- Email: [email protected]
Awareness during anaesthesia: a prospective case study
ARTICLES
Awareness during anaesthesia: a prospective case study
Rolf H Sandin, Gunnar Enlund, Peter Samuelsson, Claes Lennmarken
Summary Background Patients who are given general anaesthesia are not guaranteed to remain unconscious during surgery. Knowledge about the effectiveness of current protective measures is scarce, as is our understanding of patients’ responses to this complication. We did a prospective case study to assess conscious awareness during anaesthesia. Methods 11 785 patients who had undergone general anaesthesia were interviewed for awareness on three occasions: before they left the post-anaesthesia care unit, and 1–3 days and 7–14 days after the operation. Findings We identified 18 cases of awareness and one case of inadvertent muscle blockade that had occurred before unconsciousness. Incidence of awareness was 0·18% in cases in which neuromuscular blocking drugs were used, and 0·10% in the absence of such drugs. 17 cases of awareness were identified at the final interview, but no more than 11 would have been detected if an interview had been done only when the patients left the post-anaesthesia care unit. Four non-paralysed patients recalled intraoperative events, but none had anxiety during wakefulness or had delayed neurotic symptoms. This finding contrasts with anaesthesia with muscle relaxants, during which 11 of 14 patients had pain, anxiety, or delayed neurotic symptoms. After repeated discussion and information, the delayed neurotic symptoms resolved within 3 weeks in all patients. Analysis of individual cases suggests that a reduced incidence of recall of intraoperative events would not be achieved by monitoring of end-tidal anaesthetic gas concentration or by more frequent use of benzodiazepines. Interpretation The inability to prevent awareness by conventional measures may advocate monitoring of cerebral activity by neurophysiological techniques. However, the sensitivity of such techniques is not known, and in the light of our findings, at least 861 patients would need to be monitored to avoid one patient from suffering due to awareness during relaxant anaesthesia. Lancet 2000; 355: 707–11 See Commentary page xxx
Department of Anaesthesia and Intensive Care, Länssjukhuset, Kalmar, Sweden (R H Sandin MD, P Samuelsson MD); and Department of Anaesthesia and Intensive Care, Vrinnevisjukhuset, Norrköping, Sweden (G Enlund MD, C Lennmarken MD) Correspondence to: Dr Rolf Sandin, Department of Anaesthesia and Intensive Care, Länssjukhuset, S-39185 Kalmar, Sweden (e-mail: [email protected])
THE LANCET • Vol 355 • February 26, 2000
Introduction Recall of intraoperative events in patients under general anaesthesia is rare (0·1–0·7%), although few prospective studies have assessed the incidence of awareness during surgical procedures.1–4 Up to 54% of patients worry about the possibility of pain, paralysis, and mental distress during surgery.5 Previous studies have not recorded important background data, such as whether neuromuscular-blocking drugs were used or whether end-tidal anaesthetic gas concentration (ETAGC) was monitored. Also, memory for awareness may be delayed,1 and follow-up may have been too short to identify all cases. We did a prospective study to assess patients’ recall of awareness during surgery.
Patients and methods This study, which was approved by the Ethics Committee of Linköping University Hospital, was done in two Swedish hospitals. Clinical routine at the two hospitals includes an interview for awareness on discharge from the post-anaesthesia care unit in all patients older than 15 years who have had general anaesthesia. This interview, modified from Brice and colleagues6 and described elsewhere,1 is given by the nurses who attended the patients, and is recorded in the anaesthesia record. Dedicated nursing staff for the study recorded data from the routine interview in the post-anaesthesia care unit and repeated the interview 1–3 days and 7–14 days after anaesthesia. Inpatients were interviewed face-to-face, whereas patients who had left the hospital were contacted by telephone. Patients with possible memories of intraoperative events at any of the three interviews were given an additional interview to obtain data on awareness. All four investigators were involved in the interviewing of patients, and together they classified cases into one of three categories according to the likelihood of true recall: definite cases required that the recalled event was confirmed by attending personnel; for probable cases, the investigators were convinced that the memory was real, but no confirmation could be obtained; and possible cases, who were unable to recall any event definitely indicative of true awareness. A reported dream alone was not classified as awareness. Patients were enrolled between 1997 and 1998. Inclusion criteria were as follows: general anaesthesia with laryngeal mask airway, tracheal intubation, or jet ventilation; age of at least 16 years; ability to communicate in Swedish or English; no overt psychiatric disorder thought to interfere with the reliability of the interview; ability to attend at least one of the subsequent interviews; and agreement to participate. All patients who fulfilled the inclusion criteria were entered in the study after informed consent had been obtained. Planned and emergency surgery were included. Of 12 179 eligible patients, 394 were not included because they could not be contacted for either of the last two interviews, leaving 11 785 patients to be assessed. We assessed the cases of awareness or awake paralysis with reference to whether additional monitoring, education, or increased vigilance could have prevented awareness. Prevention was assessed only if an alternative method would have avoided wakefulness under the same circumstances. Additional neurophysiological monitoring was not allowed in any patient
707
ARTICLES
Total (n=11 785)
Patients with awareness (n=19)
Demography Age (years) Height (cm) Weight (kg) ASA physical status Male/female
48 (19) 170 (9) 74 (15) 1·4 (0·6) 4732/7053
45 (18) 170 (9) 78 (16) 1·4 (0·6) 7/12
Surgery Duration of anaesthesia (min) Elective/acute and emergency surgery*
99 (65) 9388/2397
95 (66) 14/5
ASA=American Society of Anesthesiologists. Data are means (SD) or numbers. The patient with awake paralysis is included among the awareness cases. *No cardiac surgery, transplantation, or neurosurgery was done at the hospitals during the study period. All trauma surgery is included, except for neurotrauma.
Table 1: Baseline data who was likely to fulfil the inclusion criteria, since our aim was to assess the present situation and the possible need for additional monitoring.
