Portal de Periódicos da CAPES

Absence of memory for intra-operative information during surgery with total intravenous anaesthesia †

2001; Elsevier BV; Volume: 86; Issue: 2 Linguagem: Inglês

10.1093/bja/86.2.196

1471-6771

I.F. Russell, M. Wang,

Cardiac, Anesthesia and Surgical Outcomes

While using the isolated forearm technique, we wished to determine whether patients who did not respond to commands during general anaesthesia with a total intravenous technique (propofol and alfentanil with atracurium) had any evidence of post-operative explicit or implicit memory. Forty women undergoing major gynaecological surgery were randomized, in a double-blind design, to hear two different tapes during surgery. Psychological tests of explicit and implicit memory were conducted within 2 h of surgery. There was no evidence of implicit or explicit memory, nor any recall, in the seven women who responded to commands during surgery. We conclude that during total intravenous anaesthesia with propofol and alfentanil, there is no evidence that learning takes place when anaesthesia is adequate. Furthermore, with this anaesthetic technique, it would seem that'provided any period of patient responsiveness is short and that unconsciousness is induced rapidly again'there is no evidence of implicit or explicit memory. While using the isolated forearm technique, we wished to determine whether patients who did not respond to commands during general anaesthesia with a total intravenous technique (propofol and alfentanil with atracurium) had any evidence of post-operative explicit or implicit memory. Forty women undergoing major gynaecological surgery were randomized, in a double-blind design, to hear two different tapes during surgery. Psychological tests of explicit and implicit memory were conducted within 2 h of surgery. There was no evidence of implicit or explicit memory, nor any recall, in the seven women who responded to commands during surgery. We conclude that during total intravenous anaesthesia with propofol and alfentanil, there is no evidence that learning takes place when anaesthesia is adequate. Furthermore, with this anaesthetic technique, it would seem that'provided any period of patient responsiveness is short and that unconsciousness is induced rapidly again'there is no evidence of implicit or explicit memory. Many publications suggest that patients can recall information presented to them during surgery; these studies are discussed in detail in several reviews.1Ghoneim MM Block RI Learning and consciousness during general anesthesia.Anesthesiology. 1992; 76: 279-305Crossref PubMed Scopus (16) Google Scholar, 2Andrade J Learning during anaesthesia: a review.Br J Psychol. 1995; 86: 479-506Crossref PubMed Scopus (80) Google Scholar, 3Merikle PM Daneman M Memory for unconsciously perceived events: evidence from anesthetized patients.Conscious Cogn. 1996; 5: 525-541Crossref PubMed Scopus (67) Google Scholar, 4Wang M Learning, memory and awareness during anaesthesia.in: Adams AP Cashman JN Recent Advances in Anaesthesia and Analgesia. Churchill Livingstone, Edinburgh1998: 83-106Google Scholar However, while there is agreement that memory for new information exists after surgery, there is debate as to the level of consciousness of the patients when the intra-operative information was presented to them (i.e. were the patients adequately anaesthetized?). Merikle and Daneman3Merikle PM Daneman M Memory for unconsciously perceived events: evidence from anesthetized patients.Conscious Cogn. 1996; 5: 525-541Crossref PubMed Scopus (67) Google Scholar performed a meta-analysis of the literature and stated that there is 'considerable evidence that specific information is both perceived during anaesthesia and remembered following surgery'. However, the authors assume that 'patients who are undergoing general anaesthesia are … unconscious of all external events for the entire duration of surgery'. The studies included in the meta-analysis provide no evidence that patients were in fact anaesthetized (i.e. unconscious and unresponsive) when information was presented. This is a problem with much research in this area. Andrade2Andrade J Learning during anaesthesia: a review.Br J Psychol. 