364
J. Raeder
low doses of anaesthetic drugs, and were awake during surgery without being able to
open eyes or move, thus they could be left in a state of intense anxiety during ongoing
surgery without anyone noticing. This phenomenon of so-called “awareness” was
reported in about 1 patient per 2000 general anaesthetics with curare, not in any
way life-threatening or frequent, but very traumatic for those involved [11]. In most
of these cases opioids were used as anti-nociceptive agents, ensuring stable and low
readings of BP and HR, but not reliable sleep. Thus, both the awareness cases but also
cases of overdosing with subsequent slow recovery called for better monitoring than
just looking at BP, HR, minor movements and signs of sympathetic stimulation. It
has also been shown in the elderly patients that overdose of general anaesthetic drugs
may increase the incidence of postoperative delirium [12] and cognitive decline [2].
As the central nervous system is the target organ of general anaesthetic drugs, ways
of monitoring brain function directly, such as the EEG [6], were looked for.
The raw EEG is basically a continuous flow of electrical signals from 20–40 electrodes placed scattered across the full scalp. The spontaneous EEG have some characteristics which allow the expert to define which state the individual patient belongs
to along an axis from: excited → quiet awake → asleep → unconscious. However,
the complete EEG is very cumbersome to both register and interpret, thus simplified
ways of looking at EEG were looked for. One approach was to look at the frontal
EEG response (amplitude, frequency and delay) to a standardized stimulus, e.g. sound
click, as with the auditory avoked potential (AEP). Another approach was to look
for ways to interpret the spontaneous EEG by automatic computer algorithms, basically looking at frequencies, amplitudes and phase positions during different states
of depth of anaesthesia. Fortunately, the single and easily accessible frontal EEG lead
was useful in this context, as there is a frontal shift of EEG power during general
anaesthesia. However, it was soon evident that the EEG changes were complex
and not linear with dose, and EEG changes were different with different drugs and
drug mixtures, even though they produced the same clinical depth of general anaesthesia [1, 10] Also, there were problems of electrical signals from forehead muscles
interfering with the EEG and problems of surgery and nociception interfering with
the EEG algorithms. The problems were attempted to be solved by complex signal
analyses, such as with the entrophy, narcotrend, cerebral state monitor and other
devices [9]. The most successful device proved to be the BIS monitor, launched in
the 1990ties, with a single frontal EEG reading, a secret and patented algorithm for
EEG interpretation, and a simple output of a score from 0–100. With a score value
below 60 the patient was asleep, and with a value of more than 60 for more than
5 min there was a high chance of the patients being awake and aware, while values
in the 90–100 range were seen in fully awake individuals. Although the BIS has
gained widespread use and proved to reduce the incidence of awareness [3] as well
as avoiding cases of inadvertent overdosing, there are problems with this device as
well. The signal is delayed with 1–2 min, it may be disturbed by muscle activity, it
may be inaccurate with low-power EEG in the elderly, it does not work predictably
with nitrous oxide or ketamine and it is inaccurate in monitoring the antinociceptive
part of general anaesthesia.
J. Raeder
low doses of anaesthetic drugs, and were awake during surgery without being able to
open eyes or move, thus they could be left in a state of intense anxiety during ongoing
surgery without anyone noticing. This phenomenon of so-called “awareness” was
reported in about 1 patient per 2000 general anaesthetics with curare, not in any
way life-threatening or frequent, but very traumatic for those involved [11]. In most
of these cases opioids were used as anti-nociceptive agents, ensuring stable and low
readings of BP and HR, but not reliable sleep. Thus, both the awareness cases but also
cases of overdosing with subsequent slow recovery called for better monitoring than
just looking at BP, HR, minor movements and signs of sympathetic stimulation. It
has also been shown in the elderly patients that overdose of general anaesthetic drugs
may increase the incidence of postoperative delirium [12] and cognitive decline [2].
As the central nervous system is the target organ of general anaesthetic drugs, ways
of monitoring brain function directly, such as the EEG [6], were looked for.
The raw EEG is basically a continuous flow of electrical signals from 20–40 electrodes placed scattered across the full scalp. The spontaneous EEG have some characteristics which allow the expert to define which state the individual patient belongs
to along an axis from: excited → quiet awake → asleep → unconscious. However,
the complete EEG is very cumbersome to both register and interpret, thus simplified
ways of looking at EEG were looked for. One approach was to look at the frontal
EEG response (amplitude, frequency and delay) to a standardized stimulus, e.g. sound
click, as with the auditory avoked potential (AEP). Another approach was to look
for ways to interpret the spontaneous EEG by automatic computer algorithms, basically looking at frequencies, amplitudes and phase positions during different states
of depth of anaesthesia. Fortunately, the single and easily accessible frontal EEG lead
was useful in this context, as there is a frontal shift of EEG power during general
anaesthesia. However, it was soon evident that the EEG changes were complex
and not linear with dose, and EEG changes were different with different drugs and
drug mixtures, even though they produced the same clinical depth of general anaesthesia [1, 10] Also, there were problems of electrical signals from forehead muscles
interfering with the EEG and problems of surgery and nociception interfering with
the EEG algorithms. The problems were attempted to be solved by complex signal
analyses, such as with the entrophy, narcotrend, cerebral state monitor and other
devices [9]. The most successful device proved to be the BIS monitor, launched in
the 1990ties, with a single frontal EEG reading, a secret and patented algorithm for
EEG interpretation, and a simple output of a score from 0–100. With a score value
below 60 the patient was asleep, and with a value of more than 60 for more than
5 min there was a high chance of the patients being awake and aware, while values
in the 90–100 range were seen in fully awake individuals. Although the BIS has
gained widespread use and proved to reduce the incidence of awareness [3] as well
as avoiding cases of inadvertent overdosing, there are problems with this device as
well. The signal is delayed with 1–2 min, it may be disturbed by muscle activity, it
may be inaccurate with low-power EEG in the elderly, it does not work predictably
with nitrous oxide or ketamine and it is inaccurate in monitoring the antinociceptive
part of general anaesthesia.