Results Baseline data of the 11 785 patients included in the study are shown in table 1. Anaesthesia and clinical procedures 11 440 (97·1%) patients had been given an opioid before anaesthesia and 2084 (17·7%) had received a benzodiazepine. No patient was given scopolamine. Anaesthesia was induced with propofol in 6473 (54·9%) patients, thiopentone in 5248 (44·5%), sevoflurane in 36 (0·31%), and ketamine in 28 (0·24%). Potent inhalation agents were used in 11 454 (97·2%) patients (sevoflurane in 7642 [66·7%], isoflurane in 3810 [33·3%], desflurane in two). Nitrous oxide was used in 11 034 (93·6%) patients given inhalation anaesthesia. Total intravenous anaesthesia with propofol was used Patient Demography
Surgery
Sex Age Weight Height ASA (years) (kg) (cm) grade
in 325 (2·8%) patients, 288 (89·2%) without nitrous oxide. Total intravenous anaesthesia with thiopentone was used in two patients. Intraoperative epidural analgesia with local anaesthetics was used in 346 (4·4%) of 7796 intubated patients. ETAGC was monitored in 9528 (83·2%) patients with inhalational anaesthesia. The airway was maintained by endotracheal intubation in 7796 (66·2%) patients (by neuromuscular block in 7737) and by laryngeal mask airway in 3977 (33·8%) patients (by neuromuscular block in four). Jet ventilation through a bronchoscope was used in 12 patients, all of whom received neuromuscular block. Among 7753 patients given neuromuscular block, 672 (8·7%) received only a single dose of suxamethonium for intubation. Recall of awareness 18 patients (11 [0·16%] women and seven [0·15%] men) reported awareness (six definite, eight probable, and four possible). One further woman (patient 4) had awake paralysis. Demographic data and details of surgery and anaesthesia for the 19 patients with awareness or awake paralysis are shown in table 2. Descriptions of patients’ experiences are listed in table 3. 14 patients with awareness had had previous surgery, of whom one had previously had awareness during surgery (patient 19). Neuromuscular block was used in 14 of 18 patients who reported awareness. However, none of the patients anaesthetised by total intravenous anaesthesia reported awareness during surgery. The incidence of awareness was similar for men and women. Overall incidence was 0·18% when neuromuscular block was used, and 0·10% in the absence of relaxant drugs.
Anaesthesia Co-induction
Induction
Maintenance N2O
ETAGC NMB monitoring
Benzodiazepine
Opioid
Propofol
Isoflurane
Yes
No
Rocuronium ETT
130 80
None None
Opioid Opioid
Propofol Sevoflurane
Isoflurane Sevoflurane
No Yes
No Yes
Vecuronium ETT None LMA
115
Opioid phentiazine
Opioid
Thiopentone
Sevoflurane
Yes
Yes
Rocuronium ETT
60
Opioid phentiazine None None
Opioid
Thiopentone
Sevoflurane
Yes
Yes
Rocuronium ETT
None Opioid
Thiopentone Thiopentone
Sevoflurane Sevoflurane
Yes Yes
Yes Yes
Opioid phentiazine None None
Opioid
Propofol
Sevoflurane
Yes
Yes
Suxamethon ETT Suxamethon, ETT rocuronium None LMA
Opioid Opioid
Thiopentone Thiopentone
Sevoflurane Sevoflurane
Yes Yes Yes Yes
Benzodiazepine Opioid phentiazine Opioid
Opioid Opioid
Thiopentone Thiopentone
Isoflurane Sevoflurane
Yes Yes
Yes Yes
Suxamethon Suxamethon, rocuronium Rocuronium Rocuronium
Opioid
Thiopentone
Sevoflurane
Yes
Yes
Rocuronium ETT
160
Opioid
Opioid
Propofol
Sevoflurane
Yes
Yes
None
59 45
Opioid None
Opioid None
Propofol Ketamine
Sevoflurane Isoflurane
Yes Yes
Yes Yes
None LMA Suxamethon ETT
Benzodiazepine
Opioid
Thiopentone
Isoflurane
Yes
Yes
Suxamethon ETT
None Opioid
Opioid Opioid
Propofol Thiopentone
Isoflurane Sevoflurane
Yes Yes
No Yes
Vecuronium ETT Rocuronium ETT
Duration of Premedication anaesthesia (min)
1
F
39
62
158
1
2 3
M F
60 34
84 60
173 162
1 1
4
F
51
85
174
1
5
F
21
56
161
1
Laparoscopic sterilisation Eye surgery Reconstructive breast surgery Laparotomy, resection of renal cyst Tonsillectomy
22
6 7
F M
30 23
73 87
169 185
1 1
Caesarean section Appendectomy
45 60
8
M
30
90
180
1
Hand surgery
55
9 10
F M
67 84
80 76
166 180
1 3
Hand surgery Hemicolectomy
60 140
11 12
M F
73 56
88 104
170 168
2 1
Colostomy Parathyroidectomy
246 248
13
F
40
65
162
1
55
14
M
31
54
172
2
15 16
M F
47 28
106 ?
187 ?
2 2
17
F
62
98
163
2
18 19
F F
31 45
64 70
162 163
1 1
Laparoscopic sterilisation Osteosynthesis, arm fracture Hand surgery Retrieval of retained placenta Laparoscopic cholecystectomy Tonsillectomy Laparoscopy
130 40 55
Airway
ETT ETT ETT ETT
LMA
ASA=American Society of Anesthesiologists; ETT=endotracheal intubation; LMA=laryngeal mask airway; NMB=neuromuscular block; suxamethon-=suxamethonium.
Table 2: Details of demographic data and surgery from patients who experienced awareness during surgery
708
THE LANCET • Vol 355 • February 26, 2000
ARTICLES Patient
Experience
1
Suddenly noticed endotracheal intubation, but felt no discomfort. Was scared. Tried to move but found it impossible. Heard somebody say “she is not asleep”.
2
Suddenly aware that he was awake, but felt no discomfort. Tried to move his feet. Heard somebody say “he is awake”.
3
Recognised the surgeon’s voice, heard noise from instruments, understood the situation, found herself unable to move. She thought “it will soon be over”, and soon became unconscious again.
4
Paralysed before she became unconscious. Realised something was wrong, although did not understand what. Tried to move but was unable. Repeated explanations were required. Anxiety gradually disappeared over 2 weeks.
5
Soon after anaesthetist had said “it is going to taste like garlic”, she heard “now it is time to put the tube down”, and also remembered another discussion. Soon fell unconscious again.
6
Probably only temporarily sedated by induction dose. Was suddenly widely awake and felt a cramping pain in the neck (suxamethonium).
7
Regained consciousness and found himself in some kind of turbulence. Felt pain in throat and felt like he was suffocating, but still did not panic. Understood what was happening.
8
Regained consciousness. Heard surgeon ask “can I cut here?” Felt someone manipulate wounded arm. No pain and no anxiety. No further memories.
9
Felt incision, realised what was happening and soon after lost consciousness.
10
Saw people dressed in green (operating room staff), felt severe abdominal pain, tried to talk and move, but was unable. Very agitated immediately after anaesthesia, said he had had a bad dream.
11
Heard voices, but could not recall what was said.
12
Heard someone say “is very narrow”. The patient did have a small mouth. After surgery, a tooth was found to have been damaged, possibly during intubation. However, a definite confirmation could not be obtained.
13
Realised she was awake when she should not be. She tried to get attention, tried to talk and move but was unable. Heard voices and noise from surgical instruments.
14
Suddenly saw a big, round lamp. Tried to move his arms and scream, but was unable. Had impression that surgeon was about to finish. Soon was unconscious again.
15
Suddenly felt pain in operated hand, understood situation, but did not worry since he realised he was “fading away” again.