1995; 86: 479-506Crossref PubMed Scopus (80) Google Scholar states that during 'clinically adequate anaesthesia it is not yet clear that patients were actually unconscious when stimuli were presented'. At present the only direct method of detecting a responsive state during surgery is the isolated forearm technique (IFT). By using a tourniquet on one arm to prevent muscle relaxants paralysing the hand muscles, the IFT allows a patient to respond, should they become conscious during surgery, to verbal commands such as 'Open and close the fingers of your right hand.'5Tunstall ME Detecting wakefulness during general anaesthesia for caesarean section.Br Med J. 1977; 1: 1321Crossref PubMed Scopus (148) Google Scholar, 6Tunstall ME The reduction of amnesic wakefulness during general anaesthesia for caesarean section.Anaesthesia. 1979; 34: 316-319Crossref PubMed Scopus (53) Google Scholar, 7Russell IF Comparison of wakefulness with two anaesthetic regimens. Total IV v balanced anaesthesia.Br J Anaesth. 1986; 58: 965-968Crossref PubMed Scopus (54) Google Scholar, 8Russell IF Midazolam–alfentanil: an anaesthetic? An investigation using the isolated forearm technique.Br J Anaesth. 1993; 70: 42-46Crossref PubMed Scopus (94) Google Scholar, 9Russell IF Wang M Absence of memory for intraoperative information during surgery under adequate general anaesthesia.Br J Anaesth. 1997; 78: 3-9Crossref PubMed Scopus (37) Google Scholar All electronic brain monitors assess consciousness indirectly and the IFT is the 'gold standard' against which all such monitors must be tested.10Jessop J Jones JG Conscious awareness during general anaesthesia'what are we attempting to monitor?.Br J Anaesth. 1991; 66: 635-637Crossref PubMed Scopus (20) Google Scholar 11Russell IF Studies of memory during anesthesia using the isolated forearm technique.in: Ghoneim MM Awareness During Anesthesia. Butterworth Heinemann, Woburn2000Google Scholar Use of the IFT has shown that some patients can respond to commands during surgery without any other outward sign of responsiveness.5Tunstall ME Detecting wakefulness during general anaesthesia for caesarean section.Br Med J. 1977; 1: 1321Crossref PubMed Scopus (148) Google Scholar, 6Tunstall ME The reduction of amnesic wakefulness during general anaesthesia for caesarean section.Anaesthesia. 1979; 34: 316-319Crossref PubMed Scopus (53) Google Scholar, 7Russell IF Comparison of wakefulness with two anaesthetic regimens. Total IV v balanced anaesthesia.Br J Anaesth. 1986; 58: 965-968Crossref PubMed Scopus (54) Google Scholar, 8Russell IF Midazolam–alfentanil: an anaesthetic? An investigation using the isolated forearm technique.Br J Anaesth. 1993; 70: 42-46Crossref PubMed Scopus (94) Google Scholar, 9Russell IF Wang M Absence of memory for intraoperative information during surgery under adequate general anaesthesia.Br J Anaesth. 1997; 78: 3-9Crossref PubMed Scopus (37) Google Scholar If patients are given information during such an awake responsive state, it would not be surprising that sophisticated psychological tests detected evidence of memory for this information after surgery. A previous study using the IFT with inhalational anaesthesia (nitrous oxide and halothane) found no evidence of post-operative recall for information presented to patients at a time when they were unresponsive to commands.9Russell IF Wang M Absence of memory for intraoperative information during surgery under adequate general anaesthesia.Br J Anaesth. 1997; 78: 3-9Crossref PubMed Scopus (37) Google Scholar Investigations of total intravenous anaesthesia (TIVA) with etomidate7Russell IF Comparison of wakefulness with two anaesthetic regimens. Total IV v balanced anaesthesia.Br J Anaesth. 1986; 58: 965-968Crossref PubMed Scopus (54) Google Scholar and midazolam8Russell IF Midazolam–alfentanil: an anaesthetic? An investigation using the isolated forearm technique.Br J Anaesth. 1993; 70: 42-46Crossref PubMed Scopus (94) Google Scholar found 7% and 72% incidence of responsiveness, respectively, during surgery compared with zero incidence during inhalational anaesthesia.9Russell IF Wang M Absence of memory for intraoperative information during surgery under adequate general anaesthesia.Br J Anaesth. 