors such as haemodynamic stability, chest drainage and physiological variables such as body temperature that could have affected recovery. These are not dependent on anaesthetic management alone. Moreover, consciousness cannot be the sole predictor of successful tracheal extubation. Three patients, two in the study group and one in the control group, were extubated after >10h. Two of these patients – one each in both groups – required moderate inotropic support in the immediate postoperative period, while one in the study group had chronic obstructive airway disease and was left intubated for 24 h.

There are few studies evaluating the role of BIS monitoring in cardiac surgical patients. Barr and colleagues assessed the BIS as a monitor of the depth of anaesthesia during fentanyl and midazolam anaesthesia in patients undergoing coronary artery bypass surgery [7]. They found that during clinically adequate anaesthesia, the BIS varied considerably and they could not relate it to drug concentration. Similarly, Doi and colleagues found BIS values to be quite variable during cardiac anaesthesia during CPB [8]. In both studies the drugs were not titrated to achieve an objective end-point but were given in fixed, predetermined doses. The variable BIS values achieved may indicate the variability in the pharmacokinetics and pharmacodynamics in different individuals undergoing cardiac surgery. In contrast, Schmidlin and colleagues found the BIS to be the only EEG measurement that paralleled the clinical course of anaesthesia in patients undergoing CABG with propofol-fentanyl anaesthesia during hypothermic CPB [9].

The rise in the BIS values during the institution of bypass may be due to alteration in the brain-blood concentration gradient for anaesthetic drugs because of haemodilution. The rise was not significant in the BIS-titrated group. The anaesthesiologist might have foreseen the impending ‘light’ anaesthesia from the values and trends of the BIS and increase the anaesthetic concentration accordingly. There was no significant difference in MAP between the two groups during this period. The rise in BIS, at the institution of CPB in the control group, may also be due to lowering of anaesthetic delivery by the anaesthesiologist in order to maintain arterial pressure; while in the study group, the anaesthesiologist adjusted the vasoactive drugs to maintain pressure while keeping the BIS constant.

The observed jump in the BIS values at the time of termination of CPB may be due to the temperature rise in the brain which raises the anaesthetic requirement. There is also a tendency on the part of the anaesthesiologist to concentrate more on haemodynamics than the depth of anaesthesia at this particular moment. As there was no significant rise in BIS values whilst instituting or terminating CPB in the study group, it may be concluded that BIS monitoring – and titrating anaesthetic drugs accordingly – is likely to avoid the chance of inadequate depth of anaesthesia. Fortunately, only one patient in our control group reported awareness and that was at the time of sternotomy.

This study failed to demonstrate any significant improvement in the time to recovery when the BIS was used to titrate the anaesthetic administration to supplement routine physiological monitoring. However, before discrediting the concept of BIS monitoring for this purpose, we should consider factors that might have overshadowed the possible advantages of such monitoring in this regard. Residual neuromuscular blockade could have affected the assessment of recovery. The function of the neuromuscular junction was only monitored until the recovery of all four twitches following the train-of-four stimuli and all patients had recovered four twitches in the first 10 min after cessation of anaesthesia. It was presumed that patients' ability to respond to verbal commands would be unaffected by this partial neuromuscular blockade. However, there is still the possibility that some patients may not have been able to respond earlier than 10 min because of the effect of the muscle relaxant when all four twitches had not recovered. In general technology assessment studies it is known that technology influences those practitioners that use it for a small number of their patients thereby improving the outcome for all [10]. No measures were taken in this study to addresses the influence of learning so that it is not biased against new technology. Moreover, the BIS by itself is shown to be an effective teaching tool [11]. Although in our study the patients were randomly assigned to the use of the BIS, or not, statistical techniques were not applied to determine whether the outcome in fact differed. Moreover, only one anaesthesiologist supervised the anaesthetic procedure in all patients. Recruitment of fewer patients may also be the reason for the differences being insignificant. In this study, the patients were followed up for 24 h after tracheal extubation to determine the incidence of awareness. Following up patients for longer periods after operation is more desirable to determine the true incidence of recall [12].

In conclusion, the BIS monitor in patients undergoing cardiac surgery (CABG and valve repair/replacement) under moderate hypothermic CPB decreases the incidence of haemodynamic disturbances and facilitates the titration of adequate amounts of volatile anaesthetic during whole-body perfusion with the extracorporeal circulation.


1. Vernon JM, Lang E, Sebel PS, Manberg P. Prediction of movement using bispectral electroencephalographic analysis during propofol/alfentanil or isoflurane/alfentanil anesthesia. Anesth Analg 1995; 80: 780-785.
2. Liu J, Singh H, White PF. Electro encephalogram bispectral analysis predicts the depth of midazolam-induced sedation. Anesthesiology 1996; 84: 64-69.
3. Kearse Jr LA, Manberg P, Chamoun N, Debros F, Zaslavsky A. Bispectral analysis of the electroencephalogram correlates with patient movement to skin incision during propofol/nitrous oxide anesthesia. Anesthesiology 1994; 81: 1365-1370.
4. Song D, Joshi GP, White PF. Titration of volatile anesthetics using bispectral index facilitates recovery after ambulatory anesthesia. Anesthesiology 1997; 87: 842-848.
5. Gan TJ, Glass PS, Windsor A, et al. Bispectral index monitoring allows faster emergence and improved recovery from propofol, alfentanil and nitrous oxide anesthesia. Anesthesiology 1997; 87: 808-815.
6. Hug CC. Does opioid ‘Anesthesia’ exist? Anesthesiology 1990; 73: 1-4.
7. Barr G, Anderson RE, Samuelsson S, Owall A, Jakobsson JG. Fentanyl and midazolam anaesthesia for coronary bypass surgery: a clinical study of bispectral electroencephalogram analysis, drug concentrations and recall. Br J Anaesth 2000; 84: 749-752.
8. Doi M, Gajraj RJ, Mantzaridis H, Kenny GNC. Effect of cardiopulmonary bypass and hypothermia on electroencephalographic variables Anaesthesia 1997; 52: 1048-1055.
9. Schmidlin D, Haer P, Schmid ER. Monitoring level of sedation and bispectral EEG analysis between hypothermic and normothermic cardiopulmonary bypass. Br J Anaesth 2000; 86: 769-776.
10. Roizen MF, Toledano A. Technology assessment and the learning contamination bias. Anesth Analg 1994; 79: 410-412.
11. Kumar A, De Deyne C, Struys M, Vundelinckx G, Heylen R. Use of bispectral EEG monitoring to evaluate training in anesthesia. Br J Anaesth 1999; 82 (Suppl): A47.
12. Sandin RH. Awareness during anaesthesia: a prospective case study. Lancet 2000; 355: 707-711.

CARDIAC SURGICAL PROCEDURES; myocardial revascularization; coronary artery bypass; DIAGNOSTIC TECHNIQUES; NEUROLOGICAL; electroencephalography; SURGICAL PROCEDURES; OPERATIVE; extracorporeal circulation; cardiopulmonary bypass

© 2003 European Society of Anaesthesiology