16
Saw people dressed in green, saw tombstones, thought she was attending her own funeral. Heard voices speak slowly. No impression to have been unconscious at all. Anxiety gradually disappeared over 3 weeks.
17
Sharp pain and a sense of “darkness”. Postoperative anxiety disappeared over following 3 weeks.
18
Heard someone say “not quite finished yet . . . give more” and heard noise from surgical instruments. Felt mask pressed onto her face (on awakening?).
19
Heard a voice call for “more anaesthetics”. Saw a lamp.
Table 3: Patients’ experiences of awareness during surgery
Patient
1 2 3 4* 5 6* 7 8 9 10 11 12 13 14 15 16 17 18 19
Perception
Motor function
Six patients recalled awareness at the first interview, before they left the post-anaesthesia care unit. In another patient, later found to have had awareness, this interview was taken to be inconclusive (patient 17). In four patients who later reported memories, no result from the first interview was recorded. 12 patients had recall at the second interview, including five of the six patients who had memories at the first interview. One patient had forgotten about the intraoperative events that she had reported at the first interview (patient 12). Awareness was confirmed in three of four patients in whom no result from the first interview had been recorded. One of these patients firmly denied having had any memories until just before the second interview (patient 19). In addition, three patients who denied recall of awareness at the first interview, reported at the second interview that they had had awareness during surgery. The patient who gave an inconclusive answer at the first interview also reported recall. 17 of 18 patients with awareness had recall of events at the third interview. Five patients had not previously been identified as having been aware. In one of these patients, no result from the first interview was recorded, and no recall was evident at the second interview (patient 5). The patient who had forgotten the memories that she disclosed at the first interview could remember them again (patient 12). One patient who had no recall at the first interview (patient 13), but remembered intraoperative events at the second, had once again forgotten (12 days after surgery), although 24 days after surgery she had detailed memories. One patient thought he would not have remembered at all had he not been questioned (patient 15). The patient who had awake paralysis remembered at all three interviews about being paralysed before anaesthesia. Auditory and visual perceptions were present in 12 and five patients with awareness, respectively (table 4). Pain was reported by seven patients, two of whom reported severe pain (patients 10 and 17). 11 patients with awareness were convinced that they had tried to move, but only one realised he was able to do so (patient 2). Two patients who were unable to move had received no neuromuscular block (patients 3 and 14), although one of these patients had restrained arms (patient 14). Three patients denied making any attempt to move despite their understanding of the situation.
Mental reaction
Assessment
Auditory
Visual
Pain
Tried to move
Able to move
Immediate understanding
Immediate anxiety
Delayed symptoms
Confirmed event
Probability
Detection at interview
Yes Yes Yes Yes Yes ? Yes Yes No ? Yes Yes Yes No No Yes No Yes Yes
No ? Yes Yes No No No No No Yes ? No No Yes No Yes No No Yes
No No No No No Yes Yes No Yes Yes Yes No No No Yes No Yes No No
Yes Yes Yes Yes Yes Yes Yes No ? Yes Yes No Yes Yes No Yes No No No
No Yes No No No No No .. ? No No .. No No .. No .. .. ..
Yes Yes Yes No Yes Yes Yes Yes Yes No Yes No Yes No Yes No No No Yes
Yes No No Yes Yes Yes No No No No Yes Yes Yes No No Yes Yes No No
No No No Yes Yes No No No No ? No No No No No Yes Yes No No
Yes Yes Yes Yes Yes Yes Yes No No No No No No No No No No No No
Definite Definite Definite Definite Definite Definite Definite Probable Probable Probable Probable Probable Probable Probable Probable Possible Possible Possible Possible
1, 2, 3 1, 2, 3 2, 3 1, 2, 3 3 1, 2, 3 3 2, 3 3 1, 2, 3 2, 3 1, 3 2 3 3 1, 2, 3 2, 3 2, 3 2, 3
*Probably never lost consciousness before being paralysed (definitely in the case of patient 4).
Table 4: Details of awareness obtained by interview
THE LANCET • Vol 355 • February 26, 2000
709
ARTICLES
Of 14 patients with recall and who had received neuromuscular block, nine immediately realised their situation, and six were anxious. Only one of these six patients (patient 5) had delayed symptoms (nightmares and anxiety). Among the five paralysed patients who did not understand their situation during wakefulness, three felt anxiety, two of whom also experienced delayed symptoms. The patient who had awake paralysis did not realise the situation, felt anxiety, and experienced delayed symptoms. Among the four patients with delayed symptoms after awareness or awake paralysis, only one had pain. No patient who was aware during non-relaxant anaesthesia felt anxiety during wakefulness. Three of these patients understood their situation during wakefulness and all four were free from delayed symptoms. Of the 14 awareness cases who had received neuromuscular block, 11 had unpleasant effects from their experience for one or more reasons. If the patient who had awake paralysis is taken into account, ten patients experienced anxiety during wakefulness. Delayed symptoms were evident in three of the four patients who had not been able to understand the situation, but in only one of the six who understood what was happening. Among the six paralysed patients who reported pain, only one patient had delayed symptoms (patient 17), and this patient was also among those who felt anxiety and was unable to understand the situation. All patients who experienced awareness or awake paralysis were offered repeated discussion and explanation according to their own desire, and within 3 weeks all patients declared that they were satisfied with the explanations, that they did not need any further contact, and that all their delayed symptoms had disappeared. However, one patient thought she might become worried again if she needed surgery in the future. Probable reasons for awareness For patient 2, the vaporiser was turned off after a refill during anaesthesia. For patient 4, the anaesthetist administered rocuronium before thiopentone, and assumed a time of onset similar to pancuronium. Patient 5 probably received an insufficient induction dose of thiopentone or opioid, or both, to prevent arousal during laryngoscopy. For patient 6, the anaesthetist noticed too late that thiopentone had been partially injected backwards into the fluid line, and the patient was aroused by suxamethonium-induced fasciculations. Patient 7’s awareness was caused by a difficult intubation and this was probably the reason in patient 12 as well. Patient 16 was the only patient classified as possible who had recall at all three interviews. Her memories were probably related to dreams caused by ketamine, although influences from real events cannot be ruled out. Patient 18 was suspected to have begun awakening too early. Patient 19 described a memory very similar to when she suffered from awareness 3 years previously. However, she was convinced she had been aware again. In the remaining ten patients, no obvious reason for awareness could be identified. We assumed that a neurophysiological monitoring technique capable of discrimination between consciousness and unconsciousness could have been helpful to prevent, at most, 16 of the 18 cases of awareness (not patients 6 and 16). Monitoring of
710
ETAGC was used in 15 patients with recall, and did not help in these cases. We assumed that monitoring would have been helpful in only one of the three cases of awareness among the 1928 patients without ETAGC (patient 2). The anaesthetist’s insufficient knowledge about the time of onset for rocuronium led to the case of awake paralysis (patient 4), and lack of vigilance contributed in two cases of awareness (patients 2 and 6).