1997; 78: 3-9Crossref PubMed Scopus (37) Google Scholar It has been argued that awareness is a greater risk with TIVA than with inhalational anaesthesia.12Breckenridge JL Aitkenhead AR Awareness during anaesthesia: a review.Ann R Col Surg Engl. 1983; 65: 93-96PubMed Google Scholar 13Wright PJ Dundee JW Attitudes to intravenous infusion anaesthesia.Anaesthesia. 1982; 37: 1209-1213Crossref PubMed Scopus (13) Google Scholar With the growing popularity of propofol-based TIVA, we wished to study the incidence of intra-operative responsiveness of patients anaesthetized with this technique and also if patients could recall information presented during surgery at a time when the patient was known to be unresponsive to commands. Nordstrom and colleagues14Nordstrom O Engstrom AM Persson S Sandin R Incidence of awareness in total i.v. anaesthesia based on propofol, alfentanil and neuromuscular block.Acta Anaesthesiol Scand. 1997; 41: 978-984Crossref PubMed Scopus (91) Google Scholar described a simple TIVA regimen using only one infusion pump and, equally importantly, they also provided data on the incidence of post-operative recall (5/2500). With such a well documented incidence of recall, we decided to base our study on this regimen. Approval for the study was obtained from the Hull and East Yorkshire ethics and clinical trials committee. The purpose of the trial was explained to women with no hearing difficulty presenting for major gynaecological surgery; their informed consent was obtained. After premedication with oral temazepam 20 mg, 1–2 h before surgery, the women were brought to the anaesthetic room. Intravenous access was obtained in the non-dominant arm with a 17 gauge cannula. An epidural catheter was then placed at L2/3 through an 18 gauge needle (lateral oblique approach, loss of resistance to 1 ml of 0.5% bupivacaine) and, after a negative aspiration test, a further 14–19 ml of 0.5% bupivacaine was injected incrementally over 5 min. The patient was then taken into the operating theatre, the various monitors were attached and, after a short period of baseline monitoring, general anaesthesia was induced. The general anaesthetic was based on the propofol/alfentanil infusion regimen described by Nordstrom and colleagues.14Nordstrom O Engstrom AM Persson S Sandin R Incidence of awareness in total i.v. anaesthesia based on propofol, alfentanil and neuromuscular block.Acta Anaesthesiol Scand. 1997; 41: 978-984Crossref PubMed Scopus (91) Google Scholar Propofol and alfentanil were mixed in the same syringe as follows: propofol 500 mg plus alfentanil 2500 μg to a total volume of 55 ml, giving final propofol and alfentanil concentrations of 9.1 mg ml−1 and 45.5 μg ml−1, respectively. Induction and anaesthesia were based on the propofol delivery from this mixture, with alfentanil in proportion. Induction was with propofol 2–2.5 mg kg−1, infused at a rate of 600 ml h−1 and maintenance was a 10, 8, 6, 4 mg kg−1 h−1 stepdown (at 10 min intervals) infusion regimen. The propofol/alfentanil infusion was controlled by a Graseby 3400 infusion pump (Graseby Medical Ltd, Watford, UK) and a multilumen extension tube with non-reflux valves (Vygon Octopus; Vygon, Ecouen, France) was included between the infusion lines and the intravenous cannula to prevent backflow. Suxamethonium 1–1.5 mg kg−1 was administered and endotracheal intubation was performed. When the action of suxamethonium began to wear off, atracurium 0.2–0.4 mg kg−1 was used for more prolonged neuromuscular block. The lungs were ventilated with an air–oxygen mixture containing 35–40% oxygen through an Oxford Penlon mark II ventilator in its non-rebreathing configuration and end-tidal carbon dioxide was maintained between 4 and 5%. For each patient, two identical looking 1 min continuous-loop cassette tapes (coded 'FG' for fruit group or 'VG' for vegetable group on the underside) were prepared by I.R. One to two minutes after skin incision, one of these tapes was played to the patients in a randomized double-blind manner. Randomization was achieved by means of codes contained in consecutively numbered brown envelopes. A colleague who was not involved in the trial had previously written the codes on to cards (20 cards for each code) and placed the cards