Several authors have reported sudden increases of the bispectral index values during different procedures such as endoscopic shoulder surgery , deep hypothermic circulatory arrest  or after using an upper-body blanket for warming . In addition, spurious bispectral index values can be observed when the electromyogram activity of the bispectral index monitor reaches a value of 50 . In all these reports there is a general agreement that the bispectral index monitor shows an erroneous value, which must be interpreted carefully.
In the present report we present four cases of patients who underwent general anaesthesia for lumbar arthrodesis, in which the depth of anaesthesia was assessed with the BIS A-2000™ bispectral index monitoring system (Aspect Medical Systems, Natick, MA, USA). In all patients, we observed a sudden increase in the bispectral index values coinciding with an intense surgical stimulation of the spinal roots.
These three patients were two males (66 and 41 yr) who underwent lumbar arthrodesis in the prone position and a female (60 yr) who had a resection of a thoracic vertebral body in the supine position. During surgery, when the surgeon adjusted a screw in the vertebral pedicles or in the course of the vertebral body resection there was a sudden increase of bispectral index values from 55-60 up to more than 90 while the heart rate (HR) and the blood pressure (BP) remained unaltered. The bar indicating signal quality in the monitor was of good quality, and the electromyogram did not show muscle activity. In none of the cases there were clinical signs of inadequate depth of anaesthesia (e.g. tachycardia, hypertension) and thus we did not increase the rates of infusion of remifentanil or propofol. At the time of the increase in bispectral index value, the surgeons were informed and they immediately interrupted the procedure and/or changed the direction of the screws. Afterwards the bispectral index values returned to below 60 within 30-180 s.
In all three patients general anaesthesia was achieved with an infusion of propofol (2 mg kg−1 h−1), a mixture of O2 : N2O (40 : 60) and an infusion of remifentanil (0.2-0.5 μg kg−1 min−1) which was increased or decreased in order to maintain the HR and the mean arterial pressure within a 20% range of baseline values. Neuromuscular blockade was monitored by the train-of-four (TOF) technique using a neurostimulator (Innervator 252; Fisher & Paykel Healthcare, New Zealand) with the electrodes placed on the anterior aspect of the wrist and hand. Boluses of 2 mg of cisatracurium were given to maintain an adequate level of neuromuscular blockade (two twitches of the TOF).
No changes in oxygen saturation (pulse oximetry) or body temperature (oesophageal probe) were observed at the time of the increased values of bispectral index. None of the patients had any recall evaluated 48 h after surgery.
A 69-yr-old, 80-kg male undergoing lumbar arthrodesis (L5-S1), was anaesthetized with sevoflurane (0.8% end tidal) in a mixture of O2 : N2O (40 : 60) and an infusion of alfentanil (0.72-1.0 μg kg−1min−1) administered at a rate to maintain the mean arterial pressure and HR within a 20% range of the baseline values. Neuromuscular blockade was obtained with boluses of 2 mg of cisatracurium titrated to preserve two twitches in the TOF.
At the time of rhizolysis, we observed a sudden increase in bispectral index which reached values between 90 and 100, together with an increase in mean arterial pressure from 91 to 122 mmHg and an increase in HR from 48 to 62 beats min−1. The event was not associated with cough or movement. The rate of alfentanil infusion was increased from 0.72 to 1.0 μg kg−1min−1 for a period of 10 min. BP and HR returned to the previous values (89 mmHg and 48 beats min−1) and the bispectral index value decreased to 60. Therefore, the alfentanil infusion was returned to the initial rate (0.72 μg kg−1 min−1). Afterwards, while the cardiovascular parameters remained unchanged, the bispectral index number increased again to values above 90, where it remained throughout the rest of the surgery, approximately 30 min after the event. Postoperatively, the patient stated that he dreamed about his granddaughter but there was no recall of the surgery.
These abrupt changes of the bispectral index observed during spinal surgery could have been due to a number of factors. Interferences with electrical devices in the operating room, such as the electrocautery, were considered. In all patients, we had good signal quality at the time of the event. It is possible, that the surgical tools used in this type of procedure could produce some kind of interference. However, if this were the case, changes in bispectral index would have appeared more frequently when performing this type of surgery. Moreover, when electrical interference occurs, the signal usually fades from the screen in a characteristic manner. Thus, we do not think that this type of interference could explain the change observed in our patients.
All of our patients were paralysed with cisatracurium and the electromyogram column in the bispectral index monitor showed no change in activity.
In Case 4, we observed high bispectral index values together with an increase in BP and HR. We were able to abolish the haemodynamic response and return the bispectral index value to 60 using alfentanil. This suggests the presence of pain. Afterwards, as the alfentanil infusion was returned to the initial rate, the bispectral index value increased again, a fact that could be related to the transmission of the nociceptive stimuli induced by the excision of a spinal root, to supraspinal sites. The presence of dreaming in this case could be a sign of a light level of anaesthesia .
In Cases 1-3 we observed a short increase in bispectral index values without haemodynamic changes. This could be due to a more transient stimulation and/or to the administration of remifentanil, a more effective opioid than alfentanil.
The adrenergic response induced by the intense nociceptive stimulation during spinal surgery can be blocked by the administration of opioids, combined with intravenous (i.v.) or inhalation anaesthetics. But, in order to avoid the activation of the cerebral cortex observed in these cases, it would be necessary to administer higher doses of hypnotics to maintain bispectral index values close to 40. However, the manifestation of sudden increases in bispectral index (associated or not with changes in cardiovascular parameters) could be a reliable indication of manipulation of the spinal roots or fibres. It is possible that bispectral index could be an early clinical indicator to warn the surgical team about possible nerve injury.
M. M. Puig
1Anaesthesiology Service, Hospital Universitario del Mar, UAB, Barcelona, Spain
1. Hemmerling TM, Migneault B. Falsely increased bispectral index during endoscopic shoulder surgery attributed to interferences with the endoscopic shaver device. Anesth Analg
2. Mychaskiw G, Heath BJ, Eichhorn JH. Falsely elevated bispectral index during deep hypothermic circulatory arrest. Br J Anaesth
3. Guignard B, Chauvin M. Bispectral index increases and decreases are not always signs of inadequate anesthesia. Anesthesiology
4. Baldesi O, Bruder N, Velly L, Gouin F. Spurious bispectral index values due to electromiographic activity. Eur J Anaesthesiol
5. Myles PS, Leslie K, McNeil J, Forbes A, Chan MTV, for the B-Aware trial group. Bispectral index monitoring to prevent awareness during anaesthesia: the B-Aware randomised controlled trial. The Lancet
*The results were presented at the ESA Meeting held in Lisbon, Portugal, 5-8 June 2004.