Discussion Our findings and our inability to find non-paralysed patients who had unpleasant effects during or after wakefulness suggest that awareness among patients without neuromuscular block is not a major concern, as long as conversation between operating-room staff remains respectful to the patient. Studies on awareness should therefore distinguish clearly between relaxant and non-relaxant anaesthesia. If the first interview had been done after the patients had left the post-anaesthesia care unit, no more than 11 and probably only nine of the 18 awareness cases would have been identified. There is only one previous study in which an interview was undertaken more than 72 h after anaesthesia. In that study, one patient did not recall dreaming or intraoperative events soon after anaesthesia, but remembered an unpleasant dream the next day, and had recall of operative events after 8 days.1 However, in our study, patients’ memories did not always remain constant or improve with time. In two patients, a reported memory of intraoperative events was temporarily forgotten at the next interview. A similar case was reported in a study in which one patient had recall soon after anaesthesia, but later forgot about his experiences.2 Obviously, there is no time when all cases of awareness will be detected reliably, but most previous studies may have underestimated the true number of patients with recall.1–4 This issue is not only academic, since among the three awareness cases with delayed neurotic symptoms in our study, one was not detected until day 10, and another patient may have been overlooked because he gave a vague and inconclusive answer at the first interview (patient 17). Improved knowledge and vigilance by the anaesthetist may have prevented two of the cases of awareness and also the case of awake paralysis. However, in this type of study, it is not possible to detect less obvious cases of poor anaesthetic performance, and improved education may be of greater importance to prevent awareness than indicated by the three situations we identified. There was no difference in the incidence of awareness among patients who had been given a benzodiazepine before anaesthesia and those who had not. Benzodiazepines are, in many cases, valuable for relieving anxiety before surgery, but they are also associated with a variable degree of anterograde amnesia, and their widespread use as premedication may be because of this reputation, which is shared with scopolamine. However, there is no evidence that effects of benzodiazepines or scopolamine are better than the amnesic effects of ordinary anaesthetics.7 More importantly, we believe that to carry out anaesthesia in a way that would require deliberate pre-emptive amnesia for intraoperative experiences is ethically unsound. The incidence of awareness among patients with neuromuscular block was similar irrespective of whether ETAGC was measured or not. This result is weakened
THE LANCET • Vol 355 • February 26, 2000
ARTICLES
by the fact that the patients were not randomised by whether ETAGC was used. However, it is striking that ETAGC was monitored in 15 of the 18 cases of awareness, and it is by no means clear how many, if any, of the remaining three cases would have been avoided if this technique had been used. Clearly, if the accepted ETAGC is too low, this technique will be of no value in the prevention of awareness. A safe concentration of anaesthetic gas has never been established in a sufficiently large study. Institutional guidelines at the two departments involved in our study recommend at least 1·0 MACskin incision (a measure of anaesthetic potency8) and preferably 1·3 MACskin incision of halogenated volatile anaesthetic for anaesthesia supplemented with opioids and nitrous oxide. Even if this recommendation is not sufficient to prevent awareness in some patients, a more vigorous administration of anaesthetic gas may not be possible in all cases due to haemodynamic side-effects. However, we do not believe that a general use of low concentrations of anaesthetic gas at the two departments in our study is the reason for the inability of ETAGC monitoring to prevent awareness, since the overall incidence of recall is one of the lowest reported. Our assumption that factors other than an inability to measure ETAGC are responsible for cases of awareness in contemporary anaesthetic practice is strengthened by the fact that a similar incidence of 0·2% was found with total intravenous anaesthesia during which monitoring of ETAGC is of no value.1 Thus, the extent to which ETAGC monitoring reduces the incidence of awareness is seriously questionable, and from our findings this monitoring may not help to reduce this incidence further. Monitoring techniques with alleged capability to identify intraoperative wakefulness have been proposed.9 However, no such monitoring was allowed in our study. Whether techniques such as bispectral index and midlatency auditory evoked potentials are able to reduce the incidence of awareness will probably never be proven in a randomised controlled study since about 50 000 patients would need to be included to show a significant reduction in incidence from 0·2% to 0·1%. To decide whether substantial economical investment in technical devices to reduce the incidence of awareness is justified,
THE LANCET • Vol 355 • February 26, 2000
it is essential to know not only the true current incidence of awareness, but also the proportion of patients who suffer during and after wakefulness. When the cases of awareness found in our study were scrutinised, we assumed that a neurophysiological monitoring technique with 100% sensitivity could have been helpful to avoid at most 12 of the 14 cases that occurred during relaxant anaesthesia. Nine of those 12 patients suffered in one or more ways from their experience. This means that at least 861 patients must be monitored to avoid one case of suffering due to awareness in relaxant anaesthesia, provided that the sensitivity of that particular technique is 100%, including human performance. Contributors Rolf Sandin designed the study and analysed the data. Gunnar Enlund and Peter Samuelsson were responsible for quality control, and interviewed the study participants. Claes Lennmarken and Rolf Sandin were responsible for fund raising. All investigators wrote the paper.
Acknowledgments We thank Ulla Agne, Ann-Mari Engström, Wivi Forsberg, Mats Hultman, and Gunnar Svensson for collecting and recording the data. The study was supported by a grant from Forskningsrådet i Sydöstra Sverige.
References 1
2
3
4
5
6
7 8 9
Nordström O, Engström AM, Persson S, Sandin R. Incidence of awareness in total i.v. anaesthesia based on propofol, alfentanil and neuromuscular blockade. Acta Anaesthesiol Scand 1997; 41: 978–84. Ranta SOV, Laurila R, Saario J, Ali-Melkkilä T, Hynynen M. Awareness with recall during general anesthesia: incidence and risk factors. Anesth Analg 1998; 86: 1084–89. Liu WH, Thorp TA, Graham SG, Aitkenhead AR. Perception and memory during general anaesthesia. Anaesthesia 1991; 46: 435–37. Myles PS, Williams DL, Hendrata M, Anderson H, Weeks AM. Patient satisfaction after anaesthesia and surgery: results of a prospective survey of 10 811 patients. Br J Anaesth 2000; 84: 6–10. Klafta JM, Roizen M. Current understanding of patients’ attitudes toward and preparation for anaesthesia: a review. Anaesth Analg 1996; 83: 1314–21. Brice DD, Hetherington RR, Utting JE. A simple study of awareness and dreaming during general anaesthesia. Br J Anaesth 1970; 42: 535–42. Nordström O, Sandin R. Recall during intermittent propofol anaesthesia. Br J Anaesth 1996; 76: 699–701. Quasha AL, Eger EI II, Tinker JH. Determination and applications of MAC. Anaesthesiology 1980; 53: 315–34. Pomfrett CJD. Heart rate variability, BIS and “depth of anaesthesia”. Br J Anaesth 1999; 82: 659–62.