individually into the envelopes. The envelopes were then mixed and numbered consecutively 1–40. During the trial the anaesthetic assistant opened the envelopes, wrote the envelope number and the name of the patient on each card and kept the cards in a large brown envelope until the trial was completed. The appropriate tape was chosen by the assistant, according to the code on the card, and placed in a tape recorder (Sony Professional Walkman). The remaining tape was kept hidden until the end of surgery when both tapes were then placed back in their identical plastic cases by the assistant. The tapes were personalized for each patient by inclusion of her preferred name before the command and the information. Each tape contained the same initial message which included the patient's preferred name: [Name], [Name], this is Dr Russell speaking. If you can hear me, open and close the fingers of your right hand, open and close the fingers of your right hand.' The second message differed between the tapes. One tape contained the information '[Name], [Name], this is Dr Russell speaking. Here are some special words I'd like you to remember: green pear, sharp lemon, sour gooseberry.' The second message on the other tape was '[Name], [Name], this is Dr Russell speaking. Here are some special words I'd like you to remember: frozen peas, butter beans, Chinese leaves.' To ensure that the patients heard no other extraneous information through the open dynamic headphones (Beyerdynamic DT 411), the 15 s intervals between the voice messages consisted of loud radio static. If hand movements were observed during surgery, the tape was immediately switched off and the patient spoken to directly. The patients were reassured and then the hand responses to direct commands (e.g. 'Squeeze my fingers') and to conditional commands (e.g. 'If you are in pain, squeeze my fingers twice' followed by 'If you are comfortable, squeeze my fingers once') were assessed. Anaesthesia was then deepened with a bolus of the propofol/alfentanil mixture (equivalent to propofol 0.5 mg kg−1) and the infusion rate was increased by the equivalent of propofol 1 mg kg−1 h−1 before the tape was switched on again. The infusion rate was then stepped down by 1 mg kg−1 h−1 at 10 min intervals. At the end of surgery, the tape was switched off of at the same time as the propofol/alfentanil mixture, muscle relaxant was reversed as required with atropine/neostigmine (0.6 mg/1.2 mg) and the epidural was topped up with 10 ml of 0.125% bupivacaine containing diamorphine 3 mg. Within 2 h of the end of surgery, the patients were questioned by I.R. for evidence of explicit or implicit memory. All patients were asked the same questions, without prompting, in the following order: (i) Do you remember coming into the anaesthetic room? (ii) Do you remember the epidural being inserted? (iii) Do you remember the sticky electrodes being placed? (iv) Do you remember going into the operating theatre? (v) What was the last thing you remember before going to sleep? (vi) What was the first thing you remember on waking up? (vii) Can you remember anything in between these two? (viii) Did you have any dreams? (ix) Tell me the first five fruits you can think of. (x) Tell me the first five vegetables you can think of. (xi) Now I want you to tell me the first word that comes into your head when I say a word to you: father, green, sharp, sour, frozen, butter, Chinese. Padding was placed around the dominant forearm (usually right) and an isolating tourniquet cuff wrapped around this. The standard arterial pressure cuff for routine monitoring of arterial pressure was placed on the dominant upper arm and the arm was placed on an arm board where it could be clearly observed. To restrain the hand from sudden unexpected movements, a tie was placed lightly around the fingers and the arm board. A Datex Relaxograph was used to check neuromuscular transmission under and distal to the cuff. Stimulating electrodes were placed over the ulnar and median nerves at the elbow while the sensing electrodes were placed at the wrist and over the hypothenar eminence. As soon as consciousness was lost (as judged by absence of response to command), neuromuscular transmission was checked