A reply

I thank Dr Mackenzie for his interest in my article 1. Like him, in my routine clinical practice, I have used total intravenous anaesthesia (TIVA) with minimal neuromuscular blocking drugs almost exclusively. Where we differ is that I see little, if any, place for the routine use of brain monitoring with such an anaesthetic technique. Lest some draw attention to the NICE guidelines on depth of anaesthesia monitoring 2, in this context the NICE guidelines are not applicable: this type of TIVA technique was never considered. Furthermore, despite the use of the word ‘guideline’ in the document title, and the presentation of various recommendations as regards clinical practice, we are told by the three anaesthetists on the NICE project team that the ‘guidance’ is actually a ‘technology evaluation’, not a clinical guideline 3! As regards the dose of remifentanil in the study 1, after intubation and skin incision, the ‘baseline’ remifentanil target was generally in the range of 2–4 ng.ml−1 (except for those patients where an epidural was not used, in which case the target was up to 8 ng.ml−1) and this was not increased when patients responded – only the propofol target was increased. I chose not to increase remifentanil when patients responded as this would introduce another variable whose effect would be unknown, making it more difficult to adjust further infusion rates. Whether the spikes on the EMG are due to too low a dose of remifentanil or not, it does not alter the basic issue – is it appropriate to use a brain monitor whose output is affected by the EMG to guide the administration of anaesthetic drugs, especially to reduce the doses of those drugs? I believe it is quite reasonable for Mackenzie to inform his patients that they have a low risk of awareness. In the study 1, the reason there are generally no hand movements before the patient responds to commands is because the commands were frequent (every 60 s) and so the command (with the patient's name) tends to grab the patient's attention when they first regain consciousness. Playing devil's advocate, one could also argue that the command wakes up the patient, a situation like the Heisenberg uncertainty principle: the act of establishing the level of consciousness may affect the level of consciousness. As an example, the level of consciousness at this time may be similar to a post-prandial snooze when one may be aware of the television in the background (but not the programme content), but as soon as one's name is used one is awake and responding to the question. In addition, with regard to the study 1, the fact that there is no movement before this would suggest that, despite what many would regard as minimal drug use, the combination of epidural, remifentanil and propofol was sufficient for clinically adequate analgesia. In my clinical practice, when using an isolated forearm without an automated command playing, some non-specific movement of the hand/arm is often the first indication that anaesthesia is becoming light. At this point, some of these patients will respond to command, but none who respond have had recall for surgery. Part of my teaching to trainees is the minimal use of neuromuscular blocking drugs, and should the patient move any part of the body during surgery I would then ask the patient to move that particular part, since we know it is not paralysed. If the patient responds, then clearly anaesthesia is inadequate, Conversely, if the patient does not respond, that fact provides some reassurance/guidance on the current depth of anaesthesia, should further neuromuscular blocking drug become necessary for surgery. I agree that adding a frequently repeated automatic command to Mackenzie's non-isolated forearm technique would be beneficial. My only caveat would be that this command should use the patient's preferred name. Many years ago, when I used a generic (no names) tape, two patients told me postoperatively what was on the tape: one patient had been enjoying a dream and was trying to ignore the tape as it was interrupting her enjoyment; the other did not realise that the command was meant for her! Apart from these patients’ remembering the recorded message, neither had any memory of the surgical procedure. Since using a patient-specific message over the last 20 years, this experience has never recurred.