711
Awareness during anaesthesia: a prospective case study
Rolf H Sandin, Gunnar Enlund, Peter Samuelsson, Claes Lennmarken
Summary Background Patients who are given general anaesthesia are not guaranteed to remain unconscious during surgery. Knowledge about the effectiveness of current protective measures is scarce, as is our understanding of patients’ responses to this complication. We did a prospective case study to assess conscious awareness during anaesthesia. Methods 11 785 patients who had undergone general anaesthesia were interviewed for awareness on three occasions: before they left the post-anaesthesia care unit, and 1–3 days and 7–14 days after the operation. Findings We identified 18 cases of awareness and one case of inadvertent muscle blockade that had occurred before unconsciousness. Incidence of awareness was 0·18% in cases in which neuromuscular blocking drugs were used, and 0·10% in the absence of such drugs. 17 cases of awareness were identified at the final interview, but no more than 11 would have been detected if an interview had been done only when the patients left the post-anaesthesia care unit. Four non-paralysed patients recalled intraoperative events, but none had anxiety during wakefulness or had delayed neurotic symptoms. This finding contrasts with anaesthesia with muscle relaxants, during which 11 of 14 patients had pain, anxiety, or delayed neurotic symptoms. After repeated discussion and information, the delayed neurotic symptoms resolved within 3 weeks in all patients. Analysis of individual cases suggests that a reduced incidence of recall of intraoperative events would not be achieved by monitoring of end-tidal anaesthetic gas concentration or by more frequent use of benzodiazepines. Interpretation The inability to prevent awareness by conventional measures may advocate monitoring of cerebral activity by neurophysiological techniques. However, the sensitivity of such techniques is not known, and in the light of our findings, at least 861 patients would need to be monitored to avoid one patient from suffering due to awareness during relaxant anaesthesia. Lancet 2000; 355: 707–11 See Commentary page xxx
Department of Anaesthesia and Intensive Care, Länssjukhuset, Kalmar, Sweden (R H Sandin MD, P Samuelsson MD); and Department of Anaesthesia and Intensive Care, Vrinnevisjukhuset, Norrköping, Sweden (G Enlund MD, C Lennmarken MD) Correspondence to: Dr Rolf Sandin, Department of Anaesthesia and Intensive Care, Länssjukhuset, S-39185 Kalmar, Sweden (e-mail: [email protected])
THE LANCET • Vol 355 • February 26, 2000
Introduction Recall of intraoperative events in patients under general anaesthesia is rare (0·1–0·7%), although few prospective studies have assessed the incidence of awareness during surgical procedures.1–4 Up to 54% of patients worry about the possibility of pain, paralysis, and mental distress during surgery.5 Previous studies have not recorded important background data, such as whether neuromuscular-blocking drugs were used or whether end-tidal anaesthetic gas concentration (ETAGC) was monitored. Also, memory for awareness may be delayed,1 and follow-up may have been too short to identify all cases. We did a prospective study to assess patients’ recall of awareness during surgery.
Patients and methods This study, which was approved by the Ethics Committee of Linköping University Hospital, was done in two Swedish hospitals. Clinical routine at the two hospitals includes an interview for awareness on discharge from the post-anaesthesia care unit in all patients older than 15 years who have had general anaesthesia. This interview, modified from Brice and colleagues6 and described elsewhere,1 is given by the nurses who attended the patients, and is recorded in the anaesthesia record. Dedicated nursing staff for the study recorded data from the routine interview in the post-anaesthesia care unit and repeated the interview 1–3 days and 7–14 days after anaesthesia. Inpatients were interviewed face-to-face, whereas patients who had left the hospital were contacted by telephone. Patients with possible memories of intraoperative events at any of the three interviews were given an additional interview to obtain data on awareness. All four investigators were involved in the interviewing of patients, and together they classified cases into one of three categories according to the likelihood of true recall: definite cases required that the recalled event was confirmed by attending personnel; for probable cases, the investigators were convinced that the memory was real, but no confirmation could be obtained; and possible cases, who were unable to recall any event definitely indicative of true awareness. A reported dream alone was not classified as awareness. Patients were enrolled between 1997 and 1998. Inclusion criteria were as follows: general anaesthesia with laryngeal mask airway, tracheal intubation, or jet ventilation; age of at least 16 years; ability to communicate in Swedish or English; no overt psychiatric disorder thought to interfere with the reliability of the interview; ability to attend at least one of the subsequent interviews; and agreement to participate. All patients who fulfilled the inclusion criteria were entered in the study after informed consent had been obtained. Planned and emergency surgery were included. Of 12 179 eligible patients, 394 were not included because they could not be contacted for either of the last two interviews, leaving 11 785 patients to be assessed. We assessed the cases of awareness or awake paralysis with reference to whether additional monitoring, education, or increased vigilance could have prevented awareness. Prevention was assessed only if an alternative method would have avoided wakefulness under the same circumstances. Additional neurophysiological monitoring was not allowed in any patient
707
ARTICLES
Total (n=11 785)
Patients with awareness (n=19)
Demography Age (years) Height (cm) Weight (kg) ASA physical status Male/female
48 (19) 170 (9) 74 (15) 1·4 (0·6) 4732/7053
45 (18) 170 (9) 78 (16) 1·4 (0·6) 7/12
Surgery Duration of anaesthesia (min) Elective/acute and emergency surgery*
99 (65) 9388/2397
95 (66) 14/5
ASA=American Society of Anesthesiologists. Data are means (SD) or numbers. The patient with awake paralysis is included among the awareness cases. *No cardiac surgery, transplantation, or neurosurgery was done at the hospitals during the study period. All trauma surgery is included, except for neurotrauma.
Table 1: Baseline data who was likely to fulfil the inclusion criteria, since our aim was to assess the present situation and the possible need for additional monitoring.