by observing the hand contractions as the Relaxograph went through its initial set-up procedure. When the paralysis after the intubating dose of succinylcholine began to wear off, the isolating cuff was inflated just before giving atracurium. The cuff remained inflated for 20 min before deflation. Whenever further atracurium was required, this same inflation/deflation process was followed. Neuromuscular transmission under and distal to the tourniquet was confirmed by observation of the hand and finger responses to the 'train of four' stimuli delivered every 20 s from the Relaxograph. Although the Relaxograph assesses the electrical activity of muscle groups, it was the observed hand responses that were of importance in ensuring that neuromuscular transmission was sufficient for the patient to respond. Chi-squared tests and signal detection analysis15Donaldson W Glathe H Signal detection analysis of recall and recognition memory.Can J Psychol. 1970; 24: 42-56Crossref Google Scholar were used as appropriate. Forty women were recruited into the study. Their ages, weights and duration of tape exposure, together with an indication of the surgical procedures, are shown in Table 1. The difference in the mean duration of tape exposure in the two groups just failed to reach statistical significance (P=0.06). This difference was almost entirely due to one patient in the 'fruit' group undergoing very long surgery (lasting 280 min, 2.8 sds above the mean).Table 1Age, weight and tape duration for the two groups together with the type of surgery. †Mean (range). *Mean (sd)GroupP value'Vegetable''Fruit'Age (yr)46.4 (35–45)†40.3 (26–57)†0.02Weight (kg)69.1 (15.7)*66.1 (8.9)*0.5Tape duration (min)78.9 (35.5)*109.3 (58.8)*0.06Type of gynaecological surgery (n)Abdominal procedure1918Vaginal hysterectomy12 Open table in a new tab No patient had explicit memory for any aspect of surgery or of the taped commands. Tables Table 2, Table 3, Table 4, Table 5, Table 6 indicate the results of tests of implicit memory from all 40 patients. These tables provide the 'hits' data and serial position scores for lists of both three and five exemplars. In the word association test (Table 6), there was only one 'hit', the word pair 'frozen peas'.Table 2Number of vegetable and fruit exemplar 'hits' when a list of five exemplars was requested; the total number of possible 'hits' is 60GroupVegetable 'hits' in list of five vegetablesFruit 'hits' in list of five fruitsVegetable (n=20)814Fruit (n=20)714 Open table in a new tab Table 3Number of vegetable and fruit exemplar 'hits' when a list of three exemplars was requested; the total number of possible hits is 60GroupVegetable 'hits' in list of three vegetablesFruit 'hits' in list of three fruitsVegetable (n=20)56Fruit (n=20)75 Open table in a new tab Table 4Serial position score based on list of five exemplars (score=5 for first position down to 1 for fifth position). Maximum score per patient= 12 (5+4+3). Maximum score per group=240ScoreGroupVegetableFruitVegetable (n=20)2337Fruit (n=20)2332 Open table in a new tab Table 5Serial position score based on a list of three exemplars. Score=3 for first position down to 1 for third position. Maximum score per patient= 6 (3+2+1). Maximum score per group=120ScoreGroupVegetableFruitVegetable (n=20)1010Fruit (n=20)109 Open table in a new tab Table 6Word association 'hits'. Total possible hits per group was 60. *All seven of the word association 'hits' were 'frozen peas'Word association hitsGroupVegetableFruitVegetable (n=20)4*0Fruit (n=20)3*0 Open table in a new tab Seven women, two in the 'fruit' group and five in the 'vegetable' group, responded to commands at some stage during surgery. Among them, these women responded a total of 12 times at propofol infusion rates ranging from 4 to 7 mg kg−1 h−1 and at times ranging from 34 to 125 min into surgery (Table 7). On only one occasion did one of these women express her discomfort (Table 7). There was no evidence that these seven patients had a greater rate of recall than non-responding patients (Table 8).Table 7Group allocation and weight of patients who responded to commands, the times of the responses and the infusion rates at the time of the responses from seven patients