Results Baseline data of the 11 785 patients included in the study are shown in table 1. Anaesthesia and clinical procedures 11 440 (97·1%) patients had been given an opioid before anaesthesia and 2084 (17·7%) had received a benzodiazepine. No patient was given scopolamine. Anaesthesia was induced with propofol in 6473 (54·9%) patients, thiopentone in 5248 (44·5%), sevoflurane in 36 (0·31%), and ketamine in 28 (0·24%). Potent inhalation agents were used in 11 454 (97·2%) patients (sevoflurane in 7642 [66·7%], isoflurane in 3810 [33·3%], desflurane in two). Nitrous oxide was used in 11 034 (93·6%) patients given inhalation anaesthesia. Total intravenous anaesthesia with propofol was used Patient Demography
Surgery
Sex Age Weight Height ASA (years) (kg) (cm) grade
in 325 (2·8%) patients, 288 (89·2%) without nitrous oxide. Total intravenous anaesthesia with thiopentone was used in two patients. Intraoperative epidural analgesia with local anaesthetics was used in 346 (4·4%) of 7796 intubated patients. ETAGC was monitored in 9528 (83·2%) patients with inhalational anaesthesia. The airway was maintained by endotracheal intubation in 7796 (66·2%) patients (by neuromuscular block in 7737) and by laryngeal mask airway in 3977 (33·8%) patients (by neuromuscular block in four). Jet ventilation through a bronchoscope was used in 12 patients, all of whom received neuromuscular block. Among 7753 patients given neuromuscular block, 672 (8·7%) received only a single dose of suxamethonium for intubation. Recall of awareness 18 patients (11 [0·16%] women and seven [0·15%] men) reported awareness (six definite, eight probable, and four possible). One further woman (patient 4) had awake paralysis. Demographic data and details of surgery and anaesthesia for the 19 patients with awareness or awake paralysis are shown in table 2. Descriptions of patients’ experiences are listed in table 3. 14 patients with awareness had had previous surgery, of whom one had previously had awareness during surgery (patient 19). Neuromuscular block was used in 14 of 18 patients who reported awareness. However, none of the patients anaesthetised by total intravenous anaesthesia reported awareness during surgery. The incidence of awareness was similar for men and women. Overall incidence was 0·18% when neuromuscular block was used, and 0·10% in the absence of relaxant drugs.
Anaesthesia Co-induction
Induction
Maintenance N2O
ETAGC NMB monitoring
Benzodiazepine
Opioid
Propofol
Isoflurane
Yes
No
Rocuronium ETT
130 80
None None
Opioid Opioid
Propofol Sevoflurane
Isoflurane Sevoflurane
No Yes
No Yes
Vecuronium ETT None LMA
115
Opioid phentiazine
Opioid
Thiopentone
Sevoflurane
Yes
Yes
Rocuronium ETT
60
Opioid phentiazine None None
Opioid
Thiopentone
Sevoflurane
Yes
Yes
Rocuronium ETT
None Opioid
Thiopentone Thiopentone
Sevoflurane Sevoflurane
Yes Yes
Yes Yes
Opioid phentiazine None None
Opioid
Propofol
Sevoflurane
Yes
Yes
Suxamethon ETT Suxamethon, ETT rocuronium None LMA
Opioid Opioid
Thiopentone Thiopentone
Sevoflurane Sevoflurane
Yes Yes Yes Yes
Benzodiazepine Opioid phentiazine Opioid
Opioid Opioid
Thiopentone Thiopentone
Isoflurane Sevoflurane
Yes Yes
Yes Yes
Suxamethon Suxamethon, rocuronium Rocuronium Rocuronium
Opioid
Thiopentone
Sevoflurane
Yes
Yes
Rocuronium ETT
160
Opioid
Opioid
Propofol
Sevoflurane
Yes
Yes
None
59 45
Opioid None
Opioid None
Propofol Ketamine
Sevoflurane Isoflurane
Yes Yes
Yes Yes
None LMA Suxamethon ETT
Benzodiazepine
Opioid
Thiopentone
Isoflurane
Yes
Yes
Suxamethon ETT
None Opioid
Opioid Opioid
Propofol Thiopentone
Isoflurane Sevoflurane
Yes Yes
No Yes
Vecuronium ETT Rocuronium ETT
Duration of Premedication anaesthesia (min)
1
F
39
62
158
1
2 3
M F
60 34
84 60
173 162
1 1
4
F
51
85
174
1
5
F
21
56
161
1
Laparoscopic sterilisation Eye surgery Reconstructive breast surgery Laparotomy, resection of renal cyst Tonsillectomy
22
6 7
F M
30 23
73 87
169 185
1 1
Caesarean section Appendectomy
45 60
8
M
30
90
180
1
Hand surgery
55
9 10
F M
67 84
80 76
166 180
1 3
Hand surgery Hemicolectomy
60 140
11 12
M F
73 56
88 104
170 168
2 1
Colostomy Parathyroidectomy
246 248
13
F
40
65
162
1
55
14
M
31
54
172
2
15 16
M F
47 28
106 ?
187 ?
2 2
17
F
62
98
163
2
18 19
F F
31 45
64 70
162 163
1 1
Laparoscopic sterilisation Osteosynthesis, arm fracture Hand surgery Retrieval of retained placenta Laparoscopic cholecystectomy Tonsillectomy Laparoscopy
130 40 55
Airway
ETT ETT ETT ETT
LMA
ASA=American Society of Anesthesiologists; ETT=endotracheal intubation; LMA=laryngeal mask airway; NMB=neuromuscular block; suxamethon-=suxamethonium.
Table 2: Details of demographic data and surgery from patients who experienced awareness during surgery
708
THE LANCET • Vol 355 • February 26, 2000
ARTICLES Patient
Experience
1
Suddenly noticed endotracheal intubation, but felt no discomfort. Was scared. Tried to move but found it impossible. Heard somebody say “she is not asleep”.
2
Suddenly aware that he was awake, but felt no discomfort. Tried to move his feet. Heard somebody say “he is awake”.
3
Recognised the surgeon’s voice, heard noise from instruments, understood the situation, found herself unable to move. She thought “it will soon be over”, and soon became unconscious again.
4
Paralysed before she became unconscious. Realised something was wrong, although did not understand what. Tried to move but was unable. Repeated explanations were required. Anxiety gradually disappeared over 2 weeks.
5
Soon after anaesthetist had said “it is going to taste like garlic”, she heard “now it is time to put the tube down”, and also remembered another discussion. Soon fell unconscious again.
6
Probably only temporarily sedated by induction dose. Was suddenly widely awake and felt a cramping pain in the neck (suxamethonium).
7
Regained consciousness and found himself in some kind of turbulence. Felt pain in throat and felt like he was suffocating, but still did not panic. Understood what was happening.
8
Regained consciousness. Heard surgeon ask “can I cut here?” Felt someone manipulate wounded arm. No pain and no anxiety. No further memories.
9
Felt incision, realised what was happening and soon after lost consciousness.
10
Saw people dressed in green (operating room staff), felt severe abdominal pain, tried to talk and move, but was unable. Very agitated immediately after anaesthesia, said he had had a bad dream.
11
Heard voices, but could not recall what was said.
12
Heard someone say “is very narrow”. The patient did have a small mouth. After surgery, a tooth was found to have been damaged, possibly during intubation. However, a definite confirmation could not be obtained.
13
Realised she was awake when she should not be. She tried to get attention, tried to talk and move but was unable. Heard voices and noise from surgical instruments.
14
Suddenly saw a big, round lamp. Tried to move his arms and scream, but was unable. Had impression that surgeon was about to finish. Soon was unconscious again.
15
Suddenly felt pain in operated hand, understood situation, but did not worry since he realised he was “fading away” again.
16
Saw people dressed in green, saw tombstones, thought she was attending her own funeral. Heard voices speak slowly. No impression to have been unconscious at all. Anxiety gradually disappeared over 3 weeks.
17
Sharp pain and a sense of “darkness”. Postoperative anxiety disappeared over following 3 weeks.
18
Heard someone say “not quite finished yet . . . give more” and heard noise from surgical instruments. Felt mask pressed onto her face (on awakening?).
19
Heard a voice call for “more anaesthetics”. Saw a lamp.
Table 3: Patients’ experiences of awareness during surgery
Patient
1 2 3 4* 5 6* 7 8 9 10 11 12 13 14 15 16 17 18 19
Perception
Motor function
Six patients recalled awareness at the first interview, before they left the post-anaesthesia care unit. In another patient, later found to have had awareness, this interview was taken to be inconclusive (patient 17). In four patients who later reported memories, no result from the first interview was recorded. 12 patients had recall at the second interview, including five of the six patients who had memories at the first interview. One patient had forgotten about the intraoperative events that she had reported at the first interview (patient 12). Awareness was confirmed in three of four patients in whom no result from the first interview had been recorded. One of these patients firmly denied having had any memories until just before the second interview (patient 19). In addition, three patients who denied recall of awareness at the first interview, reported at the second interview that they had had awareness during surgery. The patient who gave an inconclusive answer at the first interview also reported recall. 17 of 18 patients with awareness had recall of events at the third interview. Five patients had not previously been identified as having been aware. In one of these patients, no result from the first interview was recorded, and no recall was evident at the second interview (patient 5). The patient who had forgotten the memories that she disclosed at the first interview could remember them again (patient 12). One patient who had no recall at the first interview (patient 13), but remembered intraoperative events at the second, had once again forgotten (12 days after surgery), although 24 days after surgery she had detailed memories. One patient thought he would not have remembered at all had he not been questioned (patient 15). The patient who had awake paralysis remembered at all three interviews about being paralysed before anaesthesia. Auditory and visual perceptions were present in 12 and five patients with awareness, respectively (table 4). Pain was reported by seven patients, two of whom reported severe pain (patients 10 and 17). 11 patients with awareness were convinced that they had tried to move, but only one realised he was able to do so (patient 2). Two patients who were unable to move had received no neuromuscular block (patients 3 and 14), although one of these patients had restrained arms (patient 14). Three patients denied making any attempt to move despite their understanding of the situation.
Mental reaction
Assessment
Auditory
Visual
Pain
Tried to move
Able to move
Immediate understanding
Immediate anxiety
Delayed symptoms
Confirmed event
Probability
Detection at interview
Yes Yes Yes Yes Yes ? Yes Yes No ? Yes Yes Yes No No Yes No Yes Yes
No ? Yes Yes No No No No No Yes ? No No Yes No Yes No No Yes
No No No No No Yes Yes No Yes Yes Yes No No No Yes No Yes No No
Yes Yes Yes Yes Yes Yes Yes No ? Yes Yes No Yes Yes No Yes No No No
No Yes No No No No No .. ? No No .. No No .. No .. .. ..
Yes Yes Yes No Yes Yes Yes Yes Yes No Yes No Yes No Yes No No No Yes
Yes No No Yes Yes Yes No No No No Yes Yes Yes No No Yes Yes No No
No No No Yes Yes No No No No ? No No No No No Yes Yes No No
Yes Yes Yes Yes Yes Yes Yes No No No No No No No No No No No No
Definite Definite Definite Definite Definite Definite Definite Probable Probable Probable Probable Probable Probable Probable Probable Possible Possible Possible Possible
1, 2, 3 1, 2, 3 2, 3 1, 2, 3 3 1, 2, 3 3 2, 3 3 1, 2, 3 2, 3 1, 3 2 3 3 1, 2, 3 2, 3 2, 3 2, 3
*Probably never lost consciousness before being paralysed (definitely in the case of patient 4).
Table 4: Details of awareness obtained by interview
THE LANCET • Vol 355 • February 26, 2000
709
ARTICLES
Of 14 patients with recall and who had received neuromuscular block, nine immediately realised their situation, and six were anxious. Only one of these six patients (patient 5) had delayed symptoms (nightmares and anxiety). Among the five paralysed patients who did not understand their situation during wakefulness, three felt anxiety, two of whom also experienced delayed symptoms. The patient who had awake paralysis did not realise the situation, felt anxiety, and experienced delayed symptoms. Among the four patients with delayed symptoms after awareness or awake paralysis, only one had pain. No patient who was aware during non-relaxant anaesthesia felt anxiety during wakefulness. Three of these patients understood their situation during wakefulness and all four were free from delayed symptoms. Of the 14 awareness cases who had received neuromuscular block, 11 had unpleasant effects from their experience for one or more reasons. If the patient who had awake paralysis is taken into account, ten patients experienced anxiety during wakefulness. Delayed symptoms were evident in three of the four patients who had not been able to understand the situation, but in only one of the six who understood what was happening. Among the six paralysed patients who reported pain, only one patient had delayed symptoms (patient 17), and this patient was also among those who felt anxiety and was unable to understand the situation. All patients who experienced awareness or awake paralysis were offered repeated discussion and explanation according to their own desire, and within 3 weeks all patients declared that they were satisfied with the explanations, that they did not need any further contact, and that all their delayed symptoms had disappeared. However, one patient thought she might become worried again if she needed surgery in the future. Probable reasons for awareness For patient 2, the vaporiser was turned off after a refill during anaesthesia. For patient 4, the anaesthetist administered rocuronium before thiopentone, and assumed a time of onset similar to pancuronium. Patient 5 probably received an insufficient induction dose of thiopentone or opioid, or both, to prevent arousal during laryngoscopy. For patient 6, the anaesthetist noticed too late that thiopentone had been partially injected backwards into the fluid line, and the patient was aroused by suxamethonium-induced fasciculations. Patient 7’s awareness was caused by a difficult intubation and this was probably the reason in patient 12 as well. Patient 16 was the only patient classified as possible who had recall at all three interviews. Her memories were probably related to dreams caused by ketamine, although influences from real events cannot be ruled out. Patient 18 was suspected to have begun awakening too early. Patient 19 described a memory very similar to when she suffered from awareness 3 years previously. However, she was convinced she had been aware again. In the remaining ten patients, no obvious reason for awareness could be identified. We assumed that a neurophysiological monitoring technique capable of discrimination between consciousness and unconsciousness could have been helpful to prevent, at most, 16 of the 18 cases of awareness (not patients 6 and 16). Monitoring of
710
ETAGC was used in 15 patients with recall, and did not help in these cases. We assumed that monitoring would have been helpful in only one of the three cases of awareness among the 1928 patients without ETAGC (patient 2). The anaesthetist’s insufficient knowledge about the time of onset for rocuronium led to the case of awake paralysis (patient 4), and lack of vigilance contributed in two cases of awareness (patients 2 and 6).