with verified responses to commands. *At this time the patient indicated she was uncomfortable; the surgeon had just inserted a self-retaining retractor and was performing a thorough exploration of the upper abdominal cavityGroup of words/ patientWeightTimePropofol infusion rate(kg)(min)(mg kg–1 h–1)Vegetable groupPatient a5534*7645Patient b74554805Patient c7943Patient d67654Patient e83554755Fruit groupPatient f6466410551256Patient g66604 Open table in a new tab Table 8Number (proportion of total possible) of hits in those patients awake and responding during surgery compared with patients who did not respond during surgery. *This difference is not significant (P=0.48)Category of patients (n)Fruit or vegetable exemplars: correct hitsWord association hitsResponsive (7)5 (24%)*1 (6%)Unresponsive (33)17 (17%)*6 (6%) Open table in a new tab None of the women had any side effects from the use of the isolating tourniquet. There is debate about whether familiar or unfamiliar words are more likely to be returned in memory tests,2Andrade J Learning during anaesthesia: a review.Br J Psychol. 1995; 86: 479-506Crossref PubMed Scopus (80) Google Scholar so to cover both possibilities the exemplars presented to our patients spanned a wide range: from third through seventh to below fifteenth in their respective lists (the position of the exemplars in their respective lists had previously been established in our population of women presenting for major gynaecological surgery16Charlton PFC Implicit and explicit memory for general anaesthesia. MSc Thesis. University of Hull, Hull1991Google Scholar). We asked each patient for five exemplars and analysed both three-exemplar and five-exemplar lists. This allowed a wider range of familiarity/unfamiliarity to be assessed, increasing the chances of a positive effect. Despite this, we found no evidence of explicit or implicit memory for information presented during surgery under general anaesthesia. This result confirms previous published8Russell IF Midazolam–alfentanil: an anaesthetic? An investigation using the isolated forearm technique.Br J Anaesth. 1993; 70: 42-46Crossref PubMed Scopus (94) Google Scholar 9Russell IF Wang M Absence of memory for intraoperative information during surgery under adequate general anaesthesia.Br J Anaesth. 1997; 78: 3-9Crossref PubMed Scopus (37) Google Scholar and as yet unpublished studies11Russell IF Studies of memory during anesthesia using the isolated forearm technique.in: Ghoneim MM Awareness During Anesthesia. Butterworth Heinemann, Woburn2000Google Scholar from this department but is contrary to many other results discussed in reviews.1Ghoneim MM Block RI Learning and consciousness during general anesthesia.Anesthesiology. 1992; 76: 279-305Crossref PubMed Scopus (16) Google Scholar, 2Andrade J Learning during anaesthesia: a review.Br J Psychol. 1995; 86: 479-506Crossref PubMed Scopus (80) Google Scholar, 3Merikle PM Daneman M Memory for unconsciously perceived events: evidence from anesthetized patients.Conscious Cogn. 1996; 5: 525-541Crossref PubMed Scopus (67) Google Scholar, 4Wang M Learning, memory and awareness during anaesthesia.in: Adams AP Cashman JN Recent Advances in Anaesthesia and Analgesia. Churchill Livingstone, Edinburgh1998: 83-106Google Scholar One possible reason for our negative findings could be that the memory tests we used were unsuitable. We do not believe this is a valid reason. In a literature review carried out in 1990,16Charlton PFC Implicit and explicit memory for general anaesthesia. MSc Thesis. University of Hull, Hull1991Google Scholar seven studies were found which investigated word priming during general anaesthesia. Three of these studies obtained statistically significant evidence of word priming; in contrast to the four negative studies, these three positive studies used category association to cue recall of implicit memory. Hence this method (category association) was chosen for the current study. In addition, our experimental design uses two different probes (category association and word pair association) to cue the same primed words to increase the chances of obtaining an effect. No statistical adjustment was employed for these multiple tests, thus increasing the probability