Discussion Our findings and our inability to find non-paralysed patients who had unpleasant effects during or after wakefulness suggest that awareness among patients without neuromuscular block is not a major concern, as long as conversation between operating-room staff remains respectful to the patient. Studies on awareness should therefore distinguish clearly between relaxant and non-relaxant anaesthesia. If the first interview had been done after the patients had left the post-anaesthesia care unit, no more than 11 and probably only nine of the 18 awareness cases would have been identified. There is only one previous study in which an interview was undertaken more than 72 h after anaesthesia. In that study, one patient did not recall dreaming or intraoperative events soon after anaesthesia, but remembered an unpleasant dream the next day, and had recall of operative events after 8 days.1 However, in our study, patients’ memories did not always remain constant or improve with time. In two patients, a reported memory of intraoperative events was temporarily forgotten at the next interview. A similar case was reported in a study in which one patient had recall soon after anaesthesia, but later forgot about his experiences.2 Obviously, there is no time when all cases of awareness will be detected reliably, but most previous studies may have underestimated the true number of patients with recall.1–4 This issue is not only academic, since among the three awareness cases with delayed neurotic symptoms in our study, one was not detected until day 10, and another patient may have been overlooked because he gave a vague and inconclusive answer at the first interview (patient 17). Improved knowledge and vigilance by the anaesthetist may have prevented two of the cases of awareness and also the case of awake paralysis. However, in this type of study, it is not possible to detect less obvious cases of poor anaesthetic performance, and improved education may be of greater importance to prevent awareness than indicated by the three situations we identified. There was no difference in the incidence of awareness among patients who had been given a benzodiazepine before anaesthesia and those who had not. Benzodiazepines are, in many cases, valuable for relieving anxiety before surgery, but they are also associated with a variable degree of anterograde amnesia, and their widespread use as premedication may be because of this reputation, which is shared with scopolamine. However, there is no evidence that effects of benzodiazepines or scopolamine are better than the amnesic effects of ordinary anaesthetics.7 More importantly, we believe that to carry out anaesthesia in a way that would require deliberate pre-emptive amnesia for intraoperative experiences is ethically unsound. The incidence of awareness among patients with neuromuscular block was similar irrespective of whether ETAGC was measured or not. This result is weakened
THE LANCET • Vol 355 • February 26, 2000
ARTICLES
by the fact that the patients were not randomised by whether ETAGC was used. However, it is striking that ETAGC was monitored in 15 of the 18 cases of awareness, and it is by no means clear how many, if any, of the remaining three cases would have been avoided if this technique had been used. Clearly, if the accepted ETAGC is too low, this technique will be of no value in the prevention of awareness. A safe concentration of anaesthetic gas has never been established in a sufficiently large study. Institutional guidelines at the two departments involved in our study recommend at least 1·0 MACskin incision (a measure of anaesthetic potency8) and preferably 1·3 MACskin incision of halogenated volatile anaesthetic for anaesthesia supplemented with opioids and nitrous oxide. Even if this recommendation is not sufficient to prevent awareness in some patients, a more vigorous administration of anaesthetic gas may not be possible in all cases due to haemodynamic side-effects. However, we do not believe that a general use of low concentrations of anaesthetic gas at the two departments in our study is the reason for the inability of ETAGC monitoring to prevent awareness, since the overall incidence of recall is one of the lowest reported. Our assumption that factors other than an inability to measure ETAGC are responsible for cases of awareness in contemporary anaesthetic practice is strengthened by the fact that a similar incidence of 0·2% was found with total intravenous anaesthesia during which monitoring of ETAGC is of no value.1 Thus, the extent to which ETAGC monitoring reduces the incidence of awareness is seriously questionable, and from our findings this monitoring may not help to reduce this incidence further. Monitoring techniques with alleged capability to identify intraoperative wakefulness have been proposed.9 However, no such monitoring was allowed in our study. Whether techniques such as bispectral index and midlatency auditory evoked potentials are able to reduce the incidence of awareness will probably never be proven in a randomised controlled study since about 50 000 patients would need to be included to show a significant reduction in incidence from 0·2% to 0·1%. To decide whether substantial economical investment in technical devices to reduce the incidence of awareness is justified,
THE LANCET • Vol 355 • February 26, 2000
it is essential to know not only the true current incidence of awareness, but also the proportion of patients who suffer during and after wakefulness. When the cases of awareness found in our study were scrutinised, we assumed that a neurophysiological monitoring technique with 100% sensitivity could have been helpful to avoid at most 12 of the 14 cases that occurred during relaxant anaesthesia. Nine of those 12 patients suffered in one or more ways from their experience. This means that at least 861 patients must be monitored to avoid one case of suffering due to awareness in relaxant anaesthesia, provided that the sensitivity of that particular technique is 100%, including human performance. Contributors Rolf Sandin designed the study and analysed the data. Gunnar Enlund and Peter Samuelsson were responsible for quality control, and interviewed the study participants. Claes Lennmarken and Rolf Sandin were responsible for fund raising. All investigators wrote the paper.
Acknowledgments We thank Ulla Agne, Ann-Mari Engström, Wivi Forsberg, Mats Hultman, and Gunnar Svensson for collecting and recording the data. The study was supported by a grant from Forskningsrådet i Sydöstra Sverige.
References 1
2
3
4
5
6
7 8 9
Nordström O, Engström AM, Persson S, Sandin R. Incidence of awareness in total i.v. anaesthesia based on propofol, alfentanil and neuromuscular blockade. Acta Anaesthesiol Scand 1997; 41: 978–84. Ranta SOV, Laurila R, Saario J, Ali-Melkkilä T, Hynynen M. Awareness with recall during general anesthesia: incidence and risk factors. Anesth Analg 1998; 86: 1084–89. Liu WH, Thorp TA, Graham SG, Aitkenhead AR. Perception and memory during general anaesthesia. Anaesthesia 1991; 46: 435–37. Myles PS, Williams DL, Hendrata M, Anderson H, Weeks AM. Patient satisfaction after anaesthesia and surgery: results of a prospective survey of 10 811 patients. Br J Anaesth 2000; 84: 6–10. Klafta JM, Roizen M. Current understanding of patients’ attitudes toward and preparation for anaesthesia: a review. Anaesth Analg 1996; 83: 1314–21. Brice DD, Hetherington RR, Utting JE. A simple study of awareness and dreaming during general anaesthesia. Br J Anaesth 1970; 42: 535–42. Nordström O, Sandin R. Recall during intermittent propofol anaesthesia. Br J Anaesth 1996; 76: 699–701. Quasha AL, Eger EI II, Tinker JH. Determination and applications of MAC. Anaesthesiology 1980; 53: 315–34. Pomfrett CJD. Heart rate variability, BIS and “depth of anaesthesia”. Br J Anaesth 1999; 82: 659–62.
711