of a type I error but, despite this, we did not obtain significant results. Other possible factors to consider which could contribute to the lack of recall are the early post-operative testing and the use of temazepam premedication. The effects of temazepam on memory in this situation have not been studied but all patients remembered being in the anaesthetic room and having the epidural catheter inserted, they all remembered the attachment of electronic monitoring electrodes and being taken into the operating theatre and placed on the operating table. In previous studies, temazepam was used as premedication when patients had recall.7Russell IF Comparison of wakefulness with two anaesthetic regimens. Total IV v balanced anaesthesia.Br J Anaesth. 1986; 58: 965-968Crossref PubMed Scopus (54) Google Scholar 9Russell IF Wang M Absence of memory for intraoperative information during surgery under adequate general anaesthesia.Br J Anaesth. 1997; 78: 3-9Crossref PubMed Scopus (37) Google Scholar One could argue that the presence of residual temazepam in the immediate post-operative period should increase the probability of recall because of state-dependent memory effects. Our patients were fully responsive to verbal command and conversation at the time of testing and other researchers, using similar testing, found implicit memory when patients were tested 30–179 min after operation.17Jelicic M Bonke B Wolters G Phaf RH Implicit memory for word presented during anaesthesia.Eur J Cogn Psychol. 1992; 4: 71-80Crossref Scopus (32) Google Scholar In their meta-analysis, Merikle and Daneman3Merikle PM Daneman M Memory for unconsciously perceived events: evidence from anesthetized patients.Conscious Cogn. 1996; 5: 525-541Crossref PubMed Scopus (67) Google Scholar found that early testing was strongly associated with significant implicit memory effects. We believe we tried to favour conditions that would increase the probability of finding evidence for a priming effect. The most likely reason why we did not find priming is that anaesthesia was adequate. With the IFT we knew whether patients were responsive or not when the information was being played to them. In the majority of memory studies unconsciousness has either been assumed or been determined based on 'normal' clinical signs. In the presence of muscle relaxants, it is difficult to assess the depth of anaesthesia. The usual clinical signs of anaesthesia (heart rate, arterial pressure, sweating, tear production) are unreliable indicators of consciousness during surgery.18Cullen DJ Eger EI Stevens WC et al.Clinical signs of anaesthesia.Anesthesiology. 1972; 36: 21-36Crossref PubMed Scopus (55) Google Scholar, 19Moerman N Bonke B Oosting J Awareness and recall during general anesthesia.Anesthesiology. 1993; 79: 454-464Crossref PubMed Scopus (287) Google Scholar, 20Ponte J Neuromuscular blockers during general anaesthesia. Less may be better.Br Med J. 1995; 310: 1218Crossref PubMed Scopus (6) Google Scholar Virtually all the original clinical signs described by Snow21Snow J On the Inhalation of Vapour of Ether. Churchill, London1847Google Scholar and developed by Geudel22Guedel AE Inhalation Anaesthesia'a Fundamental Guide. MacMillan, New York1937Google Scholar rely on muscle activity: cardiovascular signs are not mentioned. When muscle relaxants are used, these original clinical signs are of limited, if any, value. The origins of using cardiovascular indices to gauge depth of anaesthesia are unclear and we are not aware of any study which has shown them to be of value in detecting consciousness. On the other hand, several studies using different anaesthetic techniques in conjunction with the IFT have shown clearly that clinical signs cannot be used to predict whether or not patients are awake and responsive during surgery.7Russell IF Comparison of wakefulness with two anaesthetic regimens. Total IV v balanced anaesthesia.Br J Anaesth. 1986; 58: 965-968Crossref PubMed Scopus (54) Google Scholar, 8Russell IF Midazolam–alfentanil: an anaesthetic? An investigation using the isolated forearm technique.Br J Anaesth. 1993; 70: 42-46Crossref PubMed Scopus (94) Google Scholar, 9Russell IF Wang M Absence of memory